NEW INNOVATIVE PESTICIDES:
AN EVALUATION OF INCENTIVES
AND DISINCENTIVES
FOR COMMERCIAL DEVELOPMENT
BY INDUSTRY
SEPTEMBER 1977
FINAL REPORT
CRITERIA AND EVALUATION DIVISION
OFFICE OF PESTICIDE PROGRAMS
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D C. 20460
-------
Document is available to the public through the National Technical
Information Service, Springfield, Virginia 22151
-------
NEW, INNOVATIVE PESTICIDES: AN EVALUATION OF INCENTIVES AND
DISINCENTIVES FOR COMMERCIAL DEVELOPMENT BY INDUSTRY
By
Stanford Research Institute
Jeanie H. Ayers
Thomas A. Blue
Ray R. Bramhall
Terry J. Braunstein
Edward E. Davis
Joseph I. DeGraw
Thomas E. Elward
Robert E. Inman
Oscar H. Johnson
Ellen B. Leaf
Fred L. Offensend
Peter D. Stent
Final Report
September 1977
Contract No. 68-01-2487
Prepared for
Economic Analysis Branch
Criteria and Evaluation Division
Office of Pesticide Programs
U.S. Environmental Protection Agency
Washington, D.C. 20460
Dr. Arnold L. Aspelin, Project Officer
EPA-540/9-77-020
-------
This report has been reviewed by the Office of Pesticide Programs of
the U.S. Environmental Protection Agency and approved for publication.
Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
-------
CONTENTS
LIST OF ILLUSTRATIONS vii
LIST OF TABLES ix
INTRODUCTION xiii
I SUMMARY AND CONCLUSIONS 1
A. Chemical Pesticides 1
B. Innovative Pesticides 2
C. Profit as an Incentive 4
D. Risk as a Disincentive 5
E. User-Assessed Innovative Pesticide Trends 6
II CRITICAL FACTORS IN PESTICIDE PRODUCT DEVELOPMENT .... 9
A. Critical Factor Identification 9
B. Selected Critical Factor Sensitivity Analyses .... 11
C. Discussion ..... 11
1. Market Size (Sales Volume) 11
2. Selling Price and Product Margin 18
3. Research and Development Costs 19
4. Receipt of Registration 20
5. Capital Costs 20
III RECOMMENDATIONS 23
IV METHOD OF APPROACH . 27
A. Preliminary Investigations 27
B. Decision Analysis 28
1. Problem Structuring 28
2. Base Case Analysis 36
3. Sensitivity Analysis 36
4. Probabilistic Analysis 39
C. Critical Factor Analysis 42
D. Information Quality and Sources 43
E. Selected References 44
V LIMITS OF INTERPRETATION 45
iii
-------
VI BACTERIA 47
A. Overview 47
B. Background 48
C. Panel Survey Analysis 50
1. Sample Copy 50
2. Synopsis of Results 50
3. Advantages and Disadvantages of Bacterial
Pesticides 51
4. Relative Costs for Research and Development:
Bacterial Versus Chemical Pesticides 54
5. Origin of Research Funds 55
6. Bacillus Thuringiensis—Future Production Levels . 55
7. Bacillus Thuringiensis—Crops Representing
Potential Expanding Markets 55
8. Bacillus Thuringiensis—Major Obstacles to
Greater Use . . . 56
9. Bacillus Thurlnglenais—Host Resistance. 56
10. Discussion 58
D. Decision Analysis of a Possible Bacterial Pesticide
Venture 60
1. Introduction 60
2. Background on the Product 61
3. Base Case Analysis 81
4. Sensitivity Analysis 87
5. Probabilistic Analysis . „ 89
E. Survey Questionnaire . . 101
F. Selected References 112
VII VIRUSES
A. Overview 115
1. Limiting Factors 115
2. Current Development Potential . 116
B. Background 118
C. Panel Survey Analysis . 121
1. Sample Copy 121
2. Prospective Viral Agents ...... 121
3. General Problems for Viral Pesticides 127
4. Role of Private Industry by Type of Company . . . 128
5. Commercial Advantages of Selected Viruses .... 133
D. Decision Analysis of a Possible Viral Pesticide
Venture 135
1. Introduction 135
2. Background on the Product 135
3. Base Case Analysis 140
iv
-------
VII VIRUSES (cont.)
4. Sensitivity Studies 145
5. Probabilistic Analysis \. 148
6. Problems to be Overcome for Successful
Commercialization 157
E. Survey Questionnaire 161
F. Selected References 172
VIII PHEROMONES 175
A. Overview 175
B. Background • 177
C. Individual Product Reviews 178
1. Forest Pests 178
2. Agricultural Pests 188
3. Human and Veterinary Pests 196
D. Decision Analysis of a Possible Pheromone Pesticide
Venture 1 197
1. Introduction 197
2. Background on the Product 197
3. Base Case 197
4. Sensitivity Studies 202
5. Probabilistic Analysis 203
E. Selected References 210
IX "CONVENTIONAL" CHEMICAL PESTICIDES 215
A. Overview 215
B. Industry Survey 216
1. General Characteristics of Successful
Pesticides 216
2. Industry Participation in Innovative Pesticides . 217
3. Specificity of Pesticide Action 217
4. Pest Resistance 218
5. Integrated Pest Management (IPM) 218
6. Research and Registration 218
7. Government Subsidies 220
8. Patentability 221
9. Nomenclature 221
C. Decision Analysis of a Possible Innovative Chemical
Pesticide Venture .... 224
1. Introduction 224
2. Background on the Product 225
3. Base Case . 230
4. Sensitivity Analysis 237
5. Probabilistic Analysis 238
D. Selected References > 245
v
-------
X OTHER AGENTS 247
A. Overview 247
B. Colletotrichum Gloeosporioides, A Fungal Pathogen
of Northern Jointvetch 247
1. Introduction 247
2. Summary 248
3. Description and Properties 249
4. Production 259
5. Market and Use Parameters 260
C. Selected References 264
APPENDICES
A. PROJECT CONTACTS 267
B. CROP-PEST SURVEY 287
vi
-------
ILLUSTRATIONS
1 Method of Approach 29
2 Assessment of Commercial Feasibility of New Pesticide
Products 31
3 Market Model Considerations 33
4 Research and Development Model 37
5 Simplified Decision Tree: New Pesticide Commercialization
Decision 40
6 Sample Probability Distribution on Profitability of New
Pesticide Product Venture 41
7 Cumulative Probability Distribution on Total Annual
Bacillus Thuringiensis Market Size at Peak 91
8 Cumulative Probability Distribution on Long Term Percent
Market Share of First Company to Provide Bacillus
Thuringiensis ^ 92
9 Cumulative Probability Distribution on Retail Price of
Bacillus Thuringiensis ... 93
10 Cumulative Probability Distribution on Production Cost
of Bacillus Thuringiensis . 94
11 Cumulative Probability Distribution on Cost of
Bacillus Thuringiensis Plant and Equipment 95
12 Decision Tree for Bacillus Thuringiensis Venture 96
13 Cumulative Probability Distribution on Net Present Value,
at Eight Percent, of Bacillus Thuringiensis Venture .... 98
14 Cumulative Probability Distribution on Retail Price of
Bacillus Thuringiensis in 1985 for a Patentable Case .... 99
15 Cumulative Probability Distribution on Net Present Value,
at Eight Percent, for the Patentable Bacillus Thuringiensis
Case 100
16 Sample Questionnaire, Bacterial Pesticides 102
17 Schematic of Decision to Commercialize Heliothis Spp. NPV . 149
18 Probability and Event Assignments on Research and Develop-
ment Outcomes and Attainment of Registration for Heliothis
Spp. NPV 150
19 Probability and Event Assignments for Market Outcomes for
Heliothis Spp. NPV 151
vii
-------
20 Decision Tree for Possible Heliothis Spp. NPV Venture . . . 153
21 Probability Distribution on Net Present Value, at Eight
Percent, of Heliothis Spp. NPV Venture 158
22 Cumulative Probability Distribution on Net Present Value,
at Eight Percent, of Heliothis Spp. NPV Venture 160
23 Sample Questionnaire, Viral Pesticides 162
24 Schematic of Gossyplure Commercialization Decision .... 205
25 Probability Assignments on Research and Development Costs
and Attainment of Registration for Gossyplure 206
26 Probability Assignments on Market Size and Selling Price
for Gossyplure 207
27 Probability Assignments on Percent of Gossyplure Market
Captured 208
28 Probability Distribution on Net Present Value, at Eight
Percent, of Gossyplure Venture 209
29 Schematic of TH 6040 Commercialization Decision 239
30 Probability Assignments on Research and Development Costs,
Building Capital, and Production Costs for TH 6040 .... 240
31 Probability Assignments on Registration, Selling Price, and
Market Size for TH 6040 241
32 Probability Distribution of Net Present Value, at Eight
Percent, of TH 6040 Venture . . . 243
B-l Sample Questionnaire, Pesticide Use Parameters ; 312
viii
-------
TABLES
1 Major Factors Affecting the Commercial Feasibility of
New Pesticide Products 10
2 Relative Importance of Factors Affecting the Commercial
Feasibility of Bacillus Thuringiensis 12
3 Relative Importance of Factors Affecting the Commercial
Feasibility of Heliothis Spp. NPV 13
4 Relative Importance of Factors Affecting the Commercial
Feasibility of Gossyplure 14
5 Relative Importance of Factors Affecting the Commercial
Feasibility of TH 6040 15
6 Ranking of Advantages and Disadvantages of Bacterial
Compared to Chemical Pesticides, Cited by Questionnaire
Respondents 52
7 Prices Paid by Users for Selected Insecticides Used on
Cole Crops 53
8 Costs of Bacterial and Chemical Pesticide Development,
Estimated by Questionnaire Respondents 54
9 Percentage of Total Insecticide-Treated Acreage Treated
with Bacillus Thuringiensis, for Selected Crops Estimated
by Questionnaire Respondents 57
10 Major Obstacles to Increased Bacillus Thuringiensis Use . . 57
11 Serological Types and Biochemical Varieties Currently
Distinguished Within Bacillus Thuringiensis (Berliner) ... 65
12 Crops Registered for B.t. Use, Pests Controlled, Recom-
mended Dosages 68
13 Base Case Maximum Potential Market Size for Bacillus
Thuringiensis 82
14 Base Case Bacillus Thuringiensis Production Costs 84
15 Base Case Research and Development Costs for Bacillus
Thuringiensis 85
16 Base Case Pre-Tax Cash Flow for Bacillus Thuringiensis
Venture ' 86
15 Important Insect Viruses with Potential as Commercial
Pesticides 120
16 Potential Viral Pesticides, Target Pests, and Principal
Crops Affected 122
ix
-------
17 Viruses with Greatest Commercial Potential Identified
by Questionnaire Respondents 124
18 Percentage of Respondents Citing Selected Viruses as
Having the Best Potential for Commercialization 125
19 Viral Pesticide Sales Estimated by Survey Respondents
(1980, 1985) 127
20 Viruses with Best Potential for Commercialization and
Their Market Share for Pest Control on Affected Crops,
as Identified by Survey Respondents 129
21 Research and Development Costs for Viral Pesticides as
Estimated by Survey Respondents 131
22 Base Case Maximum Potential Market Size for Heliothis
Spp. NPV 141
23 Base Case Research and Development Costs for Heliothis
Spp. NPV (Thousands of Dollars) 144
24 Base Case Pre-Tax Cash Flow for Heliothis Spp. NPV Venture . 146
25 Selected Single-Variable Sensitivity Studies for Heliothis
Spp. NPV 147
26 Critical Attributes of Profitable Heliothis Spp. NPV
Scenarios . 159
27 Relative Probability of Pest Suppression Programs Using
Pheromones (1980, 1985) 176
28 Use of Pheromones in Insect Control 179
29 Base Case Research and Development Costs for Gossyplure . . 200
30 Base Case Pre-Tax Cash Flow for Gossyplure Venture 201
31 Selected Single-Variable Sensitivity Studies for
Gossyplure 202
32 Base Case Maximum Potential Market Size for TH 6040 .... 231
33 Advantages and Disadvantages of TH 6040 Compared With
Other Principal Pesticides Used for Soybean Looper and
Velvetbean Caterpillar , . 232
34 Base Case Production Cost Estimate for TH 6040 233
35 Base Case Research and Development Costs for TH 6040 .... 235
36 Base Case Pre-Tajf Cash Flow for TH 6040 Venture 236
37 Selected Single-Variable Sensitivity Studies for TH 6040 . . 238
38 Certain Equivalent of TH 6040 Venture as a Function of
Risk Coefficient, a 244
x
-------
39 Crop, Ornamental, and Weed Plants That Have Been
Observed to be Immune to Infection with Colletotrichum
Gloeosporioides Under Field Conditions 252
40 Control of Northern Jointvetch with Colletotrichum
Gloeosporioides in Small Field Plots, Stuttgart,
Arkansas, 1970 257
41 Effect of Inoculation of Northern Jointvetch in Small
Field Plots with Spores of Colletotrichum Gloeosporiodes,
Stuttgart, Arkansas, 1971 258
42 Aerial Application of Colletotrichum Gloeosporiodes
Spores in Large Fields, Stuttgart, Arkansas, 1971 .... 258
B-l Respondents Expectations of 1985 Levels of Pest Control
by Pesticide Class for Selected Crops and Pests 289
B-2 Corn Diseases Identified by Survey Respondents 291
B-3 Corn Insect Pests Identified by Survey Respondents .... 292
B-4 Wheat Diseases Identified by Survey Respondents 295
B-5 Wheat Insect Pests Identified by Survey Respondents. . . . 296
B-6 Tobacco Diseases Identified by Survey Respondents 298
B-7 Tobacco Insect Pests Identified by Survey Respondents. . . 299
B-8 Peanut Diseases Identified by Survey Respondents 301
B-9 Peanut' Insect Pests Identified by Survey Respondents . . . 302
B-10 Potato Diseases Identified by Survey Respondents 305
B-ll Potato Insect Pests Identified by Survey Respondents . . . 306
B-12 Apple Diseases Identified by Survey Respondents 308
B-13 Apple Insect Pests Identified by Survey Respondents . . . 309
B-14 Citrus Diseases Identified by Survey Respondents 310
B-15 Citrus Insect Pests Identified by Survey Respondents . . . 311
xi
-------
INTRODUCTION
The Office of Pesticide Programs (OPP) within the U.S. Environmental
Protection Agency (EPA) is responsible for regulating the use of pesti-
cides in order to protect man and the environment under the Federal In-
secticide, Fungicide, and Rodenticide Act (FIFRA) as amended by the
Federal Environmental Pesticide Control Act (FEPCA). In addition, EPA
assesses the availability of potentially less hazardous chemicals or
methods of pest control which could substitute for currently registered
pesticides.
Industry, academia, and pesticide regulatory authorities have demon-
strated considerable interest in such new innovative pest control mater-
ials. However, although some biologically derived chemical controls
(including mimics) and living organisms have-been used successfully,
they currently play a small part in the pesticide market as a whole.
Some of the new agents in these and other categories appear to be poten-
tial substitutes or complements for currently used pesticides, but the
actual economic feasibility of groups or classes of these products has
not been investigated. Although several companies are conducting some
research and development (R&D) activities on such materials, there are
very few of these innovative pesticides on the commercial market.
Accordingly, EPA contracted with SRI for this study of the commer-
cial feasibility of such new innovative pesticides to determine:
• The rate at which such products might be introduced to the
commercial marketplace between 1975 and 1985.
• The critical factors that influence the incentive for in-
dustry to develop these and other pest control agents.
The work was completed and accepted by EPA in late 1975 and this
volume comprises the publicly-available results of the project. The
basic research was conducted in 1974 and 1975, and while some of the
details presented herein would be altered or modified given data since
available, SRI feels that the overall results remain valid.
The study was conducted as a multidisciplinary project employing the
following personnel from several SRI departments and divisions:*
Jeanie H. Ayers, Industrial Economist, Chemical Information
Services, Chemical Industries Center
~Personnel and positions are listed as of the date of the research;
several changes in occupation have occured since then.
xiii
-------
Ray R. Bramhall, Organic Chemist, Pharmaceutical Chemistry
Department, Life Sciences Division,
Terry J. Braunstein, Decision Analyst, Decision Analysis
Department, Economics and Management Systems Division
Edward E. Davis, Biologist, Psychobiology & Physiology
Department, Life Sciences Division
Joseph I. DeGraw, Program Manager, Pharmaceutical Chemistry
Department, Life Sciences Division
Thomas E. Elward, Manager, Animal Health Program, Life
Sciences Division
Robert E. Inman, Manager, Plant Biology Program, Life
Sciences Division
Oscar H. Johnson, Senior Industrial Economist, Chemical-
Environmental Program, Chemical Industries Center
Ellen B. Leaf, Programmer, Decision Analysis Department,
Economics and Management Systems Division
Fred L. Offensend, Decision Analyst, Decision Analysis
Department, Economics and Management Systems Division
Peter D. Stent, Agricultural Economist, Food and Agricultural
Industries Department, Economics and Management Systems
Division
The project leader was Thomas A. Blue, Manager, Agricultural Chemicals,
Chemical Industries Center.
xiv
-------
I SUMMARY AND CONCLUSIONS
A tremendous quantity of public and private research dollars has
been spent during the past several years on innovative materials and
methods for pest control. Prominent innovative techniques have included
the use of bacteria, viruses, pheromones, juvenile hormones, chitin in-
hibitors, and natural parasites and predators. EPA, considering the
large amounts of money and the publicity accorded some of these materials
and procedures (and, as indicated earlier, in conjunction with its
responsibilities under FIFRA—as amended), contracted SRI to investigate
the rate at which innovative products from these research efforts would
probably be introduced into the commercial marketplace as effective and
environmentally acceptable alternatives or complements to present pest
control tools. SRI's general conclusions follow:
A. Chemical Pesticides
In terms of materials used in pest control (as opposed to cultural
and other management techniques), synthetic chemicals have been heavily
predominant over the past 50 years in the control of insects, diseases,
and weeds. SRI believes that over the next ten years the flow of prod-
ucts from the chemical industry for these uses will continue, although
there may be temporary shortages of some chemicals in certain functional
or regional markets due to problems such as periodic physical supply
constraints and increases in pest resistance. It is also possible that
a serious constriction in the supply of selected existing efficacious
products will result from EPA cancellations or restrictions. However,
the chemical pesticide industry has been healthy and profitable, and
many major companies in the United States and abroad are maintaining or
increasing their investments in R&D and new plants, and are expected to
continue their efforts to commercially develop new chemical pesticides
that are commensurate with stabilized environmental guidelines.
In this respect, a transition in research from random-screening
techniques to a more sophisticated approach incorporating targeted
searches for control products that act on specifically identified bio-
chemical or physiological processes within target organisms appears to
be under way within the chemical industry. This reflects acknowledge-
ment that random screening may be approaching a point of diminishing
returns, and that $ more sophisticated understanding of mode-of-action
characteristics will probably be required for future successes, particu-
larly in insecticide development.
Some of the more innovative chemical products in which mode-of-
action knowledge is playing an important role—not only in the design '
1
-------
and understanding of how a product works but in how and when it must be
used—include products with ovicidal activity; and insect growth inter-
ruptors such as juvenile hormones, their analogs and antagonists, and
chitin inhibitors.
B. Innovative Pesticides
SRI concludes that, in this type of environment, products such as
bacterial, pheromonal, or viral pesticides will not be registered and
available in sufficient quantities to dramatically substitute for con-
ventional chemical pesticides over the next ten years—although some are
expected to complement the use of such chemicals in selected markets.
Of all the specific innovative agents of this type reviewed during the
course of this study, the following currently appear to have the best
chance of commercial success between 1975 and 1985:
Bacillus thurlngiensis (B.t.)—This product is already commercially
available. Current U.S. sales are approximately 1.5 million pounds
per year. Sales might triple over the next decade with major potential
increases occurring in existing markets (largely vegetables), cotton,
and in the control of forest lepidoptera. However, such anticipated
increases assume favorable results of additional research on improved
strains and on other aspects of product efficacy (and, in the cotton
market, the availability of chlordimeform).
Beyond B.t. there appear to be no bacterial-based pesticide candi-
dates likely to receive short- to medium-term development for major
markets.
Heliothis spp. NPV—This viral agent has a large potential market;
it could be targeted to control worm complexes in cotton, tobacco, toma-
toes, and sweet corn. More work has been conducted on Heliothis spp.
NPV than on any other virus as a pesticide. One private company (Sandoz)
has recently received registration for its Elcar® product, and it may
achieve a position in at least the cotton marketplace by 1980. Other
viral agents with some degree of commercial potential by 1985 include
four for the control of nonagricultural tree pests (Porthetria dispar
NPV for gypsy moths, Neodipron complex NPV for pine sawflies,
Choristoneura fumiferana NPV for spruce budworm, and Orygia pseudotsugata
NPV for Douglas fir tussock moth), plus three others for the control of
pests on cole crops, alfalfa, soybeans, and cotton (Autographa califor-
nica NPV, Plusiinae viruses, and Spodoptera complex NPV). Given that
most or all of the non-agricultural tree pest viruses would probably be
sold to and used by public agencies such as the U.S. Forest Service,
some of these products may eventually be used in larger than experimental
2
-------
quantities.* However, general problems still associated with the pro-
duction, use and efficacy, and safety and registration of viruses as
pesticides leads to the conclusion that viewed by a private company
weighing alternative investment decisions, they have limited development
potential for the next several years.
Insect Pheromones—Several insect pheromones are being used in pest
management programs for scouting, monitoring, and survey purposes. These
uses will continue and probably expand, although the tonnage markets for
the active pheromone chemicals are extremely small. In these markets,
pheromones are not subject to the constraints of EPA registration require-
ments. For actual control—true suppression by either mass trapping or
mating disruption—pheromones appear to have the greatest potential for
commercial use in controlling the following pests:
Other Agents—Additional agents and techniques for pest control in-
clude manipulation and release of sterile insects, natural parasites,
predators, and pathogens (for example, Colletotrichum gloeospoioides, a
fungal pathogen, may be used to control northern jointvetch in rice);
sound and other energy waves; and pest-resistant crop strains. [An excel-
lent review of the development of some of these agents and techniques
appears in Hearings Before a Subcommittee of the Committee of Appropria-
tions, House of Representatives, Whitten, J., Chairman, "Agriculture-
Environmental and Consumer Protection Appropriations for 1973," pp.
590-1073 (1973)].
These techniques were reviewed during the initial stages of this
contract and SRI concluded that selected application would continue on
an evolutionary basis in agricultural, nonagricultural, and public health
programs. In many cases, however, the range of activities would be
almost totally controlled by public agencies, and a more detailed assess-
ment of critical factors associated with the development of these pro-
cedures could not be effectively evaluated within the commercial
*Directly prior to the publication of this report, the USDA obtained a
registration for the use of Orygia pseudotsugata NPV in the control of
Douglas fir tussock moth.
Pest
Potential for Commercial Development
By 1980 By 1985
Pink bollworm (California)
Stored product pests
Western pine beetle
Boll weevil
Codling moth
Leaf roller complex
Cabbage looper
High chance High chance
High chance High chance
Medium chance High chance
Low chance Medium chance
Low chance Medium chance
Low chance Medium chance
Low chance Medium chance
3
-------
feasibility parameters of this report (although many of the critical
economic factors cited are likely to apply, particularly for agents that
require registration).
C.
Profit as an Incentive
The preceding conclusions were largely derived from the economic
analyses of new pesticide product ventures that private industry would
employ to screen for the most attractive of several alternative invest-
ment opportunities. This is a critical point. While specific materials
in the classes of products cited in I-B, "Innovative Pesticides," might
be quite safe, efficacious, useful, marketable, and profitable, industry
will not allocate the funds needed for the rapid development of such
innovative products if it perceives other available business options as
more attractive. In this respect it is interesting to compare the
expected net present values (pre-tax) for investments in a selected
chemical pesticide with those for the "best chance" bacteria, pheromone,
and virus products defined in this study.
Product
Chemical: TH 6040 (Dimilin®)
Bacteria: B.t.
Pheromone: gossyplure for
control of pink bollworm
Virus: Heliothis spp. NPV
Expected Net Present Value
(pre-tax)a
At Eight Percent At Fifteen Percent
$54.5 million
8.8 million
2.7 million
-2.7 million*5
$18.8 million
-6.0 million
0.0 million
-4.1 millionb
aSee IV-B-1, "Problem Structuring," for a discussion of the interpreta-
tion of expected net present value. Many chemical companies have been
using pretax discount rates of 15-20% as investment criteria.
^See note a. on p. 147.
4
-------
These figures and the specifics of their interpretation and derivation
are explained in detail in other portions of this report; they are based
on nonproprietary data and current (1974-1975) levels of information and
uncertainty, and are subject to challenge on the basis of alternative
input values. However, the relationship of the figures is more critical
than their absolute values in the context of the report. For example,
even if the net expected present value for TH 6040 were reduced by 80%,
it would still appear to be the more attractive investment alternative
on an expected value basis and would probably still be emphasized in a
private company funding program. Chemical pesticides with profitability
levels as high or higher than the level shown for TH 6040 already exist.
As long as industry perceives that there are more of these chemical-
based "winners" to be found, it will exhibit a general tendency to search
for them in preference to other types of control agents. It is recog-
nized, however, that some innovative, non-conventional products could be
promoted to a commercial position without meeting the commercial feasi-
bility standards defined in this report. This will probably occur, and
will be more likely with microbial and nonpatentable agents than for
products where a proprietary position can be established. Such occur-
rence is most likely when (1) a firm is unaware of the true commercial
feasibility and risk associated with a particular venture, (2) a firm is
risk-oriented, (3) a firm chooses to develop and promote a venture for
other than purely economic reasons, or (4) the project is largely in the
public sphere.
D. Risk as a Disincentive
It has been argued that (1) corporate size may influence which pri-
vate firms will develop what products, and (2) even if major firms con-
tinue to emphasize the development of basic chemicals, smaller companies
with greater capacity for innovation might play a significant role in
the development of agents such as bacteria, viruses, and pheromones.
This is certainly a possibility, but application of risk theory to the
various financial analyses in this study indicates that this would be
the exception rather than the rule—unless the major elements of risk
currently borne by private industry in these ventures can be shifted to
the public sector. Probability distributions on possible loss or profit
were calculated for each of the four case studies in this report and are
described in detail in other chapters. In almost all cases there were
measurable probabilities of losing at least $2 million, and in some
cases there were measurable probabilities of losing considerably more.
On this basis, the level of risk was judged to be beyond what small
firms would be willing to undertake—assuming they were actually aware
of the true risk involved or were basically not high-risk oriented.
Based on a conclusion drawn from the gossyplure analysis in VIII,
"Pheromones," some pheromones may be a general exception to this rule;
this statement also applies specifically to the chemical-producing level
of the pheromone business. While the probability of high gain in the
gossyplure example was not large, the chance of significant loss was
5
-------
also minimal, primarily because of an assumed high probability for
registration and the relatively low requirement for capital and operat-
ing costs (due to the small volumes of product required).
E. User-Assessed Innovative Pesticide Trends
The foregoing judgements result from analyses of economic attrac-:
tiveness and risk from the private industry product development/
investment/supply point of view (detailed analyses appear in later sec-
tions). An alternative attempt was made to assess the user's view of
innovative pesticide commercial feasibility/probability. This research
was based on a survey of over 170 pest control specialists for 15 crops
in 34 states. The results of this survey, which appear in Appendix B,
were unfortunately inconclusive because (1) only 40% of those surveyed
responded, (2) several respondents felt that speculating about trends
in the introduction of future pest control products and methods was
inappropriate and potentially misleading, and (3) other respondents were
not intimately familiar with work on any specific innovative materials.
Despite these qualifications, some useful inferences can be drawn:
Conventional Versus Innovative Pesticides—While panel members antic-
ipated some innovative materials for pest control over the next decade,
they expected many to be chemically based (e.g., chitin inhibitors, ovi-
cides); few specific bacterial, pheromonal, viral, or additional innova-
tive agents other than those identified in preceding paragraphs were
cited as potential pest control products. The majority of respondents
believed that between 1975 and 1985 these latter alternative controls
would not be usetf in significant amounts. Evolving resistance by spe-
cific pests to existing pesticides appeared to be a primary basis for
many members' anticipation of new products.
Chemical Pesticide Costs—There was a consensus that chemical pest
control material costs would escalate in the future. Almost all panel
members cited anticipated increases in basic petroleum and derivative
petrochemical costs as the primary reason for these increases (this
reasoning represents a panel consensus rather than a categorical state-
ment of SRI's point of view).
Demand for Pest Control—Panel members generally expected that
absolute demand for pest control in terms of acres receiving some form
of control would increase. Reasons expressed included greater substitu-
tion of capital for labor, higher crop valuations, and aversion to risk
of losses (especially in relation to high fixed cost production bases);
there was some variation in this statement, depending on the crop con-
sidered. Despite this general agreement, panel members' opinions on
the relationship between overall per-acre' pest control costs and total
per-acre crop production costs over the next decade differed
6
-------
significantly. Some variation in response related to differences in
pests, crops, and regions, but often there was simple divergence- of
opinion.
Some respondents argued that per-acre pest control costs would
increase as a percent of total per-acre production costs because material
prices would escalate faster than average per-acre crop values. Others
felt strongly that further increases in material costs would only expand
the role of extension or private pest management specialists in devising
and implementing economically efficacious control techniques that would
minimize reliance on expensive materials. Continued development of pest-
resistant crop strains was also cited as having potential to reduce
pesticide usage.
7
-------
II CRITICAL FACTORS IN PESTICIDE PRODUCT DEVELOPMENT
A. Critical Factor Identification
The preceding conclusions warrant a review of the factors determined
in this study to be most critical to the availability of private in-
dustry funding for new pesticide R&D—innovative or otherwise. Table 1
summarizes many of the significant factors that affect the assessment of
a new pesticide's commercial feasibility from an industry viewpoint. The
factors are aggregated into five major groups:
Sales Volume (Market Size)—This factor is the annual volume of
sales, in either units or dollars, which a given company would achieve
in a mature market. Built into this factor is the company's share of
the total market (which could be 100% or less), plus other factors shown
in Table 1.
Selling Price (and Product Margin)—This factor is based on dollars
per unit or per acre-treatment for each product. The selling price is
estimated from production costs, appropriate markups or reasonable mar-
gins in the marketing channel, and distribution costs.
Research and Development Costs—These costs are the sum of all the
expenditures for discovery and market introduction of a new registered
pesticide. Included are isolation, synthesis, screening, testing, pro-
cess and plant design, and registration. Depending on the product, the
costs for complying with registration requirements were generally esti-
mated to run between 15 and 35 percent of the total R&D expenditures.
Receipt of Registration—Since a considerable amount of R&D money
is invested before a product Is submitted for registration, and since
there is no guarantee that registration will be approved, the ability to
register a product in a "yes-no" sense is included as one of the major
variables. For this purpose, all monies invested in the product up to
that point are considered sunk and lost if registration is denied. (How-
ever, some valuable information usable In the development of subsequent
products may have been obtained from the development of the original
product up to the request for registration.)
Capital Costs—These are the costs associated with the construction
of a production facility of economic size, and they depend primarily on
9
-------
Table 1
MAJOR FACTORS AFFECTING THE COMMERCIAL FEASIBILITY OF NEW PESTICIDE PRODUCTS
Sales Volume
(market size)
Storage stability
Field stability
Persistence
Efficacy of control
Sophistication of user
Competitive products
Product life cycle
Mode of action
Strain selecting potential
Host specificity
Public image
Type of user
User attitudes
Deregistration of competi-
tive products
Development of new products
Government programs
Selling price
Relative price
Application methods
Resistance
Crop value
Selling Price
(and product margin)
R&D costs
Production costs
Value to user
Patentability or royalties
Distribution costs
Prices for competitive
products (availability)
Licensing fe.es (if any)
Supply-demand balances
Type of buyer
Type of contract
R&D Costs
Federal research
support
Time required for
registration
Toxicity factors
Field testing
Activity screening
Synthesis research
Cost of failures
Staff (ingenuity,
experience, luck)
Manufacturing pro-
cess complexity
Production methods
Application methods
Compatibility
Receipt of
Registration
Efficacy of control
Host specificity
Toxicity factors
Compatibility with
other pesticides
Persistence
Capital and
Production Costs
Manufacturing process complexity
Sales volume
Process R&D
Raw material cost and availability
Pollution potential
-------
the complexity of the manufacturing process. Economies of scale can be
accounted for in the decision analysis model as plant capacity adjusts
to changes in market size.
B. Selected Critical Factor Sensitivity Analyses
The relative importance of each of these five factors to four prod-
ucts studied in this project is shown in Tables 2 through 5. Each
table gives the base case net present value for the base case scenario
for each product in question, and the amount by which these base case
values are altered by designated changes in the five variables. (Base
cases are only one of several scenarios developed for each product and
are not necessarily the most probable; see the "Method of Approach"
section and the individual chapters for additional detail.)
C. Discussion
General categorical extrapolations of critical factor evaluation
from the case histroies and other data in this study to all products
in the classes of materials under consideration would be tenuous and
could easily be misleading. However, the following general observations,
based on SRI's knowledge of the industry and the findings of this study,
can be made about the various factors that influence economic incentives
to develop pesticide products in private industry:
1. Market Size (Sales Volume)
As shown in Tables 2 through'5, assuming that product margins
are sufficient to cover fixed and variable costs, variations in market
size have sizable leverage on expected gross revenue calculations, and,
therefore, on the apparent commercial feasibility of ventures. Although
costs associated with R&D and plant capital are large, overall venture
profitability appears to be less sensitive to variations in these factors
than to variations in market size (and selling price/product margin).
The following inferences and deductions can be made from this analysis.
Range of Product Activity and Target Hosts—The broader the
spectrum of product activity and/or range of the target hosts, the larger
the economic incentive for product development. These are factors over
which government appears to have little direct control. However, pub-
licly funded activities which serve to ultimately diminish basic market
size (e.g., "eradication" programs; establishment of pest control dis-
tricts where the free market utilization of chemical or other products
is restricted by edict; natural predator or insect sterilization pro-
grams) would affect estimates of obtainable market size for a new product
11
-------
Table 2
RELATIVE IMPORTANCE OF FACTORS AFFECTING
THE COMMERCIAL FEASIBILITY OF BACILLUS THURINGIENSIS
Factor
Factor Value
Net Present Value
at Eight Percent
(millions of dollars)
Percent Change in
Input Variable*
Percent
Change in Net
Present Value*
Average Percent
Change of Net Present
Value Given 1Z Change
in Input Variablet
Market size
Base case
Test 1
Test 2
Selling price
Base case
Test 1
Test 2
Registration
Base use
Test 1
Resea-h and development
costs
Base case
Test 1
Test 2
Plant capital costs
Base case
Test 1
Test 2
8.20 million lb
2.74 million lb
12.30 million lb
$5.12/lb
$4.92/lb
$5.82/lb
Obtained
Not obtained
$6.4 million
$4.3 million
$8.6 million
$2.9 million
$2.0 million
$6.0 million
$23.1
3.8
41.0
23.1
19.5
35.6
23.1
-4.4
23.1
25.3
20.9
23.1
26.0
14.0
N/A
-66 X
50
N/A
-4.02
14.0
N/A
N/A
N/A
-33
33
N/A
-31
107
N/A
-84 Z
77
N/A
-15.6
54.1
N/A
N/A
N/A
9.5
-9.5
N/A
12.5
-39.4
N/A
1.27
1.54
N/A
3.9
3.86
N/A
N/A
N/A
-0.288
-0.288
N/A
-0.403
-0.368
N/A ~ not applicable
*(Test case - base case) ? base case.
Percent change in net present value * percent change in input variable.
-------
Table 3
RELATIVE IMPORTANCE OF FACTORS AFFECTING
THE COMCRCIAL FEASIBILITY OF HELIOTHIS SPP. NPV
Factor
Factor Value
Met Present Value
at Eight Percent
(Billions of dollars)
Percent Change in
Input Variable*
Percent
Change in Net
Present Value*
Average Percent
Change of Net Present
Value Clven 1Z Change
In Input Variablet
U
Market sice
Base case
Test 1
Test 2
Test 3
Product aarginf
Base case
Test 1
Test 2
legistration
Test 1 -
ieaearch and development
costs$
Base case
Test 1
Plant capital costs?
Base case
Test 1
Test 2
5.0 Billion acre-
treataents
1.0 •illion acre-
trea Meats
2.0 ailllon acre-
treataents
10.0 Billion acre-
treataents
$0.60 per acre-
t rest Bent
$0.10 per acre-
treataent
$0.30 per acre-
treataent
Obtained
Mot obtained
$3 aillion
$6 Billion
$6 Billion
$4 Billion
$8 Billion
$10.9
-0.6
2.3
25.8
10.9
-5.2
1.4
10.9
-2.7
10.0
8.2
10.9
12.5
9.3
N/A
-80X
-60
100
N/A
-83
-50
N/A
N/A
N/A
100
N/A
-34
34
N/A
-1052
-79
137
N/A
-147
-87
N/A
N/A
N/A
-25
N/A
15
-15
N/A
1.31X
1.32
1.37
N/A
1.77
1.74
N/A
N/A
N/A
-0.25
N/A
-0.44
-0.44
N/A ¦ not applicable
*
(Test case - base case) t base case.
^Percent change in net present value * percent change in input variable.
tSee note on p. 147.
-------
Table 4
RELATIVE IMPORTANCE OF FACTORS AFFECTING
THE COMMERCIAL FEASIBILITY OF GOSSYPLURE
Factor
Factor Value
Net Present Value
at Eight Percent
(millions of dollars)
Percent Change in
Input Variable*
Percent
Change in Net
Present Value*
Average Percent
Change of Net Present
Value Given 1% Change
in Input Variablet
Market size
Base case
Test 1
Test 2
Test 3
Test 4
Selling price
Base case
Test 1
Test 2
Registration
Base case
Test 1
Research and developnent
costs
Base case
Test 1
Test 2
Plant capital costs
Base case
Test 1
Test 2
50,000 acres
10,000 acres
250,000 acres
500,000 acres
1,000,000 acres
$3.50/gram
$2.00/gram
$5.00/gram
Obtained
Not obtained
$l.i aillion
$0.9 million
$1.6 Billion
$0
$ 500,000
$1,000,000
$ 1.3
0.3
10.7
22.5
46.1
1.3
-0.8
3.3
1.3
-1.1
1.3
1.5
0.7
1.3
0.8
0.4
N/A
-802
400
900
1900
N/A
-43
43
N/A
N/A
N/A
-18
45
N/A
N/A
N/A
N/A
-54%
723
1631
3446
N/A
161
154
N/A
N/A
N/A
15
46
N/A
-38
50
N/A
0.67%
1.80
1.81
1.81
N/A
3.74
3.58
N/A
N/A
N/A
-0.83
-1.02
N/A
N/A
N/A
N/A - not applicable
*
(Test case - base case) * base case.
Percent change in net present value t percent change in input variable.
-------
Table 5
RELATIVE IMPORTANCE OF FACTORS AFFECTING
THE COMMERCIAL FEASIBILITY OF TH 6040
Factor
Factor Value
Net Present Value
at Eight Percent
(millions of dollars)
Percent Change in
Input Vairable*
Percent
Change in Net
Present Value*
Average Percent
Change of Net Present
Value Given 1Z Change
in Input Variable^
Market size
Base case
Test 1
Test 2
Selling price
Base case
Test 1
Test 2
Registration
Base case
Test 1
Plant capital costs
Base case
Test 1
Test 2
Research and development
costs
Base case
Test 1
Test 2
3.8 Billion lb
1.9 million lb
7.6 million lb
$16/lb
$13/lb
$10/lb
Obtained
Not obtained
$2.25 million
$1.50 million
$5.00 million
$ 7.4 million
$ 5.0 million
$10.0 million
$101.2
45.9
212.4
101.2
64.6
37.1
101.2
-7.7
101.2
103.4
99.1
101.2
103.8
98.4
N/A
-50Z
100
N/A
-19
-38
N/A
N/A
N/A
-34
22
N/A
-32
35
N/A
-55 Z
110
N/A
36
63
N/A
N/A
N/A
2
-2
N/A
2.5
2.8
N/A
1.1 Z
1.1
N/A
1.89
1.66
N/A
N/A
N/A
.058
.016
N/A
-.078
-.080
N/A * not applicable
*
(Test case - base case) t base case.
Percent change in net present value ^ percent change in input variable.
-------
being considered for development*. (Acreage changes influenced or
promulgated by the government would also have an effect.)
Anticipated Company Market Share—Anticipated company market
share within a particular pest control market is a critical factor that
is influenced by public policy primarily in the areas of (1) availability
of competitive products, and (2) securing proprietary positions for in-
dividual products. EPA directly influences the availability of competi-
tive materials through its registration and deregistration procedures.
Over and above a firm's basic gamble on receiving a registration (see
"Receipt of Registration" under II--C, "Discussion"), the type of regis-
tration given (e.g., general, restricted) impacts market share. The
type of registration assigned is largely a function of the physical
properties of the product as they relate to short term health hazards.
Thus, the type of registration can be assessed with reasonable accuracy
during early stages of product development. Even though these parameters
could be modified, broadening the range of products falling into the
restricted classification, it would not assure that the trend in product
development would shift away from a chemical orientation.
In recent years, the effects of product deregistration on the
market shares for certain products have been significant—for example,
deregistrations of DDT and other chlorinated hydrocarbons have expanded
*While the evolution of integrated pest management (IPM) techniques was
not reviewed in depth during this project, the possibility that such
techniques may create opportunities for innovative agents such as micro-
bials, natural parasites, and predators was considered. SRI concluded
that IPM does create an environment in the field for the use of such
materials—assuming they are available. However, this environment does
not necessarily translate into attractive business options for the pri-
vate pesticide industry. Unless the right combinations of market size/
sales volume, product margin, efficacy, usefulness, and other essential
factors are present, need does not automatically equate with business
opportunity.
These observations apply generally to situations where the private
pesticide industry assumes a major portion of the risk in sunk costs.
The probability of industry involvement in developing products for use
in IPM programs should increase if this risk can be minimized. For
example, industry already participates in the supply of selected in-
novative materials (e.g., certain pheromones and viruses) on a contract
basis to various public agencies that have borne the major burden of
product and market development. Such arrangements are likely to in-
crease in direct proportion with increases in public participation in,
and field-level control of, pest management practices.
16
-------
available markets for replacement products. Continued cancellations
would further influence future market shares for both existing and devel-
opmental products. They could also have an impact on new product devel-
opment in general. For example, if deregistration actions are prolonged
and severe, they could seriously shrink the private pesticide industry's
commitment to R&D. If cancellations remain concentrated in specific
areas of the industry, they could similarly contribute to a shrinking
of private R&D funding in those areas due to the perception of in-
creased risk. Cancellations will not, however, necessarily increase
private R&D for innovative products unless industry foresees an accept-
able rate of economic return.
Securing Proprietary Product Positions—As mentioned earlier,
the ability to secure proprietary positions for individual products, and
the related issues of patentability and product licensing affect market
share. It is acknowledged that (1) there are many products in other
industries where a proprietary position is not essential (e.g., ammonia,
grain marketing), and (2) even within specific pesticide market areas
there is significant competition between proprietary materials with
parallel activity (e.g., the corn herbicide market). This generally
indicates industry willingness to compete within parameters such as prod-
uct efficacy, marketing and management skills, and luck of discovery.
However, the registration requirements for pesticides help to create an
environment where the availability of patent (or licensing) protection
can be a significant factor in product development economics. Basically,
companies are generally unwilling to invest $5± million of front-end
money in product R&D and registration with no assurance that a competitor
will not subsequently enter the same product market without having made
a similar investment of time and working capital; the second entrant's
financial edge in product margin (which in turn can influence selling
price and market share) Is greater than the industry has been willing
to accept.
This is particularly true when the size of the total available
market is relatively small. Proprietary protection of R&D and registra-
tion investment appears more critical for small markets because of the
relatively low ratio of market size to sunk costs and the effect of time
value on a longer payout period. Thus, for potential products with
(1) inherently small markets, and (2) limited or no ability to obtain
proprietary position by patent or special licensing (e.g., microbials,
naturally occurring products), the relationship of risk to potential
project profitability is likely to be too great for the norm of private
industry to consider.
Therefore, any actions that government can take to protect ma-
terials outside present laws (e.g., microbials and other naturally oc-
curring products, products developed by public agencies) would increase
industry interest in participating in their development.
17
-------
2. Selling Price and Product Margin
SRI ascertained that probable selling prices for the products
studied in this report could be in the range of those for competitive
materials (these products would not be far along in development if their
apparent costs were prohibitive). The product margins associated with
these selling prices are defined in later chapters; while these margins
have not been definitively compared with those for all registered pesti-
cide products, SRI believes that they are comparable on a percentage
basis with general industry practice for proprietary materials.
As with market size, selling price—specifically as it influ-
ences product margin—has a major effect on potential venture attractive-
ness, and there are areas where margins might be influenced to improve
venture profitability. Selling price and margin are a function of sunk
costs (R&D and registration), production costs (raw material and other
variable costs, and capital costs), plus licensing (if any), formulation,
distribution, and other marketing cost$. They are also influenced by
product efficacy and by costs of competitive controls. The primary areas
of government influence are (1) R&D requirements for registration, and
(2) the availability and related cost of competitive controls. To the
extent that sunk costs can be shifted from the private to the public
sector, product margin and expected venture profitability are improved,
as illustrated below for B.t., gossyplure, Heliothis spp. NPV, and TH
6040.
Expected Net Present Value
at Eight Percent
(millions of dollars)
Company Pays Company Pays
Full R&D Costs No R&D Costs
B.t. $ 8.8 $15.8
Heliothis spp. NPV -2.7a 1.8a
Gossyplure 2.7 4.0
TH 6040 54.5 62.9
See note a. on p. 147.
The assumption of these development costs by the public would not only
improve the expected profitability of the above ventures, but it would
also minimize risk, which, as discussed in I-D, "Risk as a Disincentive,"
represents a primary disincentive to smaller companies that might be
potential developers of selected innovative materials. In most of the
scenarios reviewed for the four case studies in this project, the pre-
dominant risk was associated with sunk costs that could not be recovered
if registration were not obtained. Thus, any defendable reduction of this
risk load is, theoretically, an industry incentive, particularly for
smaller firms.
18
-------
3. Research and Development Costs
As discussed earlier, R&D funds constitute sunk costs expended
before the product is marketed, and if the product fails, the company
essentially loses these funds. For most pesticide products this loss
ranges between $2 million and $10 million and cannot be easily borne by
firms with low levels of capitalization. (A small company might reduce
its risk by entering a joint venture, but unless the project had an at-
tractive expected value, the firm might have difficulty finding partners.)
It is important to note, however, that while R&D costs and
their leverage on risk may be large on an absolute basis, their relative
impact on the total profitability of a successful venture is usually
small in comparison with other factors such as market size, price, and
margin—although margin includes the recovery of R&D expenses. It is
noted from the tabulation in the last subsection, however, that profit
sensitivity to R&D costs does increase as the expected net present value
decreases. This reinforces earlier statements that any defendable re-
duction in R&D would likely serve as an industry incentive in the devel-
opment of innovative products with small markets.
For the products studied in this report, registration costs
(including toxicology testing) ranged between 15 and 35 percent of the
total R&D cost and the overall impact of delaying registration was rela-
tively small compared to other factors. However, several analyses re-
vealed a significantly greater effect as the profitability of a project
increased, since successful venture scenarios lost the time value of the
profits that would have otherwise been realized in the years without
delay. This is illustrated below in four scenarios for a possible
Heliothis spp. NPV venture.
Net Present Value
at Eight Percent
(millions of dollars)
Small Market Large Market
Registration in 1978 -$3.9 $25.8
Registration in 1982 -$4.3 $17.7
£
See note a. on p. 147.
The relatively small difference in the net present values of the two
small market scenarios is almost entirely comprised of the difference in
the discounted value of the R&D expenditures, while the bulk of the $8.1
million difference between the two large market scenarios is the result
of the revenue lost during the four-year registration delay.
19
-------
Companies tend to be optimistic and eager to market products
that reach the registration process. Therefore, any shortening of the
registration process (a minimization of delays) would be well received
by the industry.
4. Receipt of Registration
As with any pesticide product, receipt of registration is a
major concern to all potential producers of innovative pesticide mate-
rials. However, for developers of materials with narrow spectrum or
relatively small markets, this uncertainty may be a particularly great
deterrent—a company might be willing to risk $5 to 10 million for a
product with an expected net present value of $50 million, but it may
not be willing to do so for a venture with an expected net present value
of only $10 million. Registration criteria generally relate to product
efficacy and safety, and industry's willingness to risk R&D funds on the
development of a product through this process is obviously enhanced if
the basic registration criteria are relatively stable. However, product
development through registration frequently takes seven to nine years,
a long time in an environment of fairly rapid technological change. If
the criteria are subject to change after a company has chosen to risk
front-end R&D money under an assumed set of criteria, an additional
element of risk is introduced that will affect the initial decision
making process. The lack of guidelines for the registration of poten-
tial innovative pesticides such as viruses and pheromones has probably
been a disincentive to private industry development. Even as guidelines
are established, some related disincentive may remain because of the pos-
sibility of alteration as new evidence is compiled; this may be particu-
larly true for viruses*.
5. Capital Costs
Capital costs represent front-end expenditures that must be
recovered if a venture is to be profitable. Available evidence indicates
that these costs may be potentially less for innovative materials requir-
ing few production steps, such as bacteria and possibly viruses, relative
to many conventional products. Offsetting this potential advantage is
the fact that products with small markets typically offer smaller levels
of profit; in these cases, any significant upward variations in antici-
pated capital costs have a larger impact on profit. For example, a
$4 million increase in capital costs for a B.t. venture as defined in
*As of 1976, EPA's Office of Pesticide Programs has obtained data as an
input to the development of registration guidelines for insect growth
regulators and pheromones, and continues to work with industry and the
scientific community toward the establishment of an appropriate set of
such guidelines.
20
-------
this report could lower Its net present value by almost 50 percent,
whereas the same capital cost Increase would only reduce the net present
value of the TH 6040 venture by three to four percent.
At the evaluative stage of a prospective venture, particularly
when innovative materials are involved, a company may be confronted with
a fair degree of uncertainty in estimating capital costs. As R&D devel-
ops, however, research on manufacturing processes progressively reduces
most of this uncertainty. Therefore, by the time a registration is
applied for, the company is reasonably assured that its latest capital
cost estimates are quite accurate. However, because of the length of
time required for development through registration, an amount of uncer-
tainty will remain about actual inflationary trends and their effect on
construction costs (few observers eight to ten years ago forecast the
degree to which plant construction costs would escalate over that time
frame, particularly in the past five years). In this context, delays
in registration associated with changing guidelines can become expensive
at the capital cost level, particularly in times of high inflation.
21
-------
Ill RECOMMENDATIONS
Summarizing from all the preceding, the following two factors have
generally been limiting on private industry's willingness to invest in
the type of innovative pesticides discussed in this report (limiting,
that is, over and above specific technical and cost factors associated
with individual products):
• Narrow spectrum, limiting market size
• Lack of proprietary protection.
The effect of these factors, singly, or together, has been (1) to gen-
erate profitability and profitability to risk scenarios that do not ap-
pear competitive with other pesticide R&D investment alternatives, and
(2) to amplify the already-sizeable risk factors for small firms that
might otherwise be expected to be innovative.
If it is advisable to stimulate private industry development of
pest control products affected by these factors, then the following
concepts will have to be changed.
1. Proprietary Protection
Under present law, naturally-occurring substances and microbes
are not patentable as new compositions (although new synthetic mimics
and new mutants of microbes may be patentable). In addition, products
developed by U.S. governmental agencies have generally not been available
for private patent or licensing under current practice. (NIH is at-
tempting to allow some products developed with its funds to be avail-
able for further development by industry on a licensing basis, but ap-
parently this may be tested in court.) As previously mentioned, the
absence of such protection—particularly for products with markets
limited by narrow spectrum—appears to be a major deterrent to serious
industrial development funding, given current practices and costs
sequired for registration and other R&D functions.
Therefore, if the government wishes to encourage such devel-
opment by private industry, it must consider what actions can equitably
be taken to provide required levels of proprietary protection. Such
actions might include exclusive licensing arrangements for products
developed by government programs or auction arrangements for naturally-
occurring agents and materials. For example, the discovery and develop-
ment of new families of synthetic pyrethroid insecticides by govern-
mental employees in England has not impeded the commercial development
23
-------
of these products by private industry, because that country has a policy
permitting semi-exclusive licensing arrangements that assures wide
market exposure plus protection incentive for the investment of develop-
ment funds by the licensors.
2. Risk
It has been determined herein that (1) the absolute size of
risk in innovative pesticide development ventures is generally larger
than what most small companies are likely to assume, and (2) the rela-
tionship of profitability to risk for most narrow spectrum products has
typically been lower than what any company would choose on an alterna-
tive investment basis. Since many of the new innovative pesticides
considered in this study are intrinsically market limited by narrow
spectrum (which thereby minimizes the ability to manipulate the most
important market profitability factors), some steps would have to be
taken to reduce risk in order to stimulate more private industry invest-
ment. Three primary ways in which such risk can be reduced, or profit-
ability to risk relationships improved, are:
• Provide some level of support for R&D costs. As noted,
these costs incurred through the registration process
are the primary funds at risk for both large and small
firms. One type of public support of such costs would
be public funding of the total R&D load for products in
selected categories (e.g., for products of a nonproprie-
tary nature), or public funding of the registration com-
ponent of these costs. Such funding could apply in all
cases, or it might be restricted to situations where
registration is not obtained; application in this latter
case might be considered attractive .assuming that a company
should be able to pay for its successes. (One possible
mechanism might be to have R&D or registration-only costs
financed by the federal government and paid back by a sur-
tax on the sales of the registered product.) However,
this type of public support could ultimately underwrite
and sustain nondiscriminate and nonproductive research
programs in industry beyond the purpose of the "subsidy."
A more suitable alternative might be to promote successful
venture profitability by underwriting R&D costs only for
those products that do receive a registration. (As an
alternative to direct funding, a system of tax credits
for R&D costs might be established.)
• Modify current product performance liability constraints
that limit product development for small markets. Pesti-
cide producers, who are reluctant to register broad spectrum
products for some uses where markets are small compared
to the potential performance liability at risk, may face
24
-------
similar constraints with narrow spectrum products. Thus,
steps that can be taken to insure against this liability
or allocate it more evenly between users, producers, and
pest control professionals at their interface should serve
as an industry incentive.
• Provide proprietary protection for products not now covered
(as already discussed).
SRI views the above as concepts warranting further study,
rather than as absolute recommendations for change. Each of the thoughts
specified, and any modifications thereof, will require extensive legal
and industry impact analyses that go well beyond the scope of this
report. However, a number of problems seem to be apparent. For
example, can special patent and licensing arrangements be equitably
extended to the pesticide industry without also being made available
to other industries? Can public funding of pesticide R&D or registra-
tion be equitably applied on company-size, market-size, or product-
type bases? Is it, in fact, necessary to artificially accelerate pri-
vate industry development of the types of pesticides in question? This
last question may be the most important of all, since it is inherently
directed at the basic structure of the pest control "industry" in this
country and the manner in which recent relationships and allocation of
resources between the public and private sectors have been—and are
likely to continue—changing as a result of (1) EPA pesticide deregis-
tration activities, (2) broader governmental involvement in overall
pest control activities, and (3) cost increases associated with the
shift from random screening to biochemically rational approaches in
new pesticide development.
25
-------
IV METHOD OF APPROACH
The primary objective of this project was to determine the innova-
tive pest control products or techniques most likely to become commerci-
ally available during the next decade, either as substitutes for, or
complements to, existing pesticides. To satisfy the demands of this ob-
jective, the project was directed at answering the following questions:
• What materials have technical potential as innovative pest
control agents?
• What are the critical factors affecting the commercial
(economic) feasibility of these products?
• What are the incentives and disincentives for industry to
commercially develop these products or mechanisms, and
which incentives or disincentives are the most closely
influenced by EPA and the policies of other government
agencies?
The diagram in Figure 1 summarizes the steps and procedures of the
project. The research development generally followed the plan outlined
in the SRI proposal, with modifications and expansions made as necessary
over the course of the project in collaboration with EPA representatives.
The diagram illustrates both the time sequence and the interrelationships
of the activities.*
There were three basic phases of the project: preliminary investi-
gations, decision analysis, and critical factor analysis.
Preliminary Investigations
To provide background for the project, a thorough review was made
of available data on the innovative pest control techniques which are
either available or are at some stage of consideration and might become
commercially available between 1975 and 1985. Information was obtained
principally through literature searches and personal interviews with
experts iri academia, government, and industry.
Six candidate agents were then chosen (with EPA review and approval)
for study preliminary to selection of the products analyzed in this
report: five insecticides—Bacillus thuringiensia, Heliothis spp. NPV,
TH 6040, Altosid®, grandlure—and one herbicide—Colletotrichum
gloeosporioides. These candidates were chosen to provide a range of
(1) synthesis and patent characteristics, '(2) modes of action and types
of pest management systems, and (3) potential economic characteristics.
27
-------
In all cases, some degree of development was preferred to provide a data
base. In addition, the candidates were reviewed as potential substi-
tutes for seven pesticides subject to EPA deregistration review.
B. Decision Analysis
The preliminary review of the candidate products provided valuable
insights into many of the factors affecting commercialization of new
pesticides. The second phase of the project was designed to refine
these insights by analyzing the decision that a company might face if
it were deciding whether to invest in the development of a new pesticide
material. This level of analysis was carried out for four of the six
candidate materials using decision analysis techniques and the best
information available to SRI, as described in IV-D, "Information Quality
and Sources."
1. Problem Structuring
The first step in the decision analysis was to develop a logi-
cal structure that would facilitate analysis of an actual commerciali-
zation decision. Figure 2 summarizes that structure and presents the
major issues and concerns affecting the commercial feasibility of new
pesticides. The diagram identifies three broad areas that affect the
commercial feasibility of a new product: market size and revenue; R&D
costs; and production, distribution, and capital costs.
The principal factors affecting market size are summarized in
the left hand portion of the figure. The maximum possible market size
is a function of the total acreage that is vulnerable to the target pest,
the value of the crop(s), pest control district policies, insurance
programs, export policies, and planting restrictions. It should be
noted that the maximum potential market size may vary over time because
of changes in the number of planted acres, trends of insect resistance,
proportion of acres infested, and price/cost relationships for the crop.
However, the actual market that the company captures will usu-
ally be less than this maximum potential. There may be competition from
different pesticide products; or, if the new product under consideration
is not patentable, there may be competition from other producers of
similar materials. As the figure shows, the market share that the com-
pany finally realizes is dependent on such specific factors as: patent
duration and product characteristics of competitive products; patent-
ability of the new product; registration requirements; and relative cost
effectiveness of the new product. A more detailed description of the
logic affecting these market size and share considerations is given in
Figure 3.
The principal factors affecting R&D costs are summarized in
the upper right portion of Figure 2. In some cases, the original
28
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
background research on a potential pesticide may have been conducted by
an academic institution or government agency; if not, the company would
incur costs for these activities. Once the product idea has been iden-
tified, the company would incur labor, material, and overhead costs as
it performs field and toxicity tests, product and process development,
and demonstration tests. The ^xtent of these costs would depend on the
product in question as well as on outside or collaborative support,
licensing arrangements, and registration requirements. Figure 4 presents
the R&D process in more detail.
The lower central portion of Figure 2 summarizes production,
distribution, and capital costs. These costs are a function of annual
sales levels as well as capital requirements, labor, materials, and
overhead.
The financial model at the far right of Figure 2 collects the
time-varying descriptions of R&D costs, revenue, and operating costs to
generate cash flow over time. This cash flow can then be evaluated for
several different sets of cost and market assumptions to determine the
potential expected profitability of the venture. There are various
methods of evaluating the relative profitability of investment projects.
No single technique suffices when major investment decisions arise, and
many company managements often adopt a combination of economic evalua-
tions to decide whether or not to invest in a new project or which of
several investment alternatives is more attractive. One of these
methods is the conventional return on investment approach which is the
ratio of net return or after-tax profit at capacity operation to gross
fixed assets; variations exist but none recognize or discriminate among
various possibilities in the timing of cash flows.
Another method involves cash recovery time or payout, i.e.,
the time required to repay the original investment assuming a given
sales pattern. Finally, there is the discounted cash flow return and
present worth approach. This method attempts to coordinate the time
value of both expenditures and revenues for a given project. In this
process, future dollars must be discounted to reflect the time value
of money. The rate at which future dollars are discounted is called
the discount rate and it reflects company's time preference for money.
The discount rate used by a particular company is determined by several
factors including the prime lending rate, the inflation rate, and the
nature of the company's investment portfolio, but it should not be
dependent on the company's size.
In this project, the latter method has been employed, using
an expression of before-tax net present value as the basic measure of
profitability. Net present value can be thought of as the amount of
present money that remains after all funds expended on the project—
including capital and on-going expenses—have been recovered at the
minimum desired rate of return. If the project fails to achieve the
desired rate of return (i.e., has a negative net present value), then
35
-------
the company would not invest in the venture and would invest in an
alternative opportunity offering at least the minimum attractive rate
of return.
The basic analyses were carried out under the assumption of
no inflation, using constant 1974 dollars. On this basis a discount
rate of eight percent before taxes was considered the minimum attractive
rate of return for a representative company considering development of
a new pesticide. (In "normal" periods of five percent inflation, the
eight percent rate would be roughly equivalent to a thirteen percent
return.) For comparative purposes, results were also given using a
fifteen percent discount rate; however, based on the assumption of no
inflation, SRI considered an eight percent rate to be the minimum
level some companies might consider for these pesticide ventures.
(Under the assumption of a long term inflation rate of five to eight
percent per year, many chemical companies have been using before-tax
discount rates of approximately 15-20 percent to evaluate new invest-
ment opportunities. Thus, the rates chosen for this study do not
negatively discriminate on the attractiveness of the ventures
analyzed.)
The logic of Figure 2 allowed a straightforward calculation
of the profitability of ventures when all input parameters were speci-
fied. These calculations were computerized to allow rapid evaluation
of alternative ventures and scenarios.
2. Base Case Analysis
In many cases there was considerable uncertainty over the in-
put parameters. Therefore, the initial step in analyzing the apparent
relative commercial attractiveness for the four candidate products was
to develop a nominal base case for each, using most-likely estimates
for the various parameters based on literature and interviews with
experts in the field. The base case provided a preliminary evaluation
of each venture's attractiveness and a reference point for comparing
alternative scenarios. It should be noted that since foreign markets
were considered more likely for some products than others, the four
base case scenarios did not always include foreign sales. However,
foreign markets were considered for all four products in the subsequent
probabilistic analyses in which hundreds of alternative scenarios were
evaluated for each product. Thus, although the base cases were not
always comparable, the expected values derived from the numerous
scenarios were.
3. Sensitivity Analysis
After the nominal base case was developed, sensitivity studies
were conducted to determine the factors that most critically affected
the venture's profitability. The studies were first conducted as single
36
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
value sensitivities and then as joint variables at their nominal levels.
The variables that shifted the project between profitable and nonprofit-
able scenarios were considered the most critical.
4• Probabilistic Analysis
In most cases, four or five uncertain variables that critically
affected the profitability of the project were identified. Subjective
probability estimates for these variables were developed, and these
distributions were used to develop decision trees similar to that in
Figure 5. A decision tree summarizes a company's commercialization de-
cision and shows the scenarios that might result if it entered the new
venture. The profitability and likelihood of each scenario can be eval-
uated by the computerized cash flow model and probability theory, given
specification of the uncertainties for each input variable.
The interested reader is directed to the references in
Section E of this chapter for more complete discussions of decision
analysis, including the techniques of how sequential decisions are
handled. As a result of evaluating the decision tree, a probability
distribution on the profitability of the project is derived (see Figure
6), given current levels of information. Different levels of information
would lead to different assessments of the project's profitability.
The probability distribution on net present value was calcu-
lated by examining several hundred possible scenarios for each venture.
A useful way to summarize the information contained in the probability
distribution is to calculate its average or expected value and obtain
the expected net present value, which expresses the attractiveness of
the venture as a single number. This quantity is calculated by weight-
ing the net present value of each of the scenarios by the likelihood of
its occurrence and then summing over all scenarios. The expected net
present value can be thought of as the average net present value of the
project if the venture were entered several hundred times, realizing
that any one venture might have a net present value significantly dif-
ferent from the average.
In some cases, a venture's profit can vary greatly, depending
on which scenario actually occurs. If some of the scenarios lead to
major losses, but the project looks attractive on an average basis, then
it is necessary to introduce concepts of risk aversion since most com-
panies weigh negative outcomes proportionately heavier than positive
gains. For the purposes of this study, a weighting function on the pos-
sible profit and loss values was used to measure the effect of risk
aversion on new product ventures. The functional form of the trans-
formation from money to this scale may be of an exponential type:
39
-------
Decision to
Fund R&D
R&D
Costs
Successful
Registration
Market
Size
Product
Margin
Profit ($)
Develop
Product
Medium
Do Not
Develop
FIGURE 5 SIMPLIFIED DECISION TREE: NEW PESTICIDE COMMERCIALIZATION
DECISION
40
-------
1.0
0.6
>
t-
O
<
ED
o
a
0.4
-5
-4
3
¦2
0
1
6
8
18
1
2
3
4
5
7
9
10
12
13
16
17
11
14
15
BEFORE-TAX NET PRESENT VALUE OF PROFIT AT 8% — millions of dollars
FIGURE 6 SAMPLE PROBABILITY DISTRIBUTION OF PROFITABILITY OF NEW PESTICIDE PRODUCT VENTURE
-------
1 - e
-x
u(x) =
1 - e
where x is money, and o is a "risk tolerance" parameter that character-
izes the strength of the desire to avoid risk.
It is possible to make a simple measurement of the risk at-
titude coefficient, a, in the above equation. If a man is indifferent
to playing or not playing a game where he calls a coin toss to win $100
or lose $50 then his risk tolerance is $100. In other words, when an
individual reaches indifference about playing or not playing a 50-50
game in which he either wins a or loses a/2, his risk tolerance is a.
Underlying this method of stating risk aversion are the con-
cepts of certain equivalence and utility theory. A company's "certain
equivalent" of a venture is the amount of money for which it would sell
the opportunity to enter the venture. In most cases a company is willing
to sell a-risky venture for an amount of money less than its expected
value. (This assumes that a firm is willing to take less than it would
earn on the average in order to avoid the possibility of loss, since
such decisions are usually not made only on the basis of favorable or
unfavorable expected values, and since losses are weighted more heavily
than gains.) Using utility theory, gains and losses stated in monetary
terms are replaced by values which are distorted to reflect these ele-
ments of subjective judgement.
Using this "utility scale," the average utility "of the lottery
can be found and a decision made, based on whether this number is greater
or less than zero. By translating back from the average utility number
to the money scale, the "certain equivalent" discussed above is then
determined. Thus the certain equivalent of a profit lottery depends on
both the lottery itself and the company's attitude on taking risks.
The references at the end of this chapter are recommended for
the reader interested in further detail on these theories.
C. Critical Factor Analysis
As a final step in the project, the major factors which act as the
incentives and disincentives for private industry to pursue the commer-
cialization of the new, innovative pesticides were delineated and ana-
lyzed. The most noticeable finding in this analysis was the similarity
between limiting factors for the different pest control materials
studied, as discussed in Chapter II.
42
-------
D. Information Quality and Sources
The elementary requirements for successful commercial development
of a pesticide by private industry are its usefulness and profitability,
with the ultimate economic measure of success being the return on invest-
ment (ROI). The manner in which these basic criteria are applied in
evaluating a product varies with th^ degree of the product's maturity.
There are three levels of maturity: (1) the evaluative (the preliminary
evaluation stage that follows successful screening), (2) the develop-
mental, and (3) the commercial. If a product is already commercial,
hard data based on experience should be available to provide a relative
optimum basis for ROI calculations. However, when potential products
are developmental or evaluative, there are usually far less accurate
data available, and in many instances there are no hard data of any
quality.
The data constraints on a study of new, innovative pesticide devel-
opment by 1985 are those associated with developmental and evaluative
materials; even in the case of an already commercial product (e.g.,
Bacillus thuringiensis), a five to ten year extrapolation based on
experience and involving more than one company contains significant
uncertainty. The use of decision analysis techniques that a typical
company might employ to place uncertainties into a reasonably manage-
able framework was discussed earlier.
Two basic methods were used to gather information: literature
search and interviews and surveys. An extensive review of the litera-
ture was conducted, primarily during the initial phases of the project.
A bibliography of selected pertinent references appears at the end of
each chapter.* To supplement this data base and to place the findings
on each of the four"products studied with decision analysis techniques
in perspective with similar materials, direct input was received from
over 180 academic, government, and industry experts (listed in Appendix
A). These contacts were made to obtain as much firsthand, unpublished
data as possible, and to ascertain the futures such experts envisioned
for selected new pesticide products.
Included are personal interviews with the heads of R&D departments
of nine large U.S. pesticide producing companies. A particular purpose
of these interviews was to:
*These references provided useful background but were not extensively
used as sources of specific data in this report. Therefore, scientific
notation or referencing of published sources has not been employed
herein. (Most of the data in this report are based on interviews or
are derived SRI estimates.)
43
-------
• Ascertain the companies' levels of interest in, and finan-
cial commitment to, innovative pesticide R&D.
• Determine industry opinions on the factors affecting com-
mercial development of the innovative pest control products
emphasized in this report.
• Make a preliminary assessment of any new products being
developed and their rate of introduction into commercial
markets over the next decade.
For selected groups of products, questionnaire surveys were con-
ducted in addition to personal interviews. Another survey on crop/pest
characteristics and trends for 15 major crops was conducted with state
entomologists and pathologists. The purpose of this survey was to:
(1) identify present pest problem areas and future trends in those
areas, (2) quantify acreages associated with the crops affected by the
pest, and (3) identify the new, innovative techniques of pest control
anticipated by users.
E* Selected References
The following is a list of pertinent references on decision analysis
and risk theory.
Howard, R. A., "The Foundations of Decision Analysis," IEEE Transactions
on Systems Science and Cybernetics, Vol. 55C-Y, No. 3
(September 1968).
Howard, R. A., "Decision Analysis in Systems Engineering," Systems
Concepts; Lectures on Contemporary Approaches to Systems,
Chapter 4, Wiley-Interscience, New York, pp. 51-85 (May 1973).
North, D. W., "A Tutorial Introduction to Decision Theory," IEEE
Transactions on Systems Science and Cybernetics, Vol. 5CC-4,
No. 3 (September 1968).
Spetzler, C. S., "The Development of Corporate Risk Policy for Capital
Investment Decision," IEEE Transactions on Systems Science and
Cybernetics, Vol. 5CC-4, No. 3 (September 1968).
44
-------
V LIMITS OF INTERPRETATION
Since this study is likely to find a readership beyond its original
audience, its qualifications must be emphasized; in addition to this
discussion, many of these qualifications are delineated throughout the
study. (Potential readers addressed here include scientists from
several disciplines [e.g., chemists, entomologists, pest control spe-
cialists], economists, politicians, agriculturalists, and bureaucrats;
R&D specialists, commercial development and marketing personnel, and
managers; and private, academic, and government personnel.)v
• The primary objectives of the study were: (1) to generally
forecast the most probable rate at which new, innovative
types of pesticides might be introduced into the commer-
cial marketplace during the next decade, and (2) to identify
and evaluate the critical factors that constitute the
primary incentives and disincentives for industry to
develop such products. It is in this spirit that the
research was conducted and that the report should be
read.
• In no manner did the research critically determine the
efficacy, usefulness, or need for any specific materials.
While certain agents were evaluated during the study, they
only served as specific examples to identify, isolate, and
analyze critical factor and problem areas for broad types
of innovative pesticides. The product analyses in this
report are not definitive, categorical statements about
the agents reviewed; nothing in this document should be
construed as information to qualify or disqualify any
product for registration.
• Data limitations were acknowledged earlier in IV-D, "Method
of Approach." SRI made every attempt within the project
budget to obtain reliable information and guidance from
experts in their respective fields. However, for some
products SRI did not have access to specific company-
proprietary data, and the Institute recognizes that
selected conclusions and product analyses are subject to
challenge on the basis of alternative input values and on
new data availability.* As part of this project, EPA was
*See note a. on p. 147.
45
-------
provided with the basic model for each of the four specific
product analyses, and individuals who wish to test alter-
native data may do so by contacting the EPA project officer.
• It was impossible to critically screen—or to even identify—
all possible innovative pest control materials that might
enter, and commercially emerge from, public and private R&D
programs during the next ten years. A categorical extrap-
olation to all classes of products from the report's find-
ings on product introduction and its evaluation of critical
factors influencing commercial development, or any other
out-of-context interpretation based on the case histories
and other data in this report, would be tenuous and possibly
misleading. However, viewed within the context of its basic
objectives and qualifications, this study presents a useful
reading on the problems, incentives, and disincentives as-
sociated with new innovative pesticide development.
• This study focused primarily on petet control in end-use
areas where a relatively free system of product develop-
ment and competitive markets attracts private industry
R&D investment; in most cases these use areas are agri-
culturally oriented markets. Therefore, the conclusions
and inferences in the study are not necessarily applicable
to other markets, such as public health, where public R&D
and control practices are dominant.
46
-------
VI BACTERIA
A. Overview
Under natural conditions, certain bacteria cause insect disease
and death. These pathogens have therefore become of interest to prac-
tical entomologists as potential commercial pest control devices. Two
such agents—Bacillus thuringiensis (Berliner) (B.t., B.T., B.T.V.) and
Bacillus popillae (Dutky)—have already been commercialized, with B.t.
approaching a volume of 1.5 million pounds per year in the United States,
mostly for agricultural uses. In this project, SRI investigated the
degree to which other bacterial-based pesticides might emerge from
R&D programs as commercial products during the next decade. A summary
of the findings follows:
• There is very small probability that presently known
bacteria, other than the two agents mentioned above,
will become new commercial pesticides by 1985. Of the
products cited by survey respondents as candidates for
potential investigation, Bacillus sphaericus and Bacillus
moritae for use on flies, mosquitos, and gnats were \
mentioned most frequently. However, there appears to
be little work being conducted on these or other non-
commercial bacterial agents.
• Most of the bacterial pesticide R&D work known to be
currently under way within industry, academia, and
government is directed toward the improvement of the
properties of B.t., primarily because the product is
already commercial and its inherent potential (based
on spectrum and target pest range) is so much greater
than that of other known bacterial agents.
• Almost all observers expected total use of B.t. in the
United States to grow fairly significantly in the
future—by at least a factor of three over the next ten
years. Most growth was anticipated in current markets
plus the cotton and forest markets. However, most
observer optimism was associated with anticipated
successes of various R&D efforts to improve active
toxins production by strain selection and to generally
improve the overall efficacy of commeirclal products;
no attempt was made in this study to assess the degree
to which such optimism is warranted.
47
-------
• A financial analysis was made of a hypothetical venture to
develop B.t., even though the product is already commercial.
This analysis provided useful insights on the critical fac-;
tors influencing private industry's decisions to develop
bacterial pesticides. Although the hypothetical venture
offered an expected net present value at eight percent
interest of $8.8 million, it was not without risk. There
was a 45 percent chance of losing money with losses as
high as $10 million. The primary uncertainties concerned
market size and market share, which were significantly
affected by lack of patentability, potential number and
types of competitors, and the number of alternative compet-
itive products.
B. Background
A number of bacteria cause disease in insects under natural condi-
tions. Therefore, it is reasonable to assume that certain of these
bacterial pathogens exert control pressures on insect populations which,
in conjunction with the effects of other pathogens, predators, and
environmental factors serve to maintain a natural balance in undisturbed
ecosystems.
Selected bacterial species that are pathogenic to certain insect
pests in agricultural systems could be developed to replace, supplement,
or reduce the use of chemical pesticides. The question of the commercial
feasibility of such developments is the subject of this chapter.
As mentioned earlier, Bacillus thuringiensis (Berliner) and Bacillus
popillae (Dutky) have already been developed as commercial insecticides
and are currently in use. Bacillus popillae (Dutky), active against
Japanese beetle larvae, is produced by Fairfax Biological Laboratory
under the trade name, Doom ; a closely related species, Bacillus
lentomorbus (Dutky), can also be used. Both cause "milky spore" disease
in larvae of certain closely related beetles of the family Scarabaeidae.
Little is known about the mode of action but the role of a toxin is
strongly suspected.
A far more important species is Bacillus thuringiensis (Berliner),
commonly referred to as B.t., B.T., and B.T.V. (in this report B.t. is
used throughout). Various commercial formulations are marketed in the
United States, by Abbott Laboratories (Dipel®) , Sandoz Incorporated
(Thuricide®), Nutrilite Products Incorporated (Biotrol®), and Rhodia
Incorporated (Bactospeine®). Available formulations include wettable
powders, dusts, liquids, baits, pressurized formulations, and formu-
lations supplemented with pyrethrins. Approximately 1.5 million pounds
of B.t. are now marketed annually in the United States, with a similar
amount marketed outside of the country.
48
-------
B.t. is active against a large number of lepidopterous larvae,
many of which are important economic pests of agricultural, ornamental,
and forest plants. The bacterium does not act as a contact poison, but
must be ingested to be effective. The mode of action is believed to
vary somewhat in different groups of lepidoptera but is largely mediated
through the action of a proteinaceous, crystalline endotoxin produced
within the vegetative bacterial cell in conjunction with sporulation.
Therefore, B.t. is not very different from a chemical insecticide; the
chemical endotoxin is the active ingredient, and the microorganism is
only the "manufacturing plant." Larvae which ingest lethal doses
usually stop feeding within an hour or two, and death follows after two
or more days.
The environmental appeal of bacterial pesticides is partly related
to their derivation from naturally occurring entities. Current commercial
formulations of B.t., at least, are environmentally nonphytotoxic, have
no apparent mammalian toxicity, and do not harm natural predators. In
addition, there is little apparent prospect that insect populations will
be successful in developing resistance to these disease agents. Finally,
they are suitably efficacious within the range of insects controlled,
and considerable promise exists for distinguishing and developing new
strains with higher potency. Disadvantages include a relatively limited
host range compared to those of several popular broad-spectrum chemical
insecticides, lack of patentability due to their natural origin, rela-
tively low field stability and persistence, and application timing
requirements.
Research on the development of bacterial pesticides (principally
B.t.) is being conducted in the United States in a number of laboratories,
including those of the U.S. Department of Agriculture's Agricultural
Research Service, commercial producers, and universities. Developmental
research is also being performed in other countries. Major research
topics addressed are strain differentiation and testing, improvements
in formulations, and improvements in method of application.
The value of B.t. as a case in point for bacterial pesticides was
recognized early in the research program. SRI felt that the problems,
utility, level of commercial success, and future potential demonstrated
by B.t. might indicate the future prospects for the entire field of
bacterial pesticides and provide a guideline for studying other bio-
logical controls.
Considerable insight was gained through a careful examination of
information on the history, properties, production, consumption, eco-
nomics, mode of action, and field characteristics of this established
bacterial pesticide. Information on B.t. was obtained through litera-
ture searches and Interviews with a large number of experts including
commercial producers.
49
-------
C. Panel Survey Analysis
1. Sample Copy
Figure 16 at the end of this chapter is a sample copy of the
bacterial pesticide questionnaire. Thirty-seven were sent out. Twenty-
four experts responded, and eighteen completed the forms (respondents
who did not complete the forms generally disqualified themselves as
insufficiently familiar with the subject to provide a satisfactory
response).
2. Synopsis of Results
The purpose of the panel survey was to solicit the opinions
of experts on critical issues affecting the successful development of
bacterial pesticides. The questionnaire consisted of two sections; the
first dealt with the concept of bacterial pesticides in general, and
the second addressed future prospects for B.t. in particular. As is often
the case with forward analysis, concrete information underlying answers
to many of the survey questions is lacking, and SRI emphasizes that panel
experts were often able to respond to some questions only with best esti-
mates based on their own knowledge and experience.
All respondents believed that bacterial pesticides have a
useful role in pest control and that efforts to develop these agents on
a commercial scale are warranted. Seventy-eight percent of the respondents
believed that bacterial pesticides will supplant chemical pesticides to
a greater degree by 1985. These responses seem to indicate that there
will be continuing interest and research activity during the next several
years—at least within the research community—in developing new and
improved bacterial pesticides. However, it was apparent that this
optimism was either philosophical or related only to B. t.; few respond-
ents saw any promising new bacterial pesticides on the short-term
horizon, and none of the products were expected to have sizable market
impacts.
Panelists were asked to list bacterial species in addition
to B.t. which held promise for commercialization by 1980 and by 1985
and eleven respondents offered suggestions. Bacillus sphaericus and
Bacillus moritae (for use on mosqultos, gnats, and flies) were mentioned
most often, with the consensus conjecturing commercialization by 1985,
but probably not by 1980. Other bacterial species mentioned were
Bacillus cereus for lice, Clostridium malacosomae for the tent caterpillar,
and Pseudomonas aeruRnosa for grasshoppers. One respondent indicated
a good possibility that other prospective bacterial candidates would
be found, but that species identification had not yet been made, and
three respondents indicated no knowledge of future prospects.
These results are compatible with the literature and do not
suggest any strong likelihood of bacterial pesticides other than B.t.
50
-------
becoming commercially available in significant quantities over the next
decade.
3. Advantages and Disadvantages of Bacterial Pesticides
The panelists were asked to choose and rank the significance
of seven (or less) advantageous and disadvantageous characteristics of
bacterial pesticides relative to synthetic chemical pesticides.
Each response was scored by assigning a value of seven to the
first-ranked characteristic and succeedingly lower values to each of
the other characteristics in order of their listing. The composite scores
are shown in Table 6. In view of an apparent low probability of new
bacterial pesticide development over the next few years, these evaluations
are quite general and can not be meaningfully applied to specific com-
mercial potential.
However, a few points do warrant comment. The relationship
of the factors "no tolerance restriction" and "mammalian toxicity" as
exemplified in the case of B.t., may be commercially more significant
than indicated. There is no tolerance restriction in the use of B.t.
(because of its low mammalian toxicity) and it can be used right up to
harvest time on food crops. This is particularly important for cole
crops, which represent B.t.'s major market area; without this freedom
from tolerance restriction its other characteristics might not be
sufficient to maintain its position in at least this market.
The characteristics listed by the panel which were considered
disadvantageous can be placed in three groups. The first group consists
of "price to user," which received the highest rating, and "lack of field
stability," which acknowledges the sensitivity of B.t. spores and other
bacteria to ultraviolet radiation. The high rank of "price to user" seems
inconsistent with a comparison of the retail price of B.t. and those of
other frequently used chemicals, as shown in Table 7. On cole crops
B-t. was more expensive per acre-treatment than parathion, malathion,
toxaphene, and carbaryl; roughly equivalent in price to dimethoate and
mevinphos; and cheaper than azinphosmethyl and methomyl. Also, a fewer
number of treatments per acre is generally required with B.t. than with
synthetic chemical contact insecticides, primarily because treatment
must be properly timed for effective control. However, the range of
insects controlled is inherent in the concept of price. Since B.t. is
limited to the control of lepidopterous pests, it offers less overall
control per acre-dollar than broad spectrum chemicals that control
aphids, bugs, and leafhoppers as well as lepidopterans.
The second group of disadvantages includes five factors of
major concern to commercial interests. Within this group are lack of
patent protection, high production and R&D costs, unfavorable user
attitudes, and efficacy of control. Since precise information on costs
for R&D and production are generally unavailable outside the industry,
51
-------
Table 6
RANKING OF
ADVANTAGES
AND DISADVANTAGES
OF
BACTERIAL
COMPARED TO
CHEMICAL PESTICIDES
CITED BY QUESTIONNAIRE RESPONDENTS
Advantages
Disadvantages
Composite
Number of
Composite
Number of
Factor
Score*
Responses
Factor
Score
Responses
Low mammalian toxicity
90
17
Price to user
58
10
Host specificity
88
16
Lack of patentability
45
8
Low toxicity to predators
63
14
Field stability
41
10
Safety to wildlife
62
13
Production costs
39
9
Strain selection potential
41
11
User attitudes
39
8
Low host resistance
31
10
Research and develop-
Public image
28
8
ment costs
35
6
Low phytotoxicity
27
9
Efficacy of control
31
6
Compatibility
20
8
Host specificity
29
5
Efficacy of control
16
4
Registration costs
26
5
Federal research support
8
3
Sophistication re-
quired of user
22
7
Potential sales volume
5
1
Low potential sales
Storage stability
4
2
volume
22
6
Deregistration of chemicals
2
1
Competition from new
Price to user
I
1
chemica Is
19
5
Registration costs
1
1
Registration time
15
5
Strain stability 13 4
Unit profitability 11 3
Storage stability 10 3
Public image 6 2
Host resistance 5 2
Federal research
support 4 2
Strain selection
potential 4 1
Compatibility 4 1
ie
Each response was scored by assigning a value of seven to the first-ranked characteristic
and succeedingly lower values to each of the other characteristics in order of their listing.
52
-------
Table 7
PRICES PAID BY USERS FOR SELECTED INSECTICIDES
USED ON COLE CROPS
(California, Early 1975)
Cost/Lb AI*
Rate/Acre
Cost/Acre-Treatmen
Insecticide
(dollars)
(lb AI)
(dollars)
me thorny1
$12.78
0.22 to 0.90
$2.81 to $11.50
azinphosmethyl
7.38
1.00 to 1.50
7.38 to 11.07
mevinphos
9.77
0.25 to 0.50
2.44 to 4.88
B.t.
5.12
0.25 to 1.00
1.28 to 5.12
dimethoate
7.80
0.25 to 0.50
1.95 to 3.90
carbaryl
2.04
0.50 to 2.00
1.02 to 4.08
toxaphene
0.98
2.50
2.45
malathion
2.36
0.50 to 1.00
1.18 to 2.36
parathion
3.98
0.25 to 0.50
1.00 to 1.99
AI = Active ingredient.
Data were calculated on the basis of the AI content of commercial
formulations available for use in cole crops in California during
early 1975.
Source: Communication with Industry
SRI assumed that the panel rated these factors subjectively. In any
event, these ratings are inconsistent with cost estimates provided
by the panel in response to a later question (discussed in the next
subsection). To the extent that "user attitudes" refers to dissat-
isfaction with the relatively slow action of bacterial pesticides and
to user resistance to changing from traditional chemical treatments,
this disadvantage will probably decrease in significance over time if
promotional efforts by producers and distributors are intensified,
and if users become better acquainted with the products, their char-
acteristics of application, and their potential. The lack of patent
protection is a readily understood disadvantage to commercialization.
The ranking of "efficacy of control" as a disadvantage was prompted
by field observations that efficacy in some past instances has been
questionable.
53
-------
The third group of disadvantages includes "host specificity,"
"registration costs," "potential sales volume," "sophistication of user,"
and the "competition from new 'chemicals'." "Host specificity," when
listed as a disadvantage, may be interpreted as "limited host range."
The specificity of B.t., for instance, limits its effectiveness to the
control of certain lepidopterans. Therefore, B.t. has a rather restricted
utility, and additional chemicals may be required to attain the overall
level of pest control desired.
4. Relative Costs for Research and Development: Bacterial
Versus Chemical Pesticides
Panelists were asked to project and compare R&D costs in five
categories for developing a bacterial and a chemical pesticide. The
results of these projections are presented in Table 8. Only six of
the eighteen respondents answered the question. Four offered lower
estimates for bacteria and two offered the same estimates for bacteria
and chemicals. The considerable variation between responses reflected
a general uncertainty among panel members outside the industry on the
investment requirements necessary to bring a product to market. In
spite of the fact that R&D costs for bacterial pesticide development
had been placed high on the list of disadvantages by the full panel,
Table 8
COSTS OF BACTERIAL AND CHEMICAL PESTICIDE DEVELOPMENT
ESTIMATED BY QUESTIONNAIRE RESPONDENTS
(Thousands of Dollars)
Category Bacteria (Range) Chemical (Range)
Screening and
synthesis
$
167
($1 to $500)
$ 750
($50
to
$2,000)
Field testing
933
($200
i to $2,800)
1,050
($100
to
$2,800)
Toxicology
608
($50
to $1,500)
1,100
($400
to
> $2,500)
Process development
697
($30
to $2,000)
1,320
($20
to
$4,000)
Registration
300
($2*
to $1,000)
300
($2*
to
$1,000)
Total
$2,
,705
$4,520
*A response from the scientific community outside of the United States
54
-------
the total of the average costs estimated by the six respondents to this
question was 40 percent lower for bacterial pesticides than for chemicals.
In addition, of the four respondents who had offered lower cost estimates
for bacteria, three listed high R&D costs as a disadvantage of bacterial
insecticides. Due to these discrepancies and the wide range of responses
within each category, these estimates can be taken to indicate little
more than the general lack of good information available on this subject
from nonindustry sources.
5. Origin of Research Funds
With regard to the origin of research funds, the panel con-
sensus was that roughly equal percentages of the total funds would be
supplied by pesticide producers (34.1 percent), federal funding agencies
(31.4 percent) and in-house research budgets of federal laboratories
(28.7 percent). Only a small percentage (4.6 percent) expected any
contribution from grower groups. Seventy percent of the respondents
believed that the indicated division of funding was proper and justi-
fied. The remainder were about equally divided in opinion as to whether
the federal government or pesticide producers should support a greater
proportion of the research costs. These responses suggest a desire
for the federal government to bear major responsibility for developing
new or improved bacterial pesticides. However, it should be remembered
that the majority of respondents were research personnel, and that
their collective response may reflect a perception of the most-likely
sources of their research funds.
6. Bacillus Thuringlensis—Future Production Levels
The second section of the questionnaire pertained specifically
to B.t. The panel was first asked to estimate the annual level of B.t.
production over the next ten years. Six respondents estimated that the
level would remain fairly static (close to 1-1.5 million pounds annually),
and eleven thought that production levels would increase. Predictions
of increased production levels ranged from a low of two million pounds
per year to a high of fifty million pounds per year. The average value
of the increases was 4.1 million pounds per year (after omitting the
highest and lowest of"the estimates). For 1985, the average of all
responses, including those predicting no increase in production, was
2.6 million pounds annually.
7. Bacillus Thuringiensis—Crops Representing Potential
Expanding Markets
Panelists were asked to indicate the crops or groups of crops
that they expected would receive increased treatment with B.t., and to
estimate the B.t. market share that might be obtained in each 1980 and
1985. (Although only eleven respondents estimated increases in total
55
-------
B.t. production over the next ten years, eighteen indicated that use
would increase.) The greatest increase was predicted for forests and
crops now registered for B.t., receiving 16 and 13 positive responses,
respectively, as shown in Table 9. "Other" vegetable crops, and home
gardens and ornamentals were also mentioned by over half of the respond-
ents. All of the other crop groups listed in the questionnaire elicited
a prediction of increased use from at least one panelist.
Twelve of the panel felt qualified to specifically predict
B.t. market shares in these crops for 1980 and 1985. An analysis of
the responses showed that B.t. use is expected on 32 percent of all the
forest and B.t.-registered crop acres to be treated with insecticides
by 1985. Home gardens and ornamentals (15 percent), other vegetables
(12 percent), tobacco (12 percent), and cotton (10 percent) were also
expected to attain significant market status by 1985. Eighty percent
of the respondents believed that new, improved, more highly virulent
or more selectively active strains of B.t. would be found and developed
within the next ten years and would play a role in increasing the
future use of B.t.
8. Bacillus Thuringiensis—Major Obstacles to Greater Use
All panel members responded to the questions about major
obstacles to greater future use of B.t. A wide variety of responses
was obtained; these were categorized as much as possible, and the
frequency of responses per category was recorded. Results are shown
in Table 10. Efficacy of control and high costs were cited most often,
closely followed by lack of field stability and narrow host range,
development of new chemicals, unfavorable user attitudes, required
sophistication of the user, and the need for better methods of treatment
to expose feeding larvae to B.t. Ten miscellaneous obstacles that
could not be categorized with any of the above factors were also listed.
Some respondents felt that increases in federal research support would
be needed to overcome the obstacles, while others believed greater
contributions by, and better communication with, producers would be
the key to future improvements.
However, respondents were also of the opinion that most
obstacles would be overcome by research and improved formulations in
the near future, which would lead to increased market size and lower
prices. Thus, the type of market growth previously projected was appar-
ently perceived to be contingent on attaining these improvements.
9. Bacillus Thuringiensis—Host Resistance
Panelists were asked to estimate the probability that any
currently susceptible major target insect would develop resistance
to B.t. in the next 20 years. The consensus was that the probability
was much lower than for chemical insecticides. The 12 numerical
56
-------
Table 9
PERCENTAGE OF TOTAL INSECTICIDE-TREATED ACREAGE
TREATED WITH BACILLUS THURINGIENSIS, FOR SELECTED CROPS
ESTIMATED BY QUESTIONNAIRE RESPONDENTS
(1980, 1985)
1980 1985
Crop (percent) (percent)
Currently registered
crops 25.0% 32.1%
Other vegetables 8.4 12.6
Forests 16.9 32.5
Alfalfa 4.6 6.4
Cotton 3.4 10.5
Tobacco 7.9 12.1
Corn 3.3 5.4
Orchards and grapes 6.6 8.0
Small grains 0.8 1.7
Other field crops 1.5 1.5
Home gardens and
ornamentals 7.5 15.0
Note: Data represent an average of the responses for each crop.
It should be noted that acres treated with B.t. may also
be treated with other insecticides.
Table 10
MAJOR OBSTACLES TO INCREASED BACILLUS THURINGIENSIS USE
Number of
Obstacles Responses
Efficacy of control 9
High costs (R&D, production, price) 8
Field stability 6
Narrow host range 6
Development of new chemicals 5
User attitudes 5
Sophistication of user 4
Application methods 3
Others 1 each
(11)
57
-------
estimates offered ranged from 0 percent to 20 percent and averaged
0.9 percent. Three respondents answered that the probability was
"unlikely," "not very good," and "very low;" three offered no opinion.
10. Discussion
Based on its analysis of the panel survey, SRI believes that
there are three potential advantages of bacterial insecticides over
chemical insecticides which constitute incentives for development and
three major disadvantages which serve as disincentives.
a. Advantages of Bacterial over Chemical Pesticides
Safety—The major and overriding advantage for B.t. is
the high degree of safety associated with its use. (This is true for
strains for which use is approved; some other strains, e.g., B-exotoxin,
are toxic to mammals.) There are several facets to this advantage.
The lack of mammalian toxicity precludes hazards to applicators and
field workers, the chance of accident to the unwary who may inadvertently
come into contact with stored materials, and has allowed the waiver of
tolerance and reentry restrictions (important in protecting cole crops
from cosmetic damage up to harvest). Other advantages include safety
to wildlife, natural predators, and beneficial insects, the lack of
phytotoxicity, and the lack of persistence in the environment.
Strain Selection Potential—The second perceived advantage
is strain selection potential, which could allow periodic improvements in
commercial products by the discovery of new, more virulant or more
selectively active strains of bacteria, hopefully with little or no
mammalian toxicity. For example, the discovery of the HD-1 strain
greatly stimulated the commercial development and use of B.t. over
the last five years. (An active research program is currently under
way to differentiate additional strains. New candidates, such as strain
HD-187, have already been found that could further increase the utility
and commercial promise of B.t.)
Host Resistance—The perceived low probability of host
insect populations developing resistance to B.t. is an illustration of
the third major advantage of bacterial pesticides. The primary benefit
of this attribute is that it extends the commercial lifetime of a
product, allowing producers a longer period in which to recover their
investment and accrue profits.
58
-------
b. Disadvantages of Bacterial Opposed to Chemical
Pesticides
This list of major disadvantages does not include any
reference to cost or price since these characteristics may be variable
with time, and more the result of the developing commercial situation
than its cause. Unit costs and price are largely inverse functions of
the utility of a product and its sales volume, and hence could decrease
if utility and sales increase.
Narrow Host Range—The fact that B.t. controls only lepi-
dopterans will always be a limitation. Although lepidopterans are of
major importance as agricultural inspect pests, broad-spectrum control
methods will in most cases be chosen for use over B.t. if other consid-
erations are equal. For this reason B.t. or similar agents may be
largely relegated to a role in integrated pest control programs or in
combination with chemical insecticides. An exception to this restriction,
however, may develop in some forest insect control programs where
lepidopterans are the only target pests of economic concern.
Lack of Patent Protection—The lack of patent protection
is a readily understood deterrent to'commercialization. Arguments
against the significance of this factor have been offered by some, citing
as an example the current and past history of B.t. as a commercikl
entity. The market size of B.t., however, does not compare with that
of protected chemical insecticides. If protection were provided, a
protected producer might feel justified in investing its funds more
heavily in research and promotion to increase market potential.
Required User Sophistication—The sophistication required
of the user may also be considered a disadvantage. This factor relates
largely to the requirement for accurate timing of application with early
instar development for efficacious control. Users with little knowledge
of host development or no experience in identifying larval stages by
inspection (and who subsequently mistime their applications of B.t.) can
often be discouraged by the results. Whether or not the average user
can realistically be expected to become astute in following the develop-
ment of insect populations in his fields is an academic question. The
fact remains that in many cases he can use chemicals without having to
keep track of pest development and may choose them for this reason
alone.
59
-------
D. Decision Analysis of a Possible Bacterial Pesticide
Venture
1. Introduction
SRI conducted a preliminary analysis of the decision to enter
a bacterial pesticide business, to ascertain the commercial feasibility
of bacterial pesticide products, and to establish the relative importance
of the many factors influencing the proposed business.
The analysis focused on identification of the bases upon which
a well-managed, private company would elect to be the first firm to
develop and market a new bacterial agent known to be toxic to certain
pests in nature. If no firm is willing to risk failure to become the
first producer, then the product will never reach the market.
Bacillus thuringiensis and the milky spore bacterium, Bacillus
popillae (Dutky), are already commercialized. However, as discussed
earlier, the panel survey revealed few other likely bacterial candidates
and little information was found on the relative prospects of any such
agents. Therefore, rather than resort to pure speculation about such
products, B.t. was chosen for this analysis since commercial feasibility
has already been established.
No proprietary information on B.t. was used by SRI. Informa-
tion was obtained from public sources, interviews with chosen authorities,
and the panel survey; where necessary, the SRI team estimated missing
values. It is recognized that companies now producing and marketing
B. t. have more precise Information and may wish to take issue with certain
conclusions generated from this analysis.
The analysis considers a hypothetical development project in
isolation from a company's product line. No consideration was given
to the possibility that the company may already possess some of the
required capital equipment, such as fermentation facilities used for
other products.
A company considering a new commercial venture views it as
one possible alternative in competition with others for available invest-
ment capital. Therefore, in this analysis the value of this investment
was measured against an arbitrary alternative opportunity. The basic
alternative investment chosen was a constant interest bank, with a rate
of return equivalent to what a typical company would expect from a
moderately-profitable business venture. This analysis assumed (1) a zero
rate of inflation for all costs and prices, (2) constant 1974 dollars,
and (3) an eight percent discount rate before taxes as a minimum
acceptable rate, as described in IV-B-1, "Problem Structuring." Thus,
if the net present value at eight percent of all the positive and
negative cash flows from the venture exceeds zero, then the venture was
considered economically superior to an alternative investment yielding
an eight percent return.
60
-------
The decision analysis was conducted on B.t. according to the
methodology discussed in IV, "Method of Approach." The first step was
to construct a base case scenario. Hie data for this case consisted
of information on market size, production costs, capital costs, and
R&D costs, based on best data available to SRI. Where hard data were
proprietary or not available, the project team used its judgment based
on the available open literature and personal interviews. The following
reviews B.t.'s history and characteristics as a preliminary step prior
to base case development.
2. Background on the Product
Review
Bacillus thuringiensis (Berliner) is a spore-forming, toxin-
producing bacterium which is pathogenic to certain insects. Various
preparations of this bacterium have been tested and commercially used
as insecticides in Russia, France, the Netherlands, Canada, and the
United States—under a variety of trade names—for several years.
Current use in this country is restricted to the control of lepidop-
terous larvae. The mechanism of control affected by B.t. is largely
mediated through the ingestion and subsequent dissolution in an insect's
midgut of a crystalline endotoxin that is produced in the vegetative
cell of the bacterium in conjunction with spore formation (certain vari-
eties also produce a thermostable 8-exotoxin [fly toxin]). Since the
endotoxin is readily soluble in alkali and insoluble in acid, it is
suspected by many that it is toxic only to lepidopterans with alkaline
midgut contents.
B.t. has been registered for use in the control of 25 lepidop-
terous pests on 24 agricultural crops, plus various flowers, shrubs,
ornamental species, and forest trees in the United States. Registered
uses include that for control of the cabbage looper (Trichoplusla ni),
one of the most destructive pests of vegetable crops. The current
U.S. market is estimated to approach 1.5 million pounds annually.
History of Development
The history'of B.t. development as a microbial insecticide,
prior to the discovery of the improved HD-1 strain, is extensive (see
Norris, 1967 and Heimpel, 1967). The bacterium was first isolated in
Japan in 1902 by Ishiwata from silkworms suffering from an epidemic of
"flacherie". Beyond its devastating effect on commercial silkworm
populations, little thought was apparently given at that time to its
potential as a microbial insecticide. In 1915 the bacterium was re-
examined and named Bacillus sotto. Two strains were differentiated
based upon their pathogenicity to silkworm. At about the same time,
Berliner was isolating the organism from larvae of the flour moth
(Anagusta kuhniella). He independently named the bacterium Bacillus
thuringiensis, the currently accepted binomial, and hence became the
authority of record. With its re-isolation in 1927, ELt, first began
61
-------
to receive attention as a potential insecticide. During the 1930s,
demonstration of its wide host range prompted extensive testing in
Europe. Promising results against the European corn borer (Ostrinia
nubilalis) were obtained. Immediately prior to the outbreak of World
War II the first commercial B.t.-based insecticide, "Sporeine", was
produced in France.
American interest in B.t. grew after World War II, and impres-
sive levels of control against the alfalfa caterpillar (Colias eurytheme)
were obtained in field trials. In 1953 additional interest was stimu-
lated by Hannay's demonstration in Canada of the role of the specific
crystalline endotoxin; although the presence of the proteinaceous
crystals had been observed early in the history of B. t., their toxi-
cological character had not been suspected. Previously, the insecticidal
activity of B.t. preparations had been assumed to result from massive
pervasion of host tissues by a burgeoning bacterial population. The
uncovering of this specific contained toxic agent provided a new dimension
to insect pathology and expanded the commercial development potential
of B.t.
The first commercial B.t. insecticide in the United States
became available in 1958. Added impetus for development was provided
in 1966 when the federal government banned a series of chemical insec-
ticides from use in the control of worms on tobacco, because of toxic
residue problems. B.t. was recommended as a substitute and it sub-
sequently provided reasonable control of tobacco hornworm (Manduca sexta)
and tobacco budworm (Heliothis virescens).
However, in spite of several demonstrations of effectiveness
in insect control, use of B.t. prior to 1969 proved in many instances
to be unreliable because of variability in potency that was generally
characteristic of the commercial preparations then in use. Because
of this experience, B.t. came to be viewed with suspicion by some
authorities.
In 1969 a new strain was discovered by Dr. Howard Dulmage
(USDA, at Brownsville, Texas) which displayed a potency 15 to 30 times
higher than previous formulations. The new strain, designated HD-1,
provided a higher and more reliable degree of control, and was rapidly
incorporated into commercial formulations. Adequate methods of product
standardization were subsequently developed by Dr, Dulmage et al and
were adopted by industry. Today, all commercial preparations of B.t.
are based upon the HD-1 strain, and research is underway to develop
still more potent strains.
The three primary U.S. producers of B.t. are Abbott Laboratories,
Nutrilite Products Inc., and Sandoz Inc. (owned by Sandoz AG, Switzerland).
Abbott Laboratories produces a wettable-powder formulation labeled Dipel®
(which can also be formulated as a baited insecticide by mixing with
corn meal). Sandoz, the other major domestic producer, manufactures
both a "flowable-liquid" formulation and a wettable-powder formulation
62
-------
labeled Thuricide®*. (Sandoz's package of B.t. technology was purchased
in 1973 from the International Minerals & Chemical Corp.) Nutrilite
Products Inc. produces smaller quantities of B.t. in a wet table formula-
tion under a Biotrol® trademark; these products are exclusively distri-
buted by the Thompson-Hayward Chemical Co., which also markets a corn
meal-corn oil baited formulation.
Description and Taxonomy
Bacillus thuringiensis (Berliner) is a rod-shaped, motile
spore-forming bacterium of the B. cereus group. Several strains have
been differentiated within the species. All are mesophilic aerobic
(or facultative anaerobic) organisms producing good growth between 28°
and 35°C. Spores are produced in aging vegetative cells termed sporangia
after 24 to 48 hours of culturing. Spores are ellipsoidal to cylindrical,
paracentral to subterminal in their placement within the sporangium,
and have thin, three-layered walls. The species is gram-positive, with
simple nutritional requirements affording uncomplicated large batch
production.
The primary distinction between B.t. and other species within
the B. cereus group is that it produces a crystalline parasporal body
within the sporangium that is toxic to a large number of lepidopterous
larvae. Hence, B.t. is an insect pathogen and B. cereus is not. The
B.t. parasporal body contains one or more toxic principals which are
collectively termed the "B.t. 6-endotoxin". Curiously, the endotoxin
is the same material as that composing the middle layer of the spore
wall. Since the formation of the endotoxin crystal occurs at the same
time within the sporangium as spore formation, Dulmage and others
considered B.t. as a deficient form of B. cereus.
Bacillus thuringiensis variety thuringiensis (Berliner) has
been designated as the type strain of the B.t. complex (Heimpel and
Angus, 1958). The differentiation of strains or varieties within the
species has been accomplished both by the use of biochemical character-
istics and serological reactions to flagellar antigens (H antigen).
Biochemical strains have been differentiated on the basis of esterase
production patterns (Norris, 1964) and toxin production (Heimpel, 1967).
Toxins involved in the classification scheme include the 6-endotoxin,
and two exotoxins designated as a-exotoxin (lecithinase C, phospholipase
C) and 3-exotoxin (thermostable fly toxin). A consideration of toxin
Sandoz does not produce a baited insecticide, although its wettable-
powder formulation is sometimes used for this purpose by licensed
formulators.
63
-------
production patterns is useful since it allows the distinction between
strains having the same serological reaction.
Differentiation on the basis of serological reaction has been
developed by de Barjac and Bonnefoi (1967, 1968, 1972) at the Institut
Pasteur in Paris. Strain classification schemes based on biochemistry
and serology, respectively, are remarkably parallel and complementary.
To date a total of 15 serological strains of B.t. have been found, and
it is reasonable to expect the discovery of additional strains as
research continues. A list of the currently recognized strains is pre-
sented in Table 11.
All current commercial formulations are based on the HD-1
strain isolated by Dulmage (1970).* This new strain was originally
considered as a sub-form within B.t. variety alesti; it since has been
elevated on the basis of serological reaction to variety status
(kurstaki) by de Barjac and Lemille (1970). This strain produces a
potent level of 6-endotoxin, but no exotoxins.
Production and Standardization
B.t. insecticides have been in commercial domestic production
in one form or another since about 1960. The bacterium itself is pro-
duced by relatively-straightforward aerobic fermentation. Two major
processes are currently in use. The method most extensively used is
based on conventional deep tank fermentation, where the culture is
carried through the sporulation stage in the fermenter. The spores
and the endotoxin components of the culture are then either separated
from the "beer" for preparation into wettable powder formulations, or
are concentrated and supplemented within the "beer" itself to provide
a liquid or flowable formulation.
The other process is "semi-solid" fermentation. The bacteria
are first allowed to increase vegetatively by fermentation in the
absence of sporulation. The "beer" is then transferred to a bran where,
after a suitable incubation period, sporulation and endotoxin formation
occurs. The bran-bacterial mixture is then dried, milled, and packaged
as a wettable formulation. Some protein denaturation and subsequent
loss of activity occurs in this semi-solid process during milling as a
result of the heat generated by grinding.
Various baited formulations can be prepared from the wettable
powders by mixing with cornmeal or combinations of cornmeal and corn
oil. Baited formulations have been used with success against the tobacco
budworm on tobacco.
Rhodia's Bactospeine® B.t. insecticide, registered in 1975, is derived
from a different strain.
64
-------
Table 11
SEROLOGICAL TYPES AND BIOCHEMICAL VARIETIES CURRENTLY DISTINGUISHED
WITHIN BACILLUS THURINGIENSIS (BERLINER)*
Serotype Variety (strain)
I thuringiensis (Berliner)
II f inltimus
III a alesti
III a,b kurstaki (HD-1)
IV a,b sotto, and dendrolimus
IV a,c kenyae
V a,b galleriae
V a,c canadensis
VI subtoxicus, and entomocidus
VII aizawai
VIII morrisoni
IX tolworthi
X darmstadiensis
XI toumanoffi
XII thoropsoni
*
de Barjac and Bonnefoi (1973).
65
-------
All formulations have a commercially acceptable shelf life if
stored in the dark at room temperature. Since the spores are sensitive
to ultraviolet radiation (viability disappears within 36 hours when
exposed to direct sunlight on glass slides), storage life can be a
problem if precautions are not taken. Wettable powder formulations
packaged in suitable commercial containers have been exposed to light at
temperatures well above room temperature without loss of potency.
Problems of convenience have been encountered with some formu-
lations. Bran formulations have proved difficult to wet, and the swell-
ing of the wetted bran can result in clogged spray equipment. Liquid
or flowable formulations have experienced settling problems in containers
during storage, and care must be taken to thoroughly resuspend these
preparations prior to tank mixing to avoid loss of material.
Potency standardization of commercial formulations is accom-
plished by spore counts and bioassay. However, values derived from
these methods do not correlate well between batches; hence, both spore
count (viable spores per gram) and bioassay values (international units
of potency) must be declared on labels*.
A
Bioassay and quality assurance costs are major items in the cost of
production. Spore counts are determined by dilution plate techniques
wherein spores are germinated on agar plates and the number of colonies
formed are counted. Spore counts so derived are not indicators of
potency, but they are retained to insure that the formulation was pro-
duced by B.t. Consequently, only a minimum spore count is required
viz a specified IU/mg figure. Bioassay is the more reliable method,
wherein potency (activity, toxicity) of the test lot is compared to a
standard lot by feeding both in various concentrations to uniform-size
test larvae. LD^values are calculated from mortality curves after a
suitable test period (e.g., five days). Cabbage loopers are tradition-
ally used as test larvae. The bioassay standard, which is stored by the
USDA at Brownsville, Texas, and supplied upon request to commercial
producers, has been arbitrarily assigned a potency value of 18,000 Inter-
national Units of Potency (IU) per mg. The test lot's potency is
determined as follows:
LDsf) (Standard lot)
Test Lot IU = — 71 r~m x 18 »000 1U
LDj_q (test lot)
The percent active ingredient to be declared on the label is then derived
by dividing the potency of the test lot by 500,000 IU, which has been
arbitrarily designed as 100% active. A batch labeled as containing
1% active ingredient, therefore, has a potency of 5,000 IU per mg,
regardless of the actual composition of the batch in terms of the per-
cent by weight bacterial matter. The disadvantage of the bioassay method,
in spite of its reliability, is the high cost of maintaining and control-
ling cultures of test larvae (cabbage loopers) in the laboratory and con-
ducting the bioassay.
66
-------
Markets
It is estimated that approximately 1.5 million pounds of B.t.
formulations are currently produced in the United States, most of which
is used domestically. The primary domestic market area has been com-
mercial vegetable crops, primarily the cole family (Table 12). B.t.
is exempt from tolerance on all food crops for which registration has
been obtained. No preharvest or reentry periods are required since
mammalian toxicity has not been,demonstrated.
A major factor limiting B.t.'s volume of sales in cole and
other vegetable markets is the relatively small amount of acreage devoted
to these crops. The U.S. acreage of vegetable crops on which B.t. has
been registered totals about 2.5 million acres, only a portion of which
is located in geographical areas with target pest problems or where B.t.
has been approved for use. Even if future registered uses are extended
to include other vegetable crops and additional areas, the upper limit
for use on these crops is probably no more than about 3.6 million
acres (e.g., the current total acreage of vegetables and melons). Hence,
if vegetable and melon crops remain the primary markets, this relatively
narrow-spectrum insecticide cannot be expected to assume a position of
major importance in relation to total insecticide usage.
However, commercial producers are exploring other potential
markets. One area receiving attention is forest insect control, which
could conceivably extend the B.t. market by several million acres in the
United States and Canada, if sufficient efficacy can be demonstrated.
"Impressive and economic" control of gypsy moth, tussock moth, and spruce
budworm has been claimed in limited applications, and prospects for
significant expansion into this market may be encouraging. In view of
the extensive forest acreage, the prime application for the future may
well be in the area of forest insect control*.
Another potential market for B.t. is cotton, which uses more
chemical insecticides than any other domestic crop. Both the tobacco
budworm and the bollworm are susceptible to B.t.; however, because of
their feeding habits, control under field conditions has not been
effective. Further development in baited formulations, as demonstrated
against the tobacco budworm in tobacco, may improve its position in
cotton against these two serious pests. If prospects for B.t. use on
There is some evidence to indicate that the commercial forest industry
has "lobbied" against the testing of B.t. in forest insect control.
It has been suggested that strong pressures have been exerted upon
government agencies to restrict or delay testing in some areas, which
could have delayed the submission of sufficient test data to satisfy
registration requirements. While B.t. might not be able to compete with
some chemical insecticides on purely economic grounds, a political factor
may have to be considered and contended with in the development of a
marketing plan for B.t. in the area of forest insect control.
67
-------
Table 12
CROPS REGISTERED FOR B_^t. USE, PESTS CONTROLLED,
RECOMMENDED DOSAGES
Crops
Insect Pests
Alfalfa
Almonds
Artichoke
Beans
Alfalfa caterpillar
(Colias eurytheme)
Dosage (Billion IU/Acre)* Restrictions
S
WP
D
1.0
0.4-0.9
2.7-5.4
West Coast
states and
Arizona only
(Temporary exemption from tolerance in or on almonds granted
[expired July 23, 1975]. Fed. Reg. Vol. 139, No. 146,
July 29, 1974)
Artichoke plume moth S
(Platytilia WP
carduidactyla) D
Cabbage looper S
(Trichoplusia ni) WP
4.0-8.0
4.0-8.2
1.8-6.8
4.0-8.0
3.6-7.2
Broccoli
Cabbage looper
(T. ni)
Imported cabbageworm
(Pieris rapae)
S
WP
D, Pr
Diamondback moth
(Plutella xylostella) WP, Pr
S
WP
D, Pr
Brussel Cabbage looper S
sprouts (T. ni) WP, Pr
Diamondback moth Pr
(P. xylostella)
Imported cabbageworm S
fapae) WP, Pr
Cabbage Cabbage looper S
(T. ni) WP
D, Pr
4.0-8.2
3.4-8.2
1.8-8.5
1.8-6.8
2.0-6.8
1.8-6.8
1.44-6.8
4.0-8.2
3.6-7.2
1.8—6•8
2.0-6.8
1.8-5.4
4.0-8.2
3.4-8.2
1.8-8.5
Diamondback moth
(£. xylostella)
WP, Pr
3.4-8.2
68
-------
Table 12 (Continued)
Crop
Insect Pests
Dosage (Billion IU/Acre)* Restrictions
Cauliflower
Imported cabbageworm
CP. rapae)
Cabbage looper
(T. ni)
S
WP
D, Pr
S
WP
D
2.0-6.8
1.8-6.8
1.44-6.8
4.0-8.2
3.4-8.2
1.8-8.5
Celery
Collards
Diamondback moth
(Z* xylostella)
Imported cabbageworm
(P. rapae)
Cabbage looper
(T. ni)
Cabbage looper
(T. ni)
WP
S
WP
D
S
WP
D'
S
WP
D
1.8-6.8
2.0-6.8
1.8-6.8
1.44-6.8
4.0-8.2
3.6-8.2
1.8-7.2
4.0-8.2
3.4-8.2
1.8-8.5
Diamondback moth
(P. xylostella)
Imported cabbageworm
(£• rapae)
WP
S
WP
D
1.8-5.4
2.0-6.8
1.8-5.4
1.44-6.8
Cotton
Cucumbers
Grapes
Cabbage looper
(T. ni)
Cabbage looper
(T. ni)
Grape leaffolder
(Desmia funeralia)
S
WP
D
S
WP
D
S
WP
D
4.0-8.0
3.6-7.2
4.0-6.8
2.0-8.2
1.8-8.2
1.44-6.0
4.0-8.2
3.6-8.2
1.44-7.2
Arizona and
California
only
Southwestern
United States
only
Omnivorous leafroller
(Platynota stultana)
S
WP
D
2.0-8.2
1.8-5.4
2.4-7.2
California
only
69
-------
Crop
Kale
Lettuce
Melons
Potatoes
Spinach
Table 12 (Continued)
Insect Pests Dosage (Billion IU/Acre)* Restrictions
Cabbage looper
(T. ni)
Diamondback moth
(]?. xylostella)
Imported cabbageworm
(£. rapae)
Cabbage looper
(T. ni)
Imported cabbageworm
(P. rapae)
Cabbage looper
(T. ni)
Mustard Cabbage looper
greens (T. ni)
Diamondback moth
(P. xylostella)
Imported cabbageworm
(P_. rapae)
Oranges California
orangedog
(Paplllo crestphontes)
Fruit tree leafroller
S
WP
D
WP
S
WP
D
S
WP
D, Pr
I
WP
D
S
WP
D
S
WP
D, Pr
WP, Pr
S
WP
D, Pr
S
WP
Cabbage looper
(T. ni)
Cabbage looper
(T. ni)
S
WP
S
WP
D
S
WP
D
A.0-8.2
3.6-8.2
1.8-6.8
1.8-5.4
2.0-6.8
1.8-5.4
1.44-6.8
4.0-8.2
3.4-8.2
1.8-8.5
6.8
4.0-6.8
2.0-8.2
1.8-8.2
1.44-6.8
4.0-8.2
3.6-8.2
1.8-5.4
1.8-5.4
2.0-6.8
1.8-5.4
1.44-6.8
2.0-8.2
1.8-3.6
4.0-8.2
3.6-7.2
2.0-8.2
3.6-8.2
1.8-7.2
4.0-8.2
3.6-8.2
2.9-6.0
Pr on bibb and
leaf lettuce only
Arizona and
California only
Ground application
only
Same as above
North Carolina
Virginia only
70
-------
Crop
Tobacco
Tomatoes
Turnip
greens
Walnuts
Table 12 (Continued)
Insect Pests Dosage (Billion IU/Acre)* Restrictions
Cabbage looper
(1- ni>
Tobacco hornworm
(Manduca sexta)
S
WP
S
WP
D
Tobacco budworm S
(Heliothis virescens) WP
Bait
Cabbage looper
(T. ni)
Tomato hornworm
(Manduca quinquem-
aculata)
Tomato fruitworm
(Heliothis zea)
Cabbage looper
(T. ni)
Diamondback *mo th
(P.- xylostella)
Imported cabbageworm
0?. rapae)
Redhumped cater-
pillar
(Schizura concinna)
Watermelons Rindworm complex
Flowers Cabbage looper
(Ornamentals)Imported cabbageworm
Leafrollers
Shrubs Bagworm
and trees (Thrydopteryx
(Ornamentals) ephemeraeformis)
S
WP
D
S
WP
D
WP
S
WP
D
WP
S
WP
D
S
WP
S
WP
WP
WP
WP
WP
4.0-6.0
3.6-5.A
1.6-6.0
1.8-5.4
1.44-5.4
1.6-6.0 Bait applied
1.8-5.4 by hand
1.5-4.8
4.0-8.2
3.6-7.4
1.8-6.8
2.0-8.2
3.6-5.4
1.44-6.8
3.6-5.4
4.0-8.2
3.4-8.2
1.8-8.5
1.8-5.4
2.0-6.8
1.8-5.4
1.44-6.8
1.0-4.0
1.8-7.2
2.0-6.0
1.8-5.4
7.35
7.35
7.35
3.6-7.2
California only
Ground application
only
Florida only
71
-------
Table 12 (Continued)
Insect Pests
Dosage (Billion IU/Acre)* Restrictions
California oakworra S
(Phryganidia cali- WP
fornica)
Elm spanworm WP
Fall cankerworm S
(Alsophila pometaria) WP
Fall webworm
S
WP
4.0-8.3
3.6-7.4
7.2
2.0-8.3
1.8-3.6
2.0-8.3
1.8-3.6
California only
Forests
Gypsy moth S
(Porthetria dispar) WP
Leafrollers S
Redhumped cater- S
pillar WP
Spring cankerworm S
(Paleacrita vernata) WP
Tent caterpillars S
(Melacosoma disstria WP
et al.)
California oakworm S
(P. californica) WP
Fall cankerworm (A. S
pometerla) and WP
Fall webworm
4.0-8.3
3.6-7.2
2.08-8.3
2.0-8.3
1.8-7.2
2.0-8.3
1.8-3.6
2,0-8.3
1.8-3.6
4.0-8.0
3.6-7.2
2.0-4.0
1.8-3.6
Northeastern
United States
only
California only
California only
Western United
States only for
Great Basin tent
caterpillar
California only
Ground application
only
Ground application
only
Gypsy moth
(!P. dispar)
Gypsy moth
(P. dispar)
S
WP
S
WP
4.0-8.0
3.6-7.2
8.0
7.2
Northeastern
United States
only
Ground application
only
Northeastern
United States only
Ground application
only
72
-------
Table 12 (Concluded)
Crop Iri3ect Pests Dosage (Billion IU/Acre)* Restrictions
Redhumped caterpillar
(S. conclnna)
Spring cankerworm
(P. vernata)
Tent caterpillars
(21 • disstria, et al.)
S 2.0-8.0
WP 1.8-7.2
S 2.0-4.0
WP 1.8-3.6
S 2.0-4.0
WP 1.8-3.6
California only
Ground application
only
Ground application
only
Western United
States only for
Great Basin tent
caterpillar. Ground
application only
*
Formulations: S ¦ Suspension, WP - Wettable Powder, D ¦ Dust, Pr » Pressurized
gas or aerosol can.
Source: EPA Compendium of Registered Pesticides, January 1974.
73
-------
extensive cotton acreage are to be seriously considered, however, devel-
opments in machine application methods will be required. Additionally,
the development of more potent formulations may be necessary.*
With tobacco, because of its controlled acreage and high per-
acre value, baited formulations are generally applied by hand. This
suggests that sweet corn might also be a potential market since insec-
ticides have been applied by hand for earworm control in this crop.
Damage to field corn by this insect has not been of sufficient economic
significance to warrant the general application (registration) of
insecticides. .
An additional market being developed is home gardens. Home
gardeners have proven to be sensitive to environmental questions and may
be more inclined toward the use of environmentally "safe" (but efficacious)
materials than profit-oriented commercial growers. All three U.S. producers
have developed and are promoting formulations specifically devised for
home-garden use. Nutrilite Products has presumably enhanced the sale-
ability of its home garden formulation (Biotrol-Plus) by including
pyrethrin, another "safe" biological insecticide.
A primary reason for B.t.'s relatively rapid growth over the
past few years was discovery of the more-potent HD-1 strain, which
provided higher and more reliable levels of control than the less-potent
strains that had previously been accepted for registration. Existing
producers (and users) were quick to adopt the new strain; this may have
been among the incentives that brought Abbott Laboratories into the
market as a new producer. The possibility of finding still more potent
strains in the future should further stimulate the market. In addition
to company-sponsored work in this area, a cooperative research effort
has been initiated by the U.S.D.A. and includes academic scientists
previously involved in B.t. research. This effort consists of simul-
taneously screening, via bioassay, some 280 different B.t. cultures
against seventeen lepidopterous insects, six species of mosquitos, a
species of biting louse, and a species of fly (for the purpose of
differentiating new strains).
*
Much of B.t.'s potential growth in the cotton market appears to have
been associated with -plans for use in combined applications with
chlordimeform, a chemical insecticide with ovicidal activity. Because
potential toxicology problems were uncovered in testing, chlordimeform
was withdrawn from the market for further testing in mid-1976—after
the body of this report was written. The resultant temporary or
permanent unavailability of this chemical is expected to have a serious
dampening feffect on B.t.'s immediate growth prospects in cotton.
74
-------
Activity
B.t. must be ingested to be effective. It does not act as
a contact poison, and its effects are slower and less visible to
the user when compared to conventional chemical controls. Thorough
coverage is required for effective control, and coverage of the under-
sides of leaves is critical for certain pests where eggs and early
instars develop on these surfaces. Pests which feed primarily on plant
terminals, e.g., the tobacco budworm (Heliothis virescens), are difficult
to control by conventional application, since new plant tissues that
develop after treatment are not protected. In these instances success
has been demonstrated with the use of baited formulations that prefer-
entially lure insects from normal feeding sites. The best example of
this practice is the control of tobacco budworm in tobacco with hand
applied baits. An alternative is repeated treatments at close time
intervals to protect newly-exposed terminals.
The timing of applications relative to instar development is
also critical for effective control. Since early instars ingest more
material per unit of body weight than do later instars, the earlier
stages are easier to control. Hence, treatments applied soon after egg
hatch are more effective than loiter treatments, provided that thorough
coverage is obtained. With proper timing and coverage at the recommended
dosage, B.t. can be as effective on registered pests as chemical insec-
ticides and is especially effective against the cabbage looper, a primary
insect pest in vegetables. Registered uses do not include the armyworm
complex, e.g., beet armyworm (Spodoptera exigua) and the cutworms, since
B» t.'s control of these pests has been less than satisfactory.
Future developments may include increased use of baited formu-
lations and improved strains to control cotton pests such as the tobacco
budworm, bollworm, and pink bollworm. (Coddling moth control on apples
might also be obtained by improvements In strains and formulation.)
While good control of bollworm and tobacco budworm infesting cotton have
been obtained with the HD-1 strain in nonbaited formulations, the required
rates of application were about five times higher than normal, resulting
in excessive cost escalation (about 20 times "normal" levels). A new
strain, HD-187, may partially overcome these problems, since its continued
refinement is expected to have about 10 times the potency of HD-1
against H. virescens.
Control activity against forest insects has been promising.
One aerial application of wettable powder formulated in molasses and
water (7 to 8 x 10^ IU/acre) has given good experimental control of
tussock moth on 20-acre plots of Douglas fir and true fir in Oregon and
Idaho over the last few years. Total treatment costs in these experiments
were $8 to $9 per acre, or about twice the typical cost for treatment
with DDT. No tree mortality occurred in treated plots, compared with
about 50 percent mortality in control plots. Prospects for control of
spruce budworm (Choristoneura fumiferana) in Canada appear excellent.
Spruce budworm control represents the largest single future prospect
75
-------
for B.t. in terms of potentially sprayable acreage. Control of gypsy
moth in the Northeast, however, has been somewhat irregular, with
good results obtained in some applications and less effective results
in others.
As with any plant protectant, B.t. is subject to the effects
of weathering once it is deposited on exposed plant surfaces. The
material can be washed from the leaves by heavy rains; this was somewhat
of a problem with molasses-water formulations used against the tussock
moth. Improvements in formulation appear to be needed to increase reten-
tion time under field conditions, especially in high rainfall areas.
Under normal field conditions, however, retention on treated plant sur-
faces has been satisfactory and comparable to most chemical insecticides.
The storage life of B.t. is commercially acceptable, although
bacterial spores are susceptible to UV radiation and decrease in via-
bility with continued exposure to direct sunlight. Generally, spore
viability cannot be demonstrated after three to five days exposure
under normal field conditions. The e'ndotoxin is resistant to the effects
of UV radiation and can persist at least up to two weeks. In general,
the effectiveness of control is more a function of coverage and timing
than of weathering and field stability.
A primary disadvantage of B.t. is its slow action compared to
"quick kill" chemical insecticides which act as contact poisons. Growers
accustomed to seeing worms drop off plants immediately after treatment
can be discouraged when use of B.t. doesn't produce the same visible
results. Objections raised on this basis may be largely subjective,
however, and potentially changeable through grower education and per-
suasion .
Toxin Production
The insecticidal action of B.t. against lepidopterous larvae
is largely caused by the ingestion of the crystalline endotoxin and
its action on the integrity and function of the gut. However, in some
cases the ingestion of spores and their subsequent vegetative prolif-
eration within the insect body are required for pathogenesis. It also
appears that the two mechanisms may complement each other in effecting
the total disease syndrome in certain insects. The exact definition of
the various mechanisms which have been proposed and demonstrated are
still being investigated, and variations within the currently accepted
modes of action will quite likely be found as research continues on the
disease syndromes in additional species of test insects. In the final
analysis, however, the gross toxicological effect of lethal dose
ingestion by susceptible larvae is the cessation of feeding within a
few hours, followed by death within two or more days. Larvae which
ingest sublethal doses of endotoxin and/or spores recover and resume
feeding following a temporary period of intoxication.
76
-------
Three separate toxins have been discovered and their effects
and properties investigated. The toxicological and commercial signif-
icance of each is discussed below.
(1) B.t. Qt-exotoxin. The a-exotoxin, reviewed by Heimpel
and Angus (1963) , is an exoenzyme, phospholipase C
(lecithinase), which is released within the insect
or to the growth medium during active vegetative
growth of the bacterium. This toxin is known to be
lethal to vertebrates. The ability of various B.t.
strains to produce this enzyme has been correlated
with pathogenicity to the larch saw fly (Pristiphora
erichsonii Hart.) and silkworm (Bombyx mori L.).
The action of the enzyme results in damage to the
gut epithelium. Several strains can produce the
toxin in vegetative cultures, including varieties
thuringiensis, sotto, dendrolimus» aizawai, and alesti.
The commercial strain, kurstaki, may also produce the
toxin in view of its close biochemical and serological
relationship to alesti, although no substantive evi-
dence of this has been published. Insect species with
highly alkaline midgut contents, such as most of the
lepidopterous larvae, are resistant to this toxin since
phospholipase has an optimum activity range of pH 6.6
to 7.4. Hence, enzyme activity may be presumed to
be severely limited or inhibited under alkaline mid-
gut conditions. No commercial significance is ascribed
to this toxin in the use of B.t. as a microbial insec-
ticide .
(2) B.t. 3-exotoxin. This toxin is nucleotide-like in struc-
ture and has a molecular weight of about 700. Its
status was recently reviewd by Faust (1972). It is
heat-stable (unaffected by autoclaving at 120°C for
15 minutes), water soluble and dialyzable, and is
secreted into the medium or the insect body during
active vegetative proliferation. This toxin is pro-
duced by B.t. serotypes, I, IVa, IVb, V, VII, IX and X.
It is not produced by the HD-1 strain (var. kurstaki),
and consequently has no role in commercial pest control
using current B.t. formulations.* It has a wider range
of activity than the endotoxin since it is toxic to dip-
teran larvae and pupae as well as to certain Lepidoptera.
It has been shown to be effective as either a contact
poison or a stomach poison.
The major character of this exotoxin is its ability to
prevent pupae of house fly larvae (Musca domestica L.)
(g)
Rhodia's Bactospeine is developed from a strain which produces the
exotoxin, but it is removed prior to formulation.
77
-------
from developing into normal adults, demonstrating a
growth regulator effect at the time of cell division
during molting or metamorphosis. Sublethal doses
cause teratological effects such as vestigial wings;
narrow, pointed abdomens in adult flies; and atrophied
or malformed buccal parts, eyes, and antennae in the
Colorado potato beetle (Leptinotarsa decemlineata).
The mode of action appears to be the inhibition of
RNA synthesis through the blocking of the ATP-dependent
enzyme, RNA polymerase.
Toxicity of ^-exotoxin compares favorably with that of
other insecticides. For example, the exotoxin LD5Q for
the housefly is 1 Mg/g, compared to 10 ug/g for dieldrin;
in the greater wax moth (Galleria mellonella L.) it is
0.55 ug/g compared to 74.2 yg/g for DDT. Pasture mos-
quito larvae (Aedes aegypti L.) are susceptible to media
containing exotoxin (LD50 = 1.9 Ug/g), and the LD5Q of
exotoxin sprayed on larvae of Pieris brassicae L. is
6.0 yg/g. Hall et al (1971) showed that the exotoxin
was highly toxic to eggs, immature stages, and adults
of the red citrus mite (Panoncychus citri), and that
residual activity suppressed population development for
at least 45 days after application. A list compiled by
Faust (1972) shows the B-exotoxin is highly toxic to 15
out of 24 species of Arthropoda treated by injection.
A low level of toxicity was demonstrated against 29
species by feeding.
Resistance in the house fly has been induced 14-fold
through feeding tests covering 50 successive generations.
Deactivation occurs in the gut of Galleria mellonella,
rendering ineffective doses up to 240 times the effective
injected dose.
Commercial utility of the B-exotoxin is restricted by
its high degree of mammalian toxicity (LD50 = 13.3 yg/g
in mice by intraperitoneal injection), by its toxicity
to beneficial insects such as the honey bee (Apis
mellifera), and by the lack of sufficient safety data.
Currently, there is essentially a zero tolerance for
the exotoxin in commercial B.t. formulations, obliging
producers either to use strains which do not produce
the exotoxin (e.g., kurstaki) or to adequately show
that the water soluble toxin has been removed during
production and formulation. (All of the older strains
of B.t. registered in the early and mid-1960s contained
measurable levels of exotoxin.) Responsibility for
monitoring formulations is left to the producers. Until
the safety of 3-exotoxin is more certain, no commercial
use of this material can be developed.
78
-------
(3) B.t. 6-endotoxin. The 6-endotoxin produced by the various
strains is the principal active component of commercial
formulations. This toxin is present only in sporulated
cultures. It is a proteinaceous crystal, octahedral and
bipyramidal in form, and is formed from cellular protein
at the same time the spore is produced in the aging
vegetative cell (sporangium). It is insoluble in water
and organic solvents", soluble in alkali, and thermolabile
(70°C for 30 minutes). Its toxicity varies between strains
but the same principle is involved. Differences in
toxicity between strains might be related to the vari-
ability in toxin solubility in the alkaline midgut juices,
due to differences in amino acid composition and sequence
in the crystal. Susceptible insects are limited to those
with alkaline midgut contents, although certain insects
with alkaline midgut contents have been observed to be
less susceptible than other species with similar pH
characteristics. Perhaps specific solubilizers (e.g.,
enzymes) also play a role.
Aspartic and glutamic acids account for approximately
25% of the amino acids present in the toxin; leucine
is also a major constituent. Some evidence suggests
that silicon forms the basic latticework upon which the
protein molecules are fixed in the crystal. This could
explain its water insolubility and alkali solubility
properties. The endotoxin can be detoxified by a variety
of protein denaturants, including formalin (de Barjac et
al, 1908).
Crystals of the endotoxin are stable either in water
suspensions of whole cultures or in suspensions of
crystals. Crystals in dried preparations apparently
retain their activity indefinitely. In one analysis,
water suspensions of crystals from variety sotto were
still toxic after ten years of storage in the dark at
3°C (Heimpel and Angus, 1963). The dissolved crystal is
less stable, but is still toxic when freshly prepared
(Heimpel, 1967). No mammalian toxicity is indicated,
even at passive doses. Test animals have included mice,
rats, dogs, rabbits, guinea pigs, cows, sheep, pigs,
man, hens, and ducks (Heimpel and Angus, 1963). Man
has been tested both by ingestion and inhalation, with
no ill effects.
(4) B.t. y-exotoxin. This fourth toxin has been
demonstrated in vitro. It is suspected of being an
enzyme, although its precise nature has not been determined
and no commercial significance has been ascribed to it.
It is reportedly produced by entomocidus, serological
type IV varieties of B.t. (Heimpel, 1967).
79
-------
Disease Syndrome
Larvae susceptible to B.t. become intoxicated and inactive
within one or two hours after ingestion. With the ingestion of lethal
doses, death follows in two or more days. The actual cause of death
may vary between insect species, but may include starvation caused by
gut paralysis and breakdown of gut epithelium tissues, gut leakage and
septicemia, a lowering of gut pH levels, and/or a raising of blood pH
levels. With the ingestion of sublethal doses, susceptible larvae can
recover after a temporary period of inactivity and resume feeding. This
suggests the action of a natural detoxification mechanism in the host.
No development of resistance to B.t. has been observed under natural
conditions or demonstrated experimentally.
The first disease symptom is a cessation of feeding caused
by gut paralysis; this occurs soon after ingestion (leading Dulmage to
suspect that the endotoxin may act in part as a nerve poison). Termi-
nation of feeding is followed by a loss of integrity in the midgut,
resulting in the separation, breakdown, and sloughing-off of the epi-
thelial cells lining the interior gut wall. It has been postulated
that one of the substrates for the endotoxin is the cementing material
which binds together the epithelium tissues. In certain cases the
sloughed-off cells cause a gut blockage resulting in death by starvation.
The main effect of gut wall damage, however, is a change in wall perme-
ability which allows the alkaline midgut contents and bacterial spores
to leak into the hemocoel, resulting in septicemia and changes in blood
pH.
Susceptible insects have been separated into four classes or
types on the basis of disease effects (Heimpel, 1967). Type I insects
undergo a rapid gut paralysis and epithelial breakdown with the ingestion
of only the endotoxin. This is followed by wall leakage, blood pH
elevation, and a general larval paralysis. The silkworm (Bombyrx mori L.)
represents this group. Most susceptible insects fall into Type II;
these insects experience gut paralysis but no apparent gut leakage,
blood pH change, or general paralysis resulting from endotoxin ingestion.
The immediate causes of death are starvation and bacterial pervasion
of body tissues. Type 111 insects require the ingestion of both the
spores and the endotoxin before toxic effects are experienced. Larvae
of the flour moth (Anagasta kunhiella) and the gypsy moth (Porthetria
dispar L.) are thought to represent this group. Insects of Type IV
are suspected to be insensitive to the endotoxin alone and susceptible
only to ingested spores. Mamestra brassicae and M. oleraceae represent
this group, although it is not fully agreed whether these insects are
completely immune to the endotoxin.
Toxicology
As mentioned in previous sections, the B.t. 6-endotoxin is
active against many lepidopterous larvae of economic pest status, is
80
-------
non-toxic to beneficial insects, and has no mammalian toxicity. In
contrast, the B.t. ^-exotoxin has a wide range of toxicity including
beneficial insects and mammals, and it is not currently tolerated in
commercial B.t. formulations. No phototoxicity due to B.t. formulations
has been observed.
3. Base Case Analysis
Market Size—B.t. is active against a broad range of lepidop-
teran larvae and the potential market for the product is dependent on
the total crop acreage that is infested by these insects at economically
treatable levels. Table 13 lists the crops for which B.t. might be used
in the future. Shown in the table are the total domestic acres planted
for each crop, SRI estimates of the portion of these acres economically
infested (where treatment is economically justified) with pests suscep-
tible to B.t., estimated growth rates for the infested acres, applica-
tion rates, and peak market shares for each crop. Recommended applica-
tion rates were taken from commercial labels of B.t. formulations.
Market share estimates were made by SRI based on the current price for
commercially offered B.t. formulations, and on considerations of the
relative advantages and disadvantages of B.t. compared to competing
controls for each crop that were drawn from panel surveys. Quantities
used are expressed as pounds of Dlpel® wettable powder equivalent. The
base case estimate includes only the U.S. domestic market.
A maximum attainable market size of 57.9 million pounds of
B.t. per year can be derived from this table. There is considerable
uncertainty in this estimate, but this will be considered in the
probabilistic analysis.
Multiplying the market share estimate for each crop by the
estimated treatable acres, the application rate, and the number of
required treatments gives the total number of pounds of B.t. that will
be sold for use on that crop in the first year of registration. These
calculations are given in the final column of Table 13. It is inappro-
priate, however, to add the individual market sizes (as listed in the
table) to obtain the total initial market size, since all of the crops
will probably not be registered in the same year. The calculations
show that the largest market is likely to be in forests, particularly
for treatment against spruce budworm. Other large contributing crops
to the base case market size are soybeans and cotton. A separate
calculation based on the initial market sizes and long-term growth
rates shows that the base case market size will ultimately reach
8.2 million pounds per year.
From this market size for the product must be derived market
share for the company. Because B.t. is a non-patentable naturally
occurring substance, it is likely that if registration is achieved
other companies will market similar products. Therefore, this analysis
assumed that the company captures 100 percent of the available market in
81
-------
Table 13
BASE CASE MAXIMUM POTENTIAL MARKET SIZE FOR BACILLUS THURINGIENSIS
00
ro
Total
Domestic Acres
Treatable
Domestic Acres
Domestic
Growth Rate
Applications
Per Acre
Dipel®' Equivalent
(pounds per
Estimated Peak
Market Share
Market Size
at First Year
Registration,
Assuming No Growth*
Alfalfa
27,000
540
ox
2
0.25
20%
54.0
Artichoke
12
12
0
6
1.00
20
14.4
Beans
414
5
0
2
1.00
10
1.0
Broccoli
54
54
0
2
1.00
20
21.6
Brussel sprouts
10
10
0
2
1.00
20
4.0
Cabbage
115
115
0
2
1.00
20
46.0
Cauliflower
32
32
0
2
1.00
20
12.8
Celery
36
36
0
2
1.00
10
7.2
Cotton
12,000
4,500
0
6
0,50
5
675.0
Cucumbers
161
16
0
2
1.00
5
1.6
Grapes
538
400
I
2
0.75
5
.3
Lettuce
230
230
0
2
1,00
20
92.0
Melons
111
111
0
2
1,00
5
11 .1
Oranges
885
44
1
2
1.00
5
0.04
Potatoes
1,324
132
0
2
1.00
2
5.3
Tobacco
900
900
0
3
0,75
25
506.25
Tomatoes
446
446
0
2
1.00
10
89.2
Walnuts
161
161
0
2
1.00
10
32.2
Watermelon
263
263
0
2
1.00
5
26.3
Forests
Tussock moth )
Spruce budworai
753,000
4,700
1
2
1.00
35
32.9
Gypsy moth )
Apple/pear
510
255
0
2
0.75
5
19.1
Sweet com
605
605
0
2
0.75
10
90.8
Field cora
62,000
620
0
2
0.75
5
46.5
Popcorn
160
32
0
2
0.75
10
4.8
Soybeans
46*000
12,500
0
2
1.00
10
2,500.0
Sugarcane
714
430
1
2
0.75
5
0.32
Sugarbeets
1,350
340
1
2
1,00
5
0.34
Small grains
101,000
1,000
0
1
1.00
5
50.0
Peas
380
38
0
2
1.00
15
11.4
Asparagus
119
6
0
2
1.00
10
1.2
Home gardens
50
30
0
2
1.00
10
6.0
*
This column is not
suamed because
not all crops are
registered in the same year.
-------
Its first year, but that its market share declines to 85 percent in the
second year, 70 percent in the third year, 55 percent in the fourth
year, and 40 percent in the fifth and all succeeding years.
Another important factor is that new products are not immedi-
ately adopted by all potential customers. There is usually a period of
gradual market penetration over several years. In this base case it
was assumed that it takes five years of linear growth for sales to move
from zero to their peak level. (The period of acceptance for the first
bacterial pesticide to be developed and registered would probably be
longer than five years; however, the hypothetical B.t. product analyzed
was assumed to be a later market entry.)
It was further assumed that once the peak market share is
reached it will be maintained for 20 years, and that sales will then
decline linearly to zero over a 14-year span. The eventual decline
in sales assumed the probable introduction of more efficacious or less
costly products, since the probability of resistance development was
judged to be low.
A constant retail price of $5.12 per lb of wettable powder
formulation was used (the recent retail price of Dipel®) . Not all of
the retail price is recovered by the manufacturer, however, (part of
it is received by middlemen), and the base case assumed that 62.5 per-
cent of $5.12 ($3.20) is actually realized by the manufacturer per
pound of product sold.
Production and Capital Costs— B.t. is currently produced by a
deep fermentation process in large fermentation tanks. The wettable
powder formulation is*essentially a spray-dried material. The principal
production costs are for capital, labor, and raw materials. Commercial
production is likely to be little more than a scaled-up version of
laboratory production. Consequently the production cost estimates were
based on current laboratory experience, adjusted for higher yields
believed to be achievable by large producers. The costs shown in
Table 14 are based on a yield of 15 grams of spray-dried product per
liter of beer.
Capital costs were estimated using traditional chemical
industry rules of thumb. The base case estimate was $2.9 million for
plant and equipment for a 1.2 million pound per year facility. A two
year time requirement to build and bring a plant on-stream was assumed,1
and it was presumed that additional capacity would be made available
.as required to accomodate increased sales levels. Working capital was
set equal to two months' production costs.
83
-------
Table 14
BASE CASE BACILLUS THURINGIENSIS PRODUCTION COSTS
Operating labor $26.84/lb
Maintenance labor $120,000/year/plant
Raw materials (nutrients, lactrose) $0.20/lb
Utilities (electricity, steam,
water) $25.50/lb
Plant overhead $240,000/year/plant
Taxes and insurance $88,000/year/plant
Sales and G&A $0.37/lb
Spray drying $0.10/lb
Packaging $0,05/lb
Storage $0.02/lb
Transportation $0.05/lb
Research and Development Costs—Table 15 provides base case
estimates for the R&D costs that would be incurred to develop B.t.
These estimates assumed that the agent has been observed to be toxic to
certain insects in nature and in the laboratory, but that little field
verification work has been done. Despite the early stage of the product's
development, it is expected that the isolation and identification costs
required will be significantly less than synthesis and screening costs
traditionally associated with chemical pesticide development programs.
The schedule in Table 15 shows R&D expenditures from 1969 to
1985. Expenditures for field testing and registration are continued
until 1985 to allow for additional registrations. However, since the
hypothetical example concerns the decision that a company would have
made in 1974 to enter a possible B.t. venture, the pre-1974 R&D expend-
itures have been compounded at eight percent interest to simulate a
lump sum expenditure in 1974 for all of the earlier work.
Cash Flow—Table 16 shows cash flow projections resulting from
the base case inputs discussed above. There is a small increase in
sales during the mature "flat" sales period due to small growth in
treatable acres for certain crops (grapes, oranges, forests, sugarcane,
and sugarbeets).
The capital cost column shows expenses for new plant and
equipment in the years when they are constructed or purchased. It was
assumed that the company is able to accurately forecast future market
growth, so that plants can increase output at the proper time. All
84
-------
Table 15
BASE CASE RESEARCH AND DEVELOPMNT COSTS FOR BACILLUS THURINGIEHSIS
(Thousands of Dollars)
1969
1970
1971
1972*
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
Isolation and
identification
150
100
50
Field tests
400
600
600
500
300
75
75
75
75
75
75
75
75
75
75
Toxicology tests
200
400
400
300
100
Formulation and
process development
50
200
200
100
Registration
25
125
200
100
50
50
50
50
50
50
50
50
50
50
50
Total
150
775
1,375
1,400
1,000
450
125
125
125
125
125
125
125
125
125
125
*
Tear of first registration.
-------
Table 16
BASE CASE PRE-TAX CASH FLOW FOR BACILLUS
THURINGIENSIS VENTURE
(Thousands
of Dollars)
Net Present Value
Sales at
Production and
Capital
Yearly
(1974
dollars)
Year
R&D Costs
$5.12/lb
Distribution Costs
Costs
Profit
At 8 percent
At 15 percent
1969
$ -150.0
$ -150.0
$ -220.4
$ -301.7
1970
-775.0
$-1,463.0
-2,238.0
-3,044.8
-3,914.3
1971
-1,375.0
-1,463.0
-2,838.0
-3,575.1
-4,316.2
1972
-1,400.0
$ 292.9
$ -568.2
-94.7
-1,770.0
-2,064.5
-2,340.8
1973
-1,000.0
747.2
-754.6
-31.1
-1,038.5
-1,121.5
-1,194.2
19 74
-450.0
1,025.9
-868.9
-19.1
-312.1
-312.1
-312.1
1975
-125.0
1,253.3
-962.2
-15.6
150.5
139.3
130.9
1976
-125.0
1,330.1
-993.7
-5.3
206.1
176.7
155.8
1977
-125.0
1,978.2
-1,259.7
-1,507.3
-913.8
-725.4
-600.8
1978
-125.0
3,204.2
-1,762.7
-1,546.8
-230.3
-169.3
-131.7
1979
-125.0
4,818.8
-2,873.2
-1,648.1
172.5
117.4
85.8
1980
-125.0
6,487.8
-3,558.0
-1,577.1
1,227.6
773.6
530. 7
1981
-125.0
8,186,2
-4,702.9
-190.8
3,167.5
1,848.2
1,190.8
1982
-125.0
9,560.0
-5,266.6
-93.9
4,074.4
2,201.3
1,331.9
L983
-125.0
10,243.2
-5,546.9
-46.7
4,524.5
2,263.4
1,286.2
1984
-125.0
10,552.7
-5,673.9
-21.2
4,732.6
2,192.1
1,169.8
1985
10,745.0
-5, 752.8
-13.2
4,979.0
2,135.4
1,070.2
1986
10,843.0
-5,793.0
-6.7
5,043.3
2,002,7
942.6
1987
10,934.8
-5,830.7
-6.3
5,097.9
1,874.5
828.5
1988
11,020.6
-5,865.9
-5.9
5,148.9
1,753.0
727. 7
1989
11,107.3
-5,901.5
-5.9
5,199.9
1,639.2
639.0
1990
11,194.9
-5,937.4
-6.0
5,251.5
1,532.9
561.2
1991
11,283.3
-5,973.7
-6.0
5,303.6
1,433.4
492.8
1992
11,372.6
-6,010.3
-1,469.1
3,893.2
974.3
314.6
1993
11,462.8
-6,047.3
-1,469.2
3,946.3
914.4
277.3
1994
11,553.8
-6,532.7
-80.9
4,940.3
1,059.9
301.9
1995
11,645.9
-6,570.4
-6.3
5,069.1
1,007.0
269.3
1996
11,738,8
-6,608,6
-6.4
5,123.9
942,5
236.7
1997
11,793.2
-6,630.9
-3.7
5,158.6
878.6
207.2
1998
11,809.1
-6,637.4
-1.1
5,170.6
815.4
180.6
1999
11,825.9
-6,644.3
-1.1
5,180.5
756.4
157.4
2000
11,805.0
-6,635.7
1.4
5,170.7
699.1
136.6
2001
11, 745.0
-6,611.1
4.1
5,138.0
643.2
118.0
2002
11,537.9
-6,526,2
14.2
5,025.9
582.6
100.4
2003
11,034.7
-5,871.7
109.1
5,272.1
565.8
91.6
2004
10,368.0
-5,598.1
45.6
4,815.5
478.6
72. 7
2005
9,649.1
-5,303.1
49.2
4,395.1
404,4
57.7
2006
8,885.6
-4,989.9
52.2
3,947.9
336.4
45.1
2007
8,106.7
-4,670.3
53.3
3,489.7
275.3
34. 7
2008
7,312.4
-3,896.4
129.0
3,545.0
258,9
30.6
2009
6,504.6
-3,564.9
55.2
2,994.9
202.6
22. 5
201*0
5,683.1
-3,227.9
56.2
2,511.4
157.3
16.4
2011
4,847.9
-2,885.1
57.1
2,019.8
117.1
11.5
2012
4,038.2
-2,552.9
55.4
1,540 .7
82.7
7.6
2013
3,254.0
-1,783.2
128.3
1,599.2
79.5
6.9
2014
2,455.5
-1,455.5
54.6
1,054.6
48.5
3.9
2015
1,687.2
-1,140.3
52.5
599.4
25.5
1.9
2016
950.3
-837.9
50.4
162.8
6.4
.5
2017
370,2
-599.9
39.7
-190.0
-6.9
-.5
2018
119.6
-497.1
17.1
-360.4
-12.2
-.8
2019
43.3
-465.8
5.2
-417.3
-13.1
-.8
2020
6.9
-450.8
2.5
-441.4
-12.8
-.7
2021
2.3
-448.9
75.1
-371.5
-10.0
-.5
Total
$134,773.5
$23,107.5
$732.4
86
-------
plants have annual capacities of 1.2 million pounds. Working capital
is,included in the capital column. During the years of increasing
sales, working capital requirements also increase and appear as negative
amounts in the column; when sales decrease, some working capital returns
to the company, and a positive column entry appears.
The columns at the right show the cash flows discounted at
eight percent and fifteen percent. The column sums give the net present
value of the venture at the indicated discount rate. A zero present
value indicates a forecast equivalent in value to a banking investment
opportunity at the same interest rate. With a pre-tax net present value
of over $23 million at eight percent in a non-inflationary world, B.t.
appears to be an attractive business proposition, although the net
present value at fifteen percent falls to only $732 thousand. However,
before deciding whether or not to undertake an expensive development
program, the company must know how these base case financial projections
would be affected by variations in the input estimates. The next section
addresses with these questions of uncertainty.
4. Sensitivity Analysis
Pest Resistance—The company might question the assumption
that pests will not develop resistance to B.t. To simulate this occur-
rence, the product life cycle was changed to a five-year buildup, a
mature period of six years instead of twenty, and a declining period
of nine years instead of fourteen. Under these conditions the net
present value at eight percent decreased by $12.8 million to a level
of $10.3 million.
Failure To Obtain Registration—Another concern might be the
possibility of failing to obtain registration after incurring the R&D
expenditures. Simulating this case resulted in a negative net present
value of -$4 million at an eight percent discount rate, suggesting
that if company management perceived a high probability of nonregistra-
tion they would discontinue development plans. Deciding on the best
course of action for the company if there were moderate chances of non-
registration involves a number of issues which are examined later in
the probabilities analysis.
Size of Market—Market size is another uncertain variable.
The uncertainty might be due to inaccuracies in the estimates of
economically affected acres, the treatment frequency, the B.t. overall
market share compared with other pesticides, and the eventual long-
term market share of the company within the B.t. market. Decreasing
the market size to one-third of the base-case estimate yields a net
present value at eight percent of $3.8 million dollars. Conversely,
if the market is 1.5 times the original estimate, the net present value
is $41 million. Clearly, market size has a major impact on the ultimate
87
-------
profitability of the venture and introduces an element of considerable
risk.
Research and Development Costs—An uncertainty of great
interest to private industry is the escalating cost of pesticide
development. Therefore, the results of higher and lower R&D costs
were tested. Increasing the R&D costs by 33 percent resulted in a
decrease in the net present value of the venture of $2.2 million
dollars at eight percent; decreasing R&D costs by one-third caused
an identical increase in net present value of $2.2 million dollars.
While two million dollars is not a trivial sum, it appears that per-
centage changes in R&D costs do not have as much leverage on the
expected profitability of this venture as do percentage changes in
market size, or the failure to obtain registration.
Delays in Registration—Another frequently mentioned obstacle
to the development of new pesticides is Relayed registration resulting
from additional testing requirements demanded by the EPA. The effect
of this problem has been examined by deferring all sales and production
by one year through an assumed registration delay. This results in a
decrease of $2.2 million to the base case net present value at eight
percent, lowering the net present value to $20.9 million, A two year
delay decreases the net present value by $4.3 million. Again, while
these are significant effects, they appear less important than other
considerations, and are unlikely to be major determinants in this
hypothetical decision of whether to develop B.t¦
Capital Costs—Mass rearing of bacteria on a commercial scale
would be a new venture for many firms. Therefore, the company would
probably have considerable uncertainty about the capital costs. If the
capital costs for a plant and associated equipment were as high as
$6.0 million instead of $2.9 million, the net present value at eight
percent for this B.t. venture would be decreased to $14 million, more
than $9 millon below the base case. If the capital costs were as low
as $2 million per plant, the net present value would be nearly $26 mil-
lion, an increase of $3 million over the base case. The uncertainty
about building costs reflected in these sensitivities, therefore,
results in a range of change of $12 million in net present value, which
is significant in terms of overall profitability.
Price--A constant price of $5.12 per pound was used In the
base case analysis. A company considering a development program in B.t.
would probably be concerned about the possibility that other participants
in the market could reduce prices (or that efficacious alternative
materials might enter the market at a low price). This possibility
was explored by lowering the price to $4.92/lb, a decrease of 20 cents
from the base case value. Under these conditions, the net present value
at eight percent declined to $19.5 million.
88
-------
Alternately, if competitive chemical products were deregistered
it might be possible to raise the price substantially. A scenario with
a price of $5.82 per lb was investigated and the net present value at
eight percent was found to be $35.6 million. It is evident that selling
price has a major impact on the profitability of the venture.
Market Share—Finally, there is the question of the company's
long-term share of the market assuming competition with other B.t.
producers. If no other manufacturer enters the market, its share would
be 100 percent. If a competitor with an advertising or distribution
advantage quickly followed the hypothetical company into the market-
place, its market share could be reduced to only 15 percent. In this
event the net present value at eight percent would be $1.1 million.
It is clear that market share also has a major impact on the venture's
profitability.
5. Probabilistic Analysis
The sensitivity analysis suggests that the B.t. project venture
is basically a good investment; only one of the single variable scenarios
resulted in a negative net present value at eight percent. However,
there were large variations in profit for some factors, and a large
number of losing scenarios could be generated by combining unfavorable
factor values.
Based on the results of the sensitivity analysis step, many
companies would terminate the analysis of the development venture and
attempt a decision; some would probably proceed with the venture even
though the project contained elements of risk. Some companies would
analyze the venture further, however, and such an analysis has been
performed in this study to provide some indications of what size
companies might be expected to develop a bacterial pesticide business.
This probabilistic analysis summarizes the current uncertainty
on the various input factors. The probability assignments were subjec-
tively determined on the basis of the relevant information at SRI's
disposal. Focus is on six major variables:
»
• Market size
• Percentage share of the B.t. market
• Price
• Production cost
• Capital costs for plant and equipment
• Registration success.
Probability distributions on each of the above variables were
encoded, using a reference wheel process. The encoding sessions were
89
-------
performed by a decision analyst on the study team working with various
technical experts at SRI who actually provided the probabilistic assess-
ments. The cumulative probability distributions encoded for each of
the first five variables are shown in Figures 7 through 11. As is often
seen when a particular variable is considered carefully, the base case
number does not always lie in the center of the probability distribution.
In particular, most of the plant and equipment cost distribution is
higher than the $2.9 million base case estimate. Also, the total mature
market size is likely to be significantly less than the base case estimate.
A probability distribution is not shown for obtaining a reg-
istration, since this is a discrete event. However, a 75 percent prob-
ability was assigned to successful registration, assuming that the
product is not the first bacterial pesticide to be registered and that
it was observed to be effective both in nature and in preliminary
laboratory tests.
Given the uncertainties encoded in Figures 7 through 11, a
decision tree can be constructed that summarizes the decisions the
company must make and the uncertain events that might occur. The tree
allows a large number of scenarios on the future of the venture to be
presented in one figure. To construct the decision tree it is necessary
to approximate each of the continuous probability distributions in
Figures 7 through 11 by simple discrete distributions. The process
is similar to making a histogram from a bell-shaped curve.
The decision tree used in this anlaysis is shown in Figure
12. Starting on the far left and most immediately in time, the company
must decide whether to begin a development program. If it decides
against the program, its "reward" is zero. If it proceeds, some years
later it will discover whether the EPA will register the product. On
the assumption that registration is denied, the company will avoid any
further expenditure and lose only $4.4 million (net present value at
eight percent). If registration is successful and the basic develop-
ment program completed, the company will have good estimates on the
plant, equipment, and operating costs required for production. With
this information the firm must make a second major decision: whether
or not to actually produce and market the B.t. pesticide. If it
chooses not to, it has lost the full cost of the R&D program ($6.3 mil-
lion). If it does go ahead, a period of years will elapse before it
knows if the venture is successful. At the end of this time there
will be a resolution of the long-term market price for the B.t. pesti-
cide, a determination of the total market size served by B.t. products,
and a long-term percentage market share established for the hypothetical
company. The tree contains 53 scenarios of the venture's final out-
come. Each scenario has a probability of occurrence determined by
multiplying the probabilities of the events in the scenario. These
probabilities are shown in the Case Probability column of Figure 12.
90
-------
1.0
H t 0.8
ui Z
^ 2
DC O
< 5
s <
0.6
0.4
iiV
ui
m
0.2
2
6
3
4
5
9
10
8
11
7
MILLIONS OF POUNDS
FIGURE 7 CUMULATIVE PROBABILITY DISTRIBUTION ON TOTAL ANNUAL BACILLUS
THURINGIENSIS MARKET SIZE AT PEAK
91
-------
1.0
0.8
u. <
Q 0.6
Z K
Ui <
u u
uu —
*¦ - 0.4
H„v
Z UJ
h ffi
•- d 0.2
J s
0
10
40
SO
60
80
90
MARKET SHARE — percent
FIGURE 8 CUMULATIVE PROBABILITY DISTRIBUTION ON LONG TERM PERCENT MARKET
SHARE OF FIRST COMPANY TO PROVIDE BACILLUS THURINGlENSfS
92
-------
1.0
C 2 0-8
a. 5
*2
Ui Q
0.6
h- <
< O
I 5
>- ~ 0.4
h ii V
5 uj
CO CD
21 0.2
0 •—
3.00
4.00
6.00
6.00
7.00
PRICE PER POUND — dollar!
FIGURE 9 CUMULATIVE PROBABILITY DISTRIBUTION ON RETAIL PRICE OF BACILLUS
THURINGIENSIS
93
-------
1.0
0.8
0.6
0.4
0.2
0
0.80
I
1.00 1.20 1.40 1.60 1.80 2.00
COST PER POUND — dollars
2.20
2.40
2.60
FIGURE 10 CUMULATIVE PROBABILITY DISTRIBUTION ON PRODUCTION COST OF
BACILLUS THURINGIENS/S
94
-------
1.00
3.00 4.00 5.00
MILLIONS OF DOLLARS
6.00
7.00
8.00
FIGURE 11 CUMULATIVE PROBABILITY DISTRIBUTION ON COST OF
BACILLUS THURINGIENSIS PLANT AND EQUIPMENT
95
-------
Nat Praaant
Building
Valua at
DttMon
Capital
Production
Markat Si>a
Eight Pareant
to Fund
Succaaaful
(MUMom of
Coat
Dacition
Sailing Priea
(Thoucanda
Marfcat Shara
Caw
(Thouaandt of
RfcD
Rag titration
Ddlant
to Markat
<»/iw
of Poundal
(SI
Probability
Doilanl
60 p
.011710
.011719
.023438
.023438
p-0.50
.011719
3.62 ^ —!-
p-h.ra 30 P
60 p
9.05
.011719
011719
011719
023436
.023438
p-0.50
011719
Lom $6.3 Million
•t 8%
011719
.011719
9.06 ^
p-0.25
p«0.50
011719
023438
023438
p-0.50
.011719
011719
P-0.25
.011719
p-0.50
.011719
p-0.25
.023438
.023438
.011719
011719
¦t 8%
.011719
.011719
P-0.25
.023438
.023438
p-0.50
.011719
.011719
.011719
p-0.50
LOM 64.4
011719
Mi lion at
.023438
cxC
p-0.60 ^
023438
p-0.50
011719
Lom 66.3 Million
it 8%
011719
.011719
p-0.50
.011719
p-0.26
.023438
.023438
p-0,50
.011719
011719
p-0.26
.011719
p-0.50
.011719
.023438
m~az;
p-0.50
023438
p-0.50
011719
011719
.3 Million
at 8%
33.54!
10,079
18,135
3.595
3,753
-6,493
2,680
-6,005
-3,658
- 7,769
-8,609
-12,910
65,396
26,681
40,629
15,324
16.513
130
34,535
10,597
18,836
3,961
4,151
-6,287
41,446
14,825
24,242
7,167
7,862
-3,498
10,585
-1,259
2,449
-4.T97
-4,500
-9.914
73,301
31,427
46.736
18,896
20.622
3,125
42.439
15,343
24,943
7,532
8,260
-3,291
AVE 8,842.3
FIGURE 12 DECISION TREE FOR BACILLUS THURINGIENS1S VENTURE
96
-------
Of the 53 scenarios, 48 involve bringing the product to
market. For each of these latter scenarios, the net present value
before tax at eight percent has been calculated. These values are
shown in the Net Present Value column of Figure 12. They can now be
reordered by profit level, and their probabilities cumulated to ascer-
tain the probability of a present value less than or equal to a given
amount. The resulting cumulative probability distribution derived
by this process is shown in Figure 13.
As indicated earlier, a zero present value is not necessarily
a bad outcome; it means "break-even" compared to an eight percent bank
investment. According to the profit lottery there is a 45 percent chance
of the net present value being less than zero, and a corresponding
55 percent chance of it being greater than zero. It is possible to
incur a net present value loss as large as $12 million or a profit as
high as $73 million. The "expected value"—the average amount of
money the firm would earn if it could repeat this venture many times—
is $8.8 million.
The hypothetical company faces a complex decision with a high
probability of a moderate loss and a possibility of large gains. However,
if the company cannot risk a loss of $5 million, it should not proceed
with the venture.
The risk of the venture and the size of the company that
could conceivably undertake it can now be analyzed with more precision
by using the concepts of certain equivalence and utility theory dis-
cussed in IV-B-4 ("Method of Approach, Probabilistic Analysis"). The
maximum risk tolerance, a, at which the hypothetical company would be
willing to proceed with the development of this bacterial pesticide
venture was found to be a $7.1 million. In other words, to proceed
with this B.t. development venture, the company should theoretically
be willing to take a risk equivalent to calling a coin toss in which
a win would be $7.1 million and a loss $3.55 million. In SRI's judg-
ment such risks are not likely to be taken by small companies. There-
fore, medium-sized or large corporations would probably develop bac-
terial pesticides.
The analysis considered whether this situation would change
appreciably if it were possible to patent or otherwise obtain exclusive
production rights to such a naturally occurring pesticide by setting
the company's B.t. market share to 100 percent and increasing R&D costs
by 10 percent (assuming that in a patentable situation government
laboratories would do less of the basic research). The price distri-
bution was also encoded, assuming a higher price for a patented product.
The changed price distribution is shown in Figure 14 and the resulting
profit lottery is shown in Figure 15; it has an expected value of
$36 million, a considerable increase from the $8.8 million for the
nonpatentable profit lottery. This probability distribition goes as
high as a $136 million gain (net present value at eight percent) and
97
-------
to
00
1.0
0.9
0-8
D
-I
<
11
ui ¥
S <
£o
ui
H fr-
ill <
* « 03
H O
< z
^¦V 0.4
> UI
H «D
0.7
0.6
°-3
m s
O
c
0.2
0.1
t—i—i—i—i—I—i—r
+ 1111
j i i L
J L
J L
J L
10 15 20 25 30 35 40
MILLIONS OF DOLLARS
45
50
55
60
65
70
75
FIGURE 13 CUMULATIVE PROBABILITY DISTRIBUTION ON NET PRESENT VALUE. AT EIGHT PERCENT.
FOR BACILLUS THURINGIENSIS VENTURE
-------
e.oo
PRICE PER POUND
6.00
7.00
dollar*
FIGURE 14 CUMULATIVE PROBABILITY DISTRIBUTION ON RETAIL PRICE OF BACILLUS
THURINGIENSIS IN 1985 FOR A PATENTABLE CASE
99
-------
o
o
50 60 70 80
MILLIONS OF DOLLARS
140
FIGURE 15 CUMULATIVE PROBABILITY DISTRIBUTION ON NET PRESENT VALUE. AT EIGHT PERCENT, FOR THE
PATENTABLE BACILLUS THURINGIENStS CASE
-------
as low as a $4.4 million loss. Patent protection, under these assump-
tions, clearly provides a more favorable business climate.
This protected scenario was tested for affect on risk tolerance.
The risk tolerance coefficient was determined to be $3.5 million, mean-
ing that to enter a B.t. development venture with a protected product,
a company must be willing to take risks comparable to flipping a coin
to wind $3.5 million or lose $1.75 million. Therefore, product pro-
tection halves the risk tolerance, and roughly halves the size of the
company that could reasonably initiate a development program for this
venture. From the standpoint of promoting the development and use of
such innovative pesticides, an exclusive license policy appears to
provide more commercial development incentive than a reduction in
registration costs or time.
E• Survey Questionnaire
A copy of the questionnaire sent to the panel of experts on the
bacterial pesticide survey is presented in Figure 16 on the following
pages.
101
-------
May 1975
BACTERIAL PESTICIQES - CRITICAL FACTORS
IN SUCCESSFUL DEVELOPMENT
While numerous species of bacteria are known to be toxic to certain economically important insect pests, relatively little
effort has been expended to develop such species as commercial insecticides. [Two notable exceptions are the crystallifer-
ous bacterium, Bacillus thuringienw Berliner (Biotrol, Dipel, Thuricide), and the milky spore organism. Bacillus popilliae
Dutky (Doom), which are produced commercially.] However, the potential for developing new pesticides based on
bacteria continues to be of interest to industry, users, researchers, and governmental regulatory agencies. This survey is
intended to solicit your views on this subject.
I. GENERAL
In this section we are interested in your impressions and opinions of the commercial potential of bacterial pesticides in
general, relative to "chemical" pesticides.
A. Please indicate your general impression of the potential of bacterial pesticides by answering the two questions below.
t. Do you believe that bacterial pesticides realistically have a place in the pesticide arsenal and that future efforts to
develop such agents on a commercial scale are warranted?
Yes No
2. Do you believe that bacterial pesticides will have made significantly greater progress in supplanting "chemical"
pesticides by 1985 than they have to-date?
Yes No
Remarks:
B. Assume that continued or increased research to develop new or improved bacterial pesticides will be supported in the
future. What percentage of the needed funds will each of the following groups provide?
% of
Interest Group Total Funds
Private Industry
Pesticide producers
Grower groups
Federal Funding Agencies
Federal Laboratories (in-house research budgets) ___________
Total 100%
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES
102
-------
Does the division of funding responsibility predicted by you agree with your opinion of how the division should be
made?
Yes No
Remarks:
C. Below is a list of factors which may bear on the eventual commercial feasibility of bacterial pesticides.
Production costs
Price to user
Unit profitability
Potential sales volume
R&D costs
Registration costs
Federal research support
Time required for registration
Patentability
Storage stability
Field stability
Efficacy of control
Sophistication of user
Strain stability (mutability)
Strain selection potential
Host specificity
Host resistance
Mammalian toxicity
Phytotoxicity
Toxicity to natural predators
Safety to wildlife
Compatibility with other pesticides
Public image
User attitudes
Deregistration of "chemicals"
Development of new "chemicals"
Other
1. From this list please select seven (7) factors (or less) which in your opinion most reflect an ad vintage of bacterial
pesticides over "chemical" pesticides. List, in the spaces provided below, the factors you have chosen in order of
decreasing advantage. Add any remarks you feel are needed to illucidate the reasons for your choice.
Rank
Advantage
Factors
Remarks
2
3
4
5
6
7
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued)
103
-------
2. Prom the same list please select seven (7) factors (or less) which you feel reflect the most disadvantageous
characteristics of bacterial pesticides relative to their "chemical" counterparts. List your choices below in order of
decreasing importance as deterrents to commercialization. Add in the spaces provided for "Remarks" any
elaboration on your choices that you feel may be instructive.
Disadvantage
Rank Factors Remarks
1
2
3
4
5
6
7
D. If a company chose to develop a new pesticide, what costs would you project in each of the categories below if the
company chose to develop (A) a bacterial pesticide, or (B) a conventional "chemical" pesticide? Assume that a high
level of promise for the bacterial agent had been observed on a single major target insect and that it could be
effectively cultured on synthetic media, but that no definitive research had been conducted beyond identifying the
bacterium and preliminary pathogenicity tests. For the chemical pesticide, assume costs of a typical program leading
to a single marketable product, but including costs of failures.
Costs ($10001
(A)
Category Bacterial Chemical
Screening and Synthesis
Field Testing
Toxicology
Formulative and Process Development
Registration
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued)
104
-------
E. While only two bacterial agents have been developed at commercial pesticide* to-date, it is possible that others may be
developed in the future. We are interested in your opinions in this area.
1. Please list below any bacterial agents which have not yet been commercialized, but which in your opinion have
definite potential for development as commercial pesticides. Please indicate the particular advantages or features of
each of your choices which you feel indicate this level of potential.
Agent Advantage
A.
B.
C.
2. While you believe that the agents you have listed have definite potential, do you think they will actually become
"commercialized" {e.g., registered and available for use):
by 1980? by 198S7
Agent Yes No Yes No
A
B
C
Remarks:
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued)
105
-------
3. For the agents you have listed as registered arid available for use by 1980 and/or 1985, for which pests and crops will
each be used (e.g., registered)?
Agent Pests Crops
A
B._
C-
Remarkt:
FIGURE 10 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued*
106
-------
11. BACILLUS THURINGIENSIS
This portion of the questionnaire pertains only to Bacillus thuringiensis {B.f.). Various formulations of this bacterial
agent have been produced as a bacterial insecticide for several years. B.t., therefore, represents a case in point of a
commercialized bacterial pesticide. We are interested in obtaining your impressions or opinions on the current and future
viability of B.t. as a commercially competitive insecticide. While we realize that hard information is not always available
and that your response may be restricted in many cases by uncertain or insufficient data, please answer the following
questions as completely as possible, using your "best guess" if necessary.
A. It has been estimated that approximately one million pounds of B.t. formulation are now marketed in the U.S.
annually. Do you think that this level will remain fairly static over the next 10 years?
Yes No
If no, to what level do you expect use to reach?
B. In your opinion, what are the major obstacles to a greater use of B.t, in the future? List below in order of decreasing
magnitude or impact.
1.
2.
3.
4.
5.
C. Which, if any, of these obstacles do you think are likely to be eliminated or mitigated by reasonable advances in
research and development? Please indicate the general mechanism by which each obstacle you list may be overcome.
Obstacle Mitigating Mechanism
1.
2.
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued)
107
-------
Obstacle
Mitigating Mechanism
3.
4.
D. Which, if any, of the obstacles listed in (B) do you regard as insurmountable? Why?
Obstacle Reason
1.
2.
3.
4.
E. If you expect the total use of B.t. (all forms or strains) to increase in the future, on which crops or crop groups listed
below do you expect this increase to occur? Indicate by checking in the appropriate space(s) below.
Crop(s) (y/ ) Remarks
Crops now registered for B.t.
Other vegetable crops
Forests
Alfalfa
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued)
108
-------
Crop(sl (>/) Remarks
Cotton
Tobacco
Corn
Orchards and grapes
Small grains
Other field crops
Home gardens and ornamentals
Others
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued)
109
-------
F. Please indicate your estimate of the market share (% of total insecticide-treated acres) that might be obtained by B.t.
in each of the following crops or crop groups by 1980, and by 1985. Assume any advances in research or technology
that you consider probable within the time frame indicated.
% Market Share
Crop(s) by 1980 by 1985
Crops now registered for B.t.
Other vegetable crops
Forests
Alfalfa
Cotton
T obacco
Corn
Orchards and grapes
Small grains
Other field crops
Home gardens and ornamentals
Others
G. Do your estimates of future market shares in (F) include any expectation on your part of new, "improved," more
highly virulent, or more selectively active strains being found or developed within the stated time frame?
Yes No,
Remarks:
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Continued)
110
-------
H. The development of resistance by certain insects to various "chemical" insecticides over time is well documented. In
your opinion, what is the probability of any major target insect developing resistance to B.t. during the next 20 years?
(If you have forecast an increase in the future use of B.t., assume that level of use is obtained during the next 20
years. If you have not forecast an increase in use, assume that the current level of use is maintained over the next
20 years.)
Probability (%)
Remarks:
4 m • •"i'i'lTliri'fc 4 j '~ 4 « * »Yi i «'• S *T
When you have completed this survey, please return to:
Thomas A. Blue
Manager, Agricultural Chemicals
Chemical Industries Center
Stanford Research Institute
Menlo Park, California 94025
Your assistance is deeply appreciated. When the final survey results ari published, we will be glad to send you a com-
plimentary copy. Please indicate below where it should be sent*:
•Th« contents of individual surveys will not be divulged but we do need to know who responded In case M have additional questions and
in order to distribute survey results.
FIGURE 16 SAMPLE QUESTIONNAIRE—BACTERIAL PESTICIDES (Concluded)
1X1
-------
F. Selected References
The following is a list of some of the more pertinent literature
references relating to bacterial pesticides that were screened during
this report.
"Budworm Target of Maine's Bacillus Thuringiensis Tests," Weeds, Trees
and Turf (March 1975).
Creighton, C. S., and T. L. McFadden, "Cabbage Caterpillars: Efficacy
of Chlordimeform and Bacillus thuringiensis in Spray Mixtures and
Comparative Efficacy of Several Chemical and B. thuringiensis
formulations," Journal of Economic Entomology, Vol. 68, No. 1,
pp. 57-60 (1975).
deBarjac, H., and A. Bonnefoi, "Classification des Souches de Bacillus
thuringiensis", Academie des Sciences Comptes Rendus Hebdomadaires
Seances, Series D> 264::1811—1813 (1967).
deBarjac, H. , and A. Bonnefoi, "A Classification of Strains of Bacillus
thuringiensis Berliner with a Key to Their Differentiation",
Journal of Invertebrate Pathology, Vol. 11, pp. 335-347 (1968).
deBarjac, H., A. Burgerjon, and A. Bonnefoi, "Detoxification of the
Crystal Toxin of Bacillus thuringiensis by Formalin", Journal of
Invertebrate Pathology, Vol. 11, pp. 335-347 (1968).
deBarjac, H., and F. Lemille, "Presence of Flagellar Antigenic Subfac-
tors in Serotype 3 of Bacillus thuringiensis", Journal of Invertebrate
Pathology, Vol. 15, pp. 139-140 (1970).
deBarjac, H., and F. Lemille, "Presence of H Antigenic Subfactors in
Serotype V of Bacillus thuringiensis with Description of a New Type:
B. thuringiensis var. canadensis", Journal of Invertebrate Pathology.
Vol. 20, pp. 212-213 (1972).
deBarjac, H., and A. Bonnefoi, "Mise au Point sur la Classification des
Bacillus thuringiensis", Entomophaga, Vol. 18, pp. 5-17 (1973)
Dulmage, H. T., Economics of Microbiol Control, Entomology Research
Division, ARS-USDA, Brownsville, Texas (19 70).
Dulmage, H. T. "Insecticidal Activity of HD-1, A New Isolate of
Bacillus thuringiensis var. alesti". Journal of Invertebrate
Pathology, Vol. 15, pp. 232-239 (1970).
Dulmage, H. T., et al, "A Proposed Standardized Bioassay for Formulations
of Bacillus thuringiensis Based on the International Unit," Journal
of Invertebrate Pathology, Vol. 18, pp. 240-245 (1971).
Faust, R. M., "The Bacillus thuringiensis fc-Exotoxin: Current Status,"
Bulletin of the Entomological Society of America, Vol. 18, No. 4,
pp. 153-156 (1973).
112
-------
Hall, I. M., D. K. Hunter, and K. Y. Arakawa, "The Effect of the B-exo-
toxin Fraction of Bacillus thuringiensis on the Citrus Red Mite",
Journal of Invertebrate Pathology, Vol. 18, pp. 359-362 (1971).
Heimpel, A. M., "A Critical Review of Bacillus thuringiensis var.
Thuringiensis Berliner and Other Crystalliferous Bacteria," Annual
Review of Entomology, Vol. 12, pp. 287-322 (1967).
Heimpel, A. M., and T. A. Angus, "Diseases Caused by Certain Spore-
forming Bacteria", Insect Pathology: An Advanced Treatise,
E. A. Steinhaus, Editor, Chapter 2, pp. 21-73, Academic Press (1963).
•*
Heimpel, A. M., and T. A. Angus, "The Taxonomy of Insect Pathogens
Related to Bacillus cereus F. and Fr.", Canadian Journal
of Microbiology, Vol. 4, pp. 531-541 (1958).
Jacques, R. P., "Methods and Effectiveness of Distribution of Microbial
Insecticides, Annuals of the New York Academy of Sciences, Vol. 217,
pp. 109-119 (1973).
Klassen, W., A Look Forward in Pest Management, paper presented at the
6th Annual Tall Timbers Conference on Ecological Animal Control
by Habitat Management, Gainesville, Florida (March 1, 1974).
Lewis, F. B., et al., "Gypsy Moth: Efficacy of Aerially-Applied Bacillus
thuringiensis," Journal of Economic Entomology, Vol. 67, No. 3,
pp. 351-354 (1974).
Norris, J. R., "The Classification of Bacillus thuringiensis", Journal of
Applied Bacteriology. Vol. 27, pp. 439-447 (1964).
Norris, J. R., "The Use of Micro-organisms for the Control of Insect
Pests", Chemistry and Industry, pp. 1941-1945 (November 18, 1967).
Smith, R. F., (ed.), "Considerations on the Safety of Certain Biological
Agents for Arthropod Control," WHO Bulletin 48, pp. 685-698 (1973).
"Two Fronts: Gypsy Moths," Weeds, Trees and Turf (March 1975).
113
-------
VII VIRUSES
A. Overview
Currently, insect viruses are not commonly used as commercial pesti-
cides, although considerable attention continues to be focused on their
development. Considerable uncertainty continues to exist about certain
factors that are critical to their development, but the following para-
graphs summarize SRI's current judgment of these agents' potential as
commercial products.
1. Limiting Factors
The following factors apply to viruses in general, but their
relative importance varies between viral complexes.
Efficacy—Viruses are almost by definition efficacious popula-
tion control agents for their specific hosts. However, this statement
applies to natural conditions and cannot necessarily be extrapolated to
the economic efficacy desired in agricultural pest management. Although
efficacy has been demonstrated under specialized conditions, for many
potential viral products large-scale field tests are still considered
inconclusive, and consistent economic efficacy has not yet been proven
over a wide range of field conditions and locations.
Since viral agents must be ingested and are slow to act, some
level of target pest crop damage generally continues for a period of
time after application. This may limit their potential use to the con-
trol of leaf feeding insects, and to crops where some level of foliar
damage is not detrimental to yield or to marketable crop quality. It
also introduces an education cost associated with convincing growers
that observed levels of such "residual" feeding are not economically
detrimental.
Product Specificity, Market Size—Since most viruses are pest
specific, market size will generally be small, although there are excep-
tions. Market size may also be adversely affected by lack of patent-
ability, potential competition from other firms, and competition in the
market from other pesticides.
115
-------
Production, Distribution, Storage, and Field-life Character-
istics—Current technology requires that viruses be produced in living
hosts. Known production schemes are labor intensive and small scale
operations dissimilar to most chemical pesticide production processes.
Viral agents are short-lived and fragile if not properly protected.
Technology has not been developed to assure adequate distribution and
storage for many agents, although there have been improvements here.
Effective field life is short, ami considerable sophistication on the
part of the user may be required to assure optimum efficacy.
Classification Schemes—There is some opinion that viruses may
exist as complexes rather than as more narrowly defined agents, and that
effective pest control viral complexes under natural conditions may
mutate along with fluctuations in the immunity systems of the target
host(s) (Falcon, L.A., D. Kennedy, 1975). These phenomena have increased
some researchers' concern about the long term safety of large scale
production and distribution of viral pesticides. They may also effec-
tively increase production costs because of the need for extensive qual-
ity control to ensure (1) that production runs generate a product of
consistent.composition, and (2) that each batch will not be out of phase
with the immunity system of target insects in each market area at a given
point in time.
Research and Development Costs—No evidence collected to date
suggests that total R&D costs for viral pesticides would be less than
those for chemical pesticides; however, the composition of these costs
would be different.
Commercial Activity—Within the private pesticide industry,
only two companies (neither of which is currently a major industry factor)
exhibit significant identifiable activity on viruses. Many firms may
not see viruses as an attractive investment. Others—for example, firms
with a heavy investment in human health protection—may be avoiding in-
volvement because of possible damage to their corporate image.
2. Current Development Potential
Although private industry does not now generally view viral
pesticides as an attractive investment opportunity, certain viral agents
may nonetheless become commercially available within the next five to
ten years. The following factors are in support of this statement.
Efficacy and Production Problems—Over 50 percent of the ob-
servers surveyed in this project felt that most viral agent efficacy
116
-------
and production problems could or would be overcome within the next
few years.*
Advantages over Chemical Pesticides—Although currently avail-
able data do not suggest that viral pesticides have any meaningful eco-
nomic advantage over traditional chemical pesticides, this could change
if (1) pest resistence to chemical products increases, (2) EPA deregis-
ters prominent chemical products, or (3) current chemical pesticide and
carrier costs rise faster than those for potential viral products.
Integrated Pest Management—Many observers feel that pest con-
trol costs will increase, causing growers to be more prone to accept
pest management practices recommended by pest management professionals.
This kind of market environment should be conducive to more widespread
use of viral pesticides—if they were available.
Public Agency Markets—The panel survey identified four viral
pesticides expected to be commercially developed against nonagricultural
tree pests: Porthetria dispar NPV for gypsy moth, Neodiprion complex
NPV for pine sawflies, Choristoneura fumiferana NPV for spruce budworm,
and Orgyia pseudotsugata NPV for Douglas fir tussock moth.+ The primary
buyer in these markets would likely be a governmental body such as the
U.S. Forest Service. The supply of such products would essentially be
outside the private agricultural market; such narrow-spectrum materials
would probably be produced for the government on a contract basis—or
possibly by the government itself. (Small quantities of some of these
viruses are currently being produced and used experimentally.)
*However, some informed contacts felt that current problems are essen-
tially the same as several years ago, and that little real progress has
been made toward safe, efficacious, and economic viral pesticides.
The availability of research funds will be the most important factor
determining whether these problems can be overcome, and private industry
is currently not oriented toward supplying this funding.
tQrgyia pseudotsugata NPV for control of Douglas fir tussock moth received
Registration directly prior to the publication of this report. It is pro-
duced for the U.S. Forest Service by Nutrilite Products in a form-
ulation containing a lignosulfonate material (Sandoz's Shade®), made
from Douglas fir bark, to protect the virus from photodegradation. The
product is applied in the 1st 3rd instar at a current cost of about
$5 per acre. A current year's growth must be sacrificed to achieve con-
trol. However, control approaching eradication is claimed, and the Forest
Service has indicated it intends to submit additional two-year test data
toward the goal of registering the product for "population control."
117
-------
Other Motivation for Development—Significant amounts of re-
search funds have already been spent on certain viruses for pesticide
use. There is some opinion that certain products may be promoted in
order to justify (1) research expenditures already made, (2) the need
for future development, and (3) possibly for peripheral and unspecified
reasons (in the case of private industry). In addition any market intro-
duction of such products could serve as a basis for establishing regis-
tration criteria for viral pesticides and hence further stimulate
future commercial work.
Most-Likely Potential Products—The project surveys identified
five products deemed most likely to be commercially developed over the
next ten years. Heliothis spp. NPV has received the most public and
private research effort, and at least one company has applied for regis-
tration.* The primary target is the large cotton worm complex market.
Autographa californica NPV, Plusiinae viruses (Trichoplusia), and
Spodoptera complex NPV were all given some chance of becoming commer-
cially available by 1985. Primary markets would be cole crops, alfalfa,
soybeans, and cotton. However, unlike Heliothis spp. NPV, no major work
is known to be currently under way within private industry on any of
these agents. Porthetria dispar
-------
The best known insect viruses belong to the genus Baculovirus and
include the nuclear polyhedroses (NPVs) and the granuloses (GVs).
Viruses have been isolated from more than 400 species of insects and
mites. Table 15 presents important viruses often identified as possible
substitutes for or supplements to chemical pesticides between now and
1985.
Important characteristics relevant to the pesticidal action of
viruses include:
• Generally pest specific
• Relatively slow acting
• Relatively short field life
• Generally propagated in living hosts.
Specificity of viral agents in infecting only target pests is technically
and ecologically desirable.* However, specificity can be a disadvantage
to the producer by limiting market sizfe and sales volume. A virus which
could control several pests would also be more attractive to users and
would not only compete with broad-spectrum chemical pesticides, but could
also compete directly with other viruses for market share. For example,
Heliothls spp. NPV and Autographa californica NPV both cause epizootics
in Hellothis virescens (budworm), but Autographa californica NPV infects
at least six other insect species while Heliothis spp. NPV is known to
only infect the larvae of five species of the genus Heliothis.
Since viruses must be ingested by target pests to cause an epizootic,
viruses as pest control mechanisms are generally slow acting. This has
two practical implications. First, viruses are more effective against
leaf feeding insects than against boring insects. Second, viruses' slow
acting control may not be appropriate for crops where physical appearance
is an important market feature or where the degree of "contamination"
with insect parts is restricted by FDA regulations (e.g., certain canned
or frozen vegetables and fruits). Other crops can sustain considerable
cosmetic damage and not be penalized in the marketplace. However, even
these crops will eventually suffer yield loss if physical damage from
insects is excessive. The important distinction is what parts of plants
are damaged by target insects, and what degree of cosmetic and yield
damage can be tolerated.
Viruses are generally short lived in the field, being both heat
labile and light sensitive. Effective ultra-violet (UV) screening tech-
nology is still developmental; with UV screens that are currently avail-
able, field half-life seldom exceeds two to three days. Short field
*Some of the more important NPVs—particularly Autographa californica
NPV and Trichoplusia nl NPV—do cross-infect to more than one specie
of insect and there is the potential possibility that some viruses
could cross-infect to higher order organisms.
119
-------
Table 15
IMPORTANT INSECT VIRUSES WITH POTENTIAL
AS COMMERCIAL PESTICIDES
Virus Complex
Agrotis, Peridroma, NPV
Argyrotaenia velutinana GV
Autographa californica NPV
Cadra (Ephestla) cautella NPV, GV
Chilo suppressalls GV, IV
Choristoneura fumlferana NPV, GV, CPV
Heliothis spp. NPV
Hemerocantpa leucostigma NPV, CPV
Laspeyresla pomonella GV
Malacosoma complex NPV
Mamestra brassicae NPV, CPV
Neodiprion complex NPV
Orgyia pseudotsugata NPV
Peridroma saucla NPV, GV
Phthorimaea operculella GV
Pieris spp. NPV, CPV, GV
Plusiinae viruses NPV
Porthetria dispar NPV
Spodoptera complex viruses NPV
Pest
Cutworms
Red-banded leafroller
Alfalfa looper, pink bollworm,
cotton leaf perforator, tobacco
budworm, cabbage looper, beet
armyworm, salt-marsh cater-
pillar, cabbageworm
Almond or fig moth
Asiatic rice borer
Spruce budworm
Bollworm, tobacco budworm,
corn earworm
White-marked tussock moth
Codling moth
Forest tent caterpillars
Mamestra brassicae
Pine sawflies
Douglas fir tussock moth
Variegated cutworm
Potato tuberworm
Crucifer caterpillars
Cabbage looper complex
Gypsy moth
Armyworms and leafworms
120
-------
life and the requirement for ingestion often make it difficult to con-
vince growers of "equivalent" field efficacy between viral and chemical
pesticides.
Virus propagation is currently achieved by infecting living larvae
of the target host; cell or tissue culture techniques are not sufficiently
developed for mass propagation methods. These iji vivo techniques are
relatively expensive and require proportionately large labor expendi-
tures, thereby affecting product costs and their relationship with alter-
native controls.
The following sections summarize:
• The results of a panel survey undertaken to identify the
factors affecting the commercialization of vir^l complexes
as pesticides.
• An economic analysis of the decision to enter a possible
viral pesticide venture.
C. Panel-Survey Analysis
1. Sample Copy
Figure 23 at the end of this chapter presents the viral pesti-
cide questionnaire. Fifty-two were mailed; forty-one contacts responded,
and twenty-eight of these completed the forms. Those who did not com-
plete the forms generally disqualified themselves as not being familiar
enough with the subject to provide a satisfactory response.
2. Prospective Viral Agents
Table 16 lists the viruses identified by respondents as having
the best chance for commercialization. Each virus was selected by at
least one respondent as having either the "best," "second best," or
"third best" chance for commercialization. The crops primarily affected
by the target pests are also listed in the table. Table 17 lists the
viruses according to the individual ratings for each response.
As indicated in Table 18, the virus that respondents thought
most likely to be developed commercially was Heliothis spp. NPV,* listed
in 79 percent of the questionnaires; Autographa californica NPV appeared
in 64 percent. It is important to note that five viruses received most
of the attention, with the remainder receiving 18 percent or less of the
citations.
*Already commercial (i.e., first registration received) as of the pub-
lication of this report.
121
-------
Table 16
POTENTIAL VIRAL PESTICIDES,
TARGET PESTS, AND PRINCIPAL CROPS AFFECTED
Virus
Hellothia spp. NPV
Target Pest(s)
Autographa callfortilca NPV
Porthetrla dlspar NPV
Heliothis zea
(bollworm, corn earworm,
tomato fruitworm)
Heliothls vlrescens
(tobacco budworm)
Autographa californlca
(alfalfa looper)
Pectlnophora gossvpiella
(pink bollworm)
Bucculatrix thurberlella
(cotton leaf perforator)
Hellothls vlrescens
(tobacco budworm)
Trlchoplusla nl
(cabbage looper)
Spodoptera exlgua
(beet armyworm)
Eatlgmene acrea
(salt-marsh caterpillar)
Plutella zvlostella
(cabbage worm)
Porthetria dlspar
(8iyP8y moth)
Principal Crop(s) Affected
Cotton, soybeans, tobacco,
corn, sorghum
Alfalfa, cotton, cole
crops, corn, sugarbeets,
tomatoes, potatoes, onions,
peas, asparagus, table
beets, lettuce, soybeans
Deciduous and evergreen
trees and shrubs of north-
eastern U.S.
Plusiinae viruses
Spodoptera Complex NPV
Trlchoplusla ni
(cabbage looper)
Spodoptera exigua
(beet armyworm)
Spodoptera litura
(cotton leafworm)
Cole crops, lettuce,
spinach,' beets, peas,
celery, parsley, potatoes,
tomatoes, soybeans, cotton
Sugarbeets, table beets,
asparagus, cotton, alfalfa,
corn, lettuce, tomatoes,
potatoes, onions, peas,
citrus
Neodiprlon spp. NPV
Neodiprlon lecontei
(red headed pine sawfly)
Neodiprlon pratti banksianae
(jack pine sawfly)
Neodiprlon pratti pratti
(Virginia pine sawfly)
Pinus spp.
122
-------
Virus
Table 16 (concluded)
Target Pest(s)
Principal Crop(s) Affected
Neodiprion spp. NPV
(continued)
Hemerocampa spp. NPV
Laspevresia pomonella GV
Agrotls NPV
Choristoneura fumlferana
Pieris spp. NPV
Neodiprion sertlfer
(European pine sawfly)
Neodiprion swalnel
(Swalne's jack pine sawfly)
Hemerocampa pseudotsugata*
(Douglas fir tussock moth)
Laspevresia pomonella
(codling moth)
Grapholltha molesta
(oriental fruit moth)
t
Agrotis segetum
Choristoneura fumiferana
(Spruce budworm)
Pieris brasslcae
(large white butterfly)
Pieris rapae
(imported cabbageworm)
Pieris spp.
Forests
Apple, pear, quince,
English walnut, peach,
apricot, plum
Cotton
Forests
Kale crops
Various vegetable crops
Cabbage
Hemerocampa pseudotsugata is actually known as the white marked tussock moth, while
Orgyla pseudotsugata is the Douglas fir tussock moth.
^Virus may be\cross-infective to other species of cutworms.
123
-------
Table 17
VIRUSES WITH GREATEST COMMERCIAL POTENTIAL IDENTIFIED BY QUESTIONNAIRE RESPONDENTS
.Juestionnaire
Number Best Chance Second Best Chance Third Best Chance
1
Porthetria dispar NPV
Neodiprion NPV
Autographa californica NPV
2
Heliothis spp. NPV
Spodoptera complex NPV
Autographa californica NPV
3
Autographs californica NPV
Porthetria dispar NPV
Heliothis spp. NPV
4
Autographa californica NPV
Plusiinae viruses
Porthetria dispar NPV
5
Heliothis spp. NPV
Plusiinae viruses
Spodoptera complex NPV
6
Heliothis spp. NPV
Plusiinae viruses
Spodoptera complex NPV
7
Heliothis spp. NPV
Autographa californica NPV
Porthetria dispar NPV
8
Plusiinae viruses
Spodoptera complex NPV
Autographa californica NPV
9
Heliothis spp. NPV
Plusiinae viruses
Orgyia pseudotsugata NPV
10
Autographa californica NPV
Heliothis spp. NPV
Porthetria dispar NPV
11
Heliothis spp. NPV
Autographa californica NPV
Spodoptera complex NPV
12
Heliothis spp. NPV
Plusiinae viruses
Spodoptera complex NPV
13
Heliothis spp. NPV
Autographa californica NPV
Spodoptera complex NPV
14
Heliothis spp. NPV
Plusiinae viruses
Autographa californica NPV
15
Heliothis spp. NPV
Plusiinae viruses
Porthetria dispar NPV
16
Plusiinae viruses
Heliothis spp. NPV
Agrotis NPV
17
Porthetria dispar NPV
Heliothis spp. NPV
Autographa californica NPV
18
Heliothis spp. NPV
Autographa californica NPV
Porthetria dispar NPV
19
Laspeyresia pomonella GV
Neodiprion NPV
Heliothis spp. NPV
20
Autographa californica NPV
Porthetria dispar NPV
Neodiprion NPV
21
Porthetria dispar NPV
Autographa californica NPV
Neodiprion NPV
22
Heliothis spp. NPV
Neodiprion NPV
Choristoneura fumiferana NPV. GV. CP
23
Heliothis spp. NPV
Plusiinae viruses
Laspeyresia pomonella GV
24
Autographa californica NPV
Heliothis spp. NPV
Choristoneura fumiferana NPV. GV. CP
25
Spodoptera complex NPV
Heliothis spp. NPV
Plusiinae viruses
26
Autographa californica NPV
Laspeyresia pomonella GV
Pieris spp. GV
27
Heliothis spp. NPV
Autographa californica NPV
Pieris spp. GV
28
Heliothis spp. NPV
Porthetria dispar NPV
Autographa californica NPV
-------
Table 18
PERCENTAGE OF RESPONDENTS CITING SELECTED VIRUSES
AS HAVING THE BEST POTENTIAL FOR COMMERCIALIZATION
Percentage of
Virus Respondents*
Heliothis spp. NPV 79%
Autographa californica NPV 64
Plusiinae viruses 39
Porthetria dispar NPV 39
Spodoptera spp. NPV 29
Neodiprion spp. NPV 18
Laspeyresia pomonella GV 11
Choristoneura fumiferana
NPV, GV, CPV 8
Pieris spp. GV 7
Agrotis NPV 4
Orgyia pseudotsugata NPV 4
*
Percentage of respondents selecting the virus
as one of three having the best chances for
commercialization.
Fifty-two percent of the respondents cited Heliothis spp. NPV
as the virus with the "best chance" for commercialization; twenty-two
percent chose Autographa californica NPV. For the virus with the "second
best" chance for commercialization, 30 percent of the respondents selected
the Plusiinae viruses, 22 percent selected Autographa californica
NPV, and 19 percent selected Heliothis spp. NPV.
Three viruses were listed by 19 percent of the respondents as
having the "third best" chance for commercialization: Autographa cali-
fornica NPV, Spodoptera complex NPV, and Porthetria dispar NPV.
125
-------
Twenty-five percent of the responses for "best," "second best,"
or "third best" chance for commercialization included one of the follow-
ing tree insect viruses: Porthetria dispar NPV, Laspeyresia pomonella
GV, Neodiprion complex NPV, Orgyia pseudotsugata NPV,* or Choristoneura
fumiferana NPV, GV, CPV.
As shown in Table 19, respondents expected the gypsy moth
virus (Porthetria dispar NPV) to be purchased and used by the govern-
ment, but the majority of sales of the other four principal viral pesti-
cides were expected to be primarily to farmers—either directly or through
pest management professionals. The channeling of sales through pest
management professionals implies an increasing use of this type of service
and a required degree of sophistication on the part of the user to effec-
tively apply viral pesticides.
Table 20 summarizes panel responses for the five viruses that
appeared most frequently in the survey results. Most of the respondents
who cited particular viruses expected Heliothis spp. NPV and Porthetria
dispar NPV to be available on a commercial basis by 1980. The majority
of the respondents did not expect the other three viruses—Autographa
californica NPV, Plusiinae viruses, and Spodoptera complex NPV—to be
commercially available until 1985 or later.
The table's market averages should be viewed with caution be-
cause they were estimated on the basis of very limited data. The range
of answers was large, and the number of responses in some instances was
small.
Table 21 presents additional detail on the R&D data presented
in Table 20. Combined respondent estimates of government and private
industry R&D costs for commercializing a viral pesticide (through first
full registration) are similar to current costs to register a chemical
pesticide. A major difference is that respondents expected a significant
share of the R&D expense to be borne by government. This has signifi-
cant implications on the attractiveness of an overall viral pesticide
venture.
However, as with the market average data, the R&D figures are
respondents' educated estimates. The number of responses in some areas
was very small, and there was a wide range in expected R&D expense; this
reflects the uncertainty involved in some of the technical, economic,
and political issues which confront viruses as pesticides.
Respondents expected the pesticide industry to bear the major-
ity of R&D expenditures for Heliothis spp. NPV, Plusiinae viruses, and
Spodoptera complex NPV. For the gypsy moth virus (Porthetria dispar NPV),
*Already commercial (i.e., first registration'received) as of the pub-
lication date of this report.
126
-------
the government was expected to bear the R&D costs, while for Autographa
callfornica NPV the panelists expected the chemical industry and govern-
ment to share about equally in R&D costs. This latter virus was the
only one where the agriculture industry was expected to contribute sig-
nificant funds for development, although its contribution was estimated
at less than ten percent.
Table 19
VIRAL PESTICIDE SALES
ESTIMATED BY SURVEY RESPONDENTS
(1980, 1985)
Sales
(percent)
1980 1985
Virus
Gr
PMP
Go
I
H
Gr
PMP
Go
I
H
Heliothis sdo. NPV
52%
42%
6%
0%
0%
51%
41%
7%
0%
1%
Autographa californica NPV
48
44
6
1
1
42
45
7
1
5
Porthetria dispar NPV
0
4
88
3
5
6
6
82
4
2
Plusiinae viruses
80
15
0
0
5
61
32
1
3
3
Spodoptera complex NPV
0
0
0
0
0
52
28
15
2
3
Gr = growers; PMP ¦» pest management professionals; Go = government;
I = industry; H ¦ home.
3. General Problems for Viral Pesticides
Considerable information was collected from respondents con-
cerning the technical, economic, and institutional factors they felt
would be most critical or limiting to the commercialization of viral
pesticides. Respondents considered two factors, product specificity
and relative cost, to be most limiting. Product specificity relates to
market size, which was also one of the most critical factors in the
financial analysis of viral pesticides. (Ironically, as described
earlier, while target pest specificity is desirable ecologically, it
can also be economically limiting. Viral pesticides, such as Autographa
californica NPV, which cross-infect to several species of economically
important pests, tend to overcome some of the limitations of product
specificity, but cross-infection may also be a disadvantage because of
possible infection to other organisms.)
127
-------
Cost to the user of viral pesticides relative to traditional
chemical pesticides (broad-spectrum) was viewed as a significant limiting
factor. It is difficult to assess how respondents defined relative cost
(price), but unquestionably production costs and narrow spectrum, as
well as market size, are significant.
Technical production problems were considered a relatively
major limiting factor, barring a breakthrough in tissue culture tech-
niques. Currently viruses must be produced in vivo, generally an expen-
sive technique. Most respondents agreed that the technical problems
associated with viral pesticides, including production efficiency,
could be overcome if the potential economic returns justified committing
large amounts of capital to develop the needed technology. However,
industry may not allocate such commitments because of the lack of patent
protection, relatively small market sizes, and other disincentives.
Respondents generally felt that the sophistication required
for use was not a major problem, and that farmer acceptance was a matter
of cost effectiveness. None of the respondents expected the general
public to exert a significant pressure against the use of viral agents
as pesticides.
Two other factors were mentioned by a few respondents which,
when examined in follow-up interviews, appeared to be important: char-
acterization of viruses and uniformity of product. These factors are
related and appear to be initial requirements for commercializing viral
agents. Virus identification and classification is essential to product
uniformity, but the science of virus identification is not yet well
defined and systematic. A sound system for identifying and classifying
viral complexes appears essential to assure product uniformity, which is
a necessary condition for commercialization.
4. Role of Private Industry by Type of Company
Respondents were asked to evaluate the role different types
of companies might be expected to play in the development of the viral
pesticides.
Small Independent Companies—These firms cannot be expected to
underwrite the original R&D necessary for market*introduction of viral
pesticides because the costs are prohibitively high relative to the
payout, as discussed in I-D, "Risk as a Disincentive." Small independents
may, however, produce limited orders on a contract basis, primarily for
government use on public lands or in public projects. They may also
produce for local markets, if they do not have to bear the majority of
the R&D and registration costs.
Because many markets for viral pesticides are small several
respondents felt that small independent companies could only produce
128
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
viral pesticides with some form of government subsidy. Since viral
products are not currently patentable, such a subsidy might act as an
extra incentive for small companies only to produce on a contract basis.
Medium-Sized Specialized Firms—Respondents expected that these
companies would be most likely to perform the private developmental work
on the first viruses that are commercialized. They have more capital
to risk than smaller companies and might realize some interesting oppor-
tunities if they could introduce a product into the marketplace.
Major Pesticide Firms—These companies have adequate capital
for the R&D required for viral pesticides, but most respondents felt
that they would adopt a "wait and see" attitude. Many of these companies
do not have in-house microbiological specialists, but would undoubtedly
set up departments to develop viral pesticides if market conditions made
the opportunity financially attractive and practical. A breakthrough in
production technology (for example, in the tissue culture area) could
also provide development incentive. Other considerations might include
(1) maintaining their shares of specific crop protection markets,
especially if resistance of certain target species to traditional prod-
ucts increases, (2) complementing an existing product line or offering
a well-rounded integrated pest management package, or (3) maintaining a
public image of ecological concern.
5. Commercial Advantages of Selected Viruses
After respondents selected the three viruses they felt had the
best chances for commercial development, they were asked why they thought
these materials had a competitive advantage or a chance for early com-
mercialization. Respondents generally gave one set of answers for their
three selected viruses. Their reasons fell largely in the five following
areas:
• Sunk R&D costs: Both government and private industry have
spent significant capital researching viruses for pesti-
cides. Some respondents felt ttfat certain governmental
agencies will promote registration of forest insect viruses
to justify past development expenditures. This could assist
commercial companies developing viral pesticides by speeding
up the establishment of a viral registration process in EPA.
Also, most of the viruses selected have an R&D history and a
considerable head start toward commercialization.
• Ecological concerns: Both the private and public sectors
of the economy have become more aware of the environment
and are increasingly conscious of their related responsibil-
ities. This has had or could have results which would aid
the commercial development of viral pesticides:
133
-------
- Government subsidies and R&D for viral pesticides.
- Increased awareness and use of integrated pest manage-
ment systems.
- Increased area-wide management.
- Changes in registration or patent laws.
Lack of suitable chemical pesticides: Some chemical
pesticides that are used against pests which are also
potential targets for viral pesticides may become either
unavailable or unsuitable. This could stimulate the devel-
opment of both viral pesticides and new chemical insecticides.
Existing chemical products could become unavailable or
unsuitable for one or more of the following reasons:
- Insect resistance.
- Deregistration.
- Price increases caused by rising production costs or
short supply. Some respondents specifically predicted
an eventual price advantage for viral pesticides, but
their reasoning appeared imprecise.
- Social and political pressures restricting release of
chemical pesticides into the environment.
Market characteristics: Many respondents felt that the
viruses which would be developed first would be for:
- Nonfood crops.
- Crops which use a large number of treatments of chemical
pesticides annually—either having a large number of
acres to be treated or requiring many treatments per acre
each growing season.
- Target pests with specific physical exposure potential
and feeding habits. Chewing insects iri particular are
the primary pests susceptible to the viral pesticides
currently being tested; no viruses to date are being
tested which affect sucking insects.
Characteristics of particular viruses: Respondents con-
sidered specific viruses with certain general character-
istics at an advantage for commercialization.
- The viruses in the Baculovirus genus (NPV and GV) are
particularly virulent and can be produced relatively
easily.
- Viruses (other than those used in government programs)
with the broadest spectrum of activity have an obvious
market advantage. (Some respondents presented the op-
posite case, preferring viruses which are highly species
specific because of their implied safety.)
134
-------
- Certain viruses are extremely virulent, particularly
Neodiprion spp. NPV, which can be used in relatively
small doses.
D. Decision Analysis of a Possible Viral Pesticide Venture
1. Introduction
The panel survey revealed a fair level of uncertainty about
the role that viruses will play as pesticide agents over the next ten
years. Nevertheless, government and industry are making decisions af-
fecting the future of viruses as pesticide agents. This section examines
the assessment of commercial feasibility for one viral agent that a firm
might make, based on current levels of information.
The nuclear polyhedrosis virus (NPV) for Heliothis spp. was
selected as the hypothetical agent because there is a reasonable amount
of available data on it in the literature and from other sources acces-
sible to SRI. Additionally, panel respbndents indicated that this virus
might have the best chance for commercialization.
The analysis examined the decisions that a hypothetical company
must make as it considers whether to develop Heliothis spp. NPV as a
commercial pesticide. The effect that a Heliothis spp. NPV product might
have on the firm's other lines of business was ignored, and it was assumed
that the company is considering the venture solely on its own apparent
merits.
2. Background on the Product
Review
From an economic loss and control cost viewpoint, the bollworm
complex (the cotton bollworm or corn earworm, Heliothis zea, and the
tobacco budworm, Heliothis virescens) is the second most important
economic "pest" on cotton (after the boll weevil) and also causes serious
damage to other crops such as corn, tomatoes, beans, alfalfa, vetch, and
many garden plants and flowers. One potential method of control for
these insects is the controlled use of the Heliothis host-specific nuclear
polyhedrosis virus complex. Considerable work on the commercial potential
for this and other viral agents as effective pest control tools continues
to be conducted by the USDA, universities, and other organizations. In
addition, Nutrilite Products Inc. and Sandoz Inc. are engaged in the
development of commercial pesticides based on Heliothis spp. NPV. Some
of the characteristics of the products are:
• Stability: Stable at temperatures less than 32°C for
one to two months; this must be increased to three to
six months to obtain a "permanent" label. Degraded by
sunlight (half-life <24 hours); half-life increased to
135
-------
two to three days when formulated with inhibitor
screen. Shelf-life of one to two years if stored
frozen.
• Formulations: Wettable powder or dust, typically
50 larval equivalents (LE) per pound. May have
ultraviolet screen added. (Note: one LE = 6xl09
polyhedral inclusion bodies, PIB. A PIB consists
of several virus particles embedded within a protein
sheath.)
• Toxicity, Mutagenicity, Teratogenicity, Carcino-
genicity : None known. To date, Heliothis spp. NPV
has been shown to be highly host-specific and non-
infective in other vertebrate and invertebrate
species examined.
• Metabolic Information: Heliothis spp. NPV is non-
infective following ingestion by mammals. It does
retain activity after passage through the digestive
system of birds and it may possibly be spread in this
manner.
• Environmental Information: Heliothis spp. NPV is a
naturally-occurring virus. Application of commercially-
developed products is usually in small amounts (40-
100 LE/acre). Half-life in open environments, par-
ticularly in sunlight, is less than 24 hours; however,
if applied to soils where it may receive some pro-
tection from radiation, it may persist for longer
periods of time. While persistence of Heliothis spp.
NPV in the .soil has not been investigated, other NPV
have been shown to persist for long periods in the
soil (up to five years).
History of Development
In 1958, Coaker attempted to use Heliothis spp. NPV as a
control for Heliothis armigera. In 1965 Ignoffo et al. compared the
effectiveness of the virus alone and with several chemical insecticides
for H. zea and H. virescens control, and Brazzel et al. used it in a
biological control program on cotton in the Mississippi Delta. Commercial
development of these products was initiated by Nutrilite Products Inc.
(distribution through Thompson-Hayward Chemical Co.), and by the Inter-
national Mineral's & Chemicals Corp. IMC eventually sold its pesticide
development operations (including Bacillus thuringiensis) to Sandoz Inc.,
a subsidiary of Sandoz AG, Switzerland, in the early 1970s. Sandoz has
applied for two patents in the production and formulation area (as
naturally-occurring materials, viruses are not patentable on a composi-
tion of matter basis). Both Nutrilite and Sandoz have received temporary
permits for use of their Heliothis spp. NPV products on cotton; both have
completed toxicity testing and need to compile efficiency data before
136
-------
applying for permanent labels on this crop. Most of the test work under
temporary label is expected to be completed by Sandoz in 1975.* It is
estimated that the total output of Heliothis spp. NPV products in 1974
was on the order of two to four thousand pounds, with a gross value of
$10,000 to $20,000. Each commercial product may have already accrued at
least $5 million worth of developmental costs.
Production
Viruses are obligate parasites and to date have been suc-
cessfully (in a commercial sense) produced in cells of the host organism.
Output of Heliothis spp. NPV in quantities for field testing is being
produced in vivo using larvae of the host insect(s). As these larvae
die, the infected cadavers are collected and the PIBs are extracted,
purified, and assayed for potency. In vivo manufacture requires the
mass rearing and controlled infection of the host species with the con-
committant problems of contamination and loss from unwanted pathogens
such as bacteria, fungi, and other viruses. In addition, bacterial
counts and 21-day mouse feeding studies and bioassays are needed for each
batch of Heliothis spp. NPV produced. If viruses are to be developed to
their full potential, better in vivo production processes will have to
be developed that offer better control, improved yield, and reduced loss.
Although attempts have been made at developing alternative in vitro pro-
cesses that might be suitable for larger-scale output, no viable com-
mercial processes are known to have been identified to date.
Markets
The future for commercial products will undoubtedly depend
on their success in cotton. This section, then, focuses on the outlook
for commercial acceptance and use on this crop. The following summarizes
the characteristics of bollworms and their infestations of this crop.
The term "bollworm" is often used generically to refer to
two very similar species of insects, Heliothis zea and Heliothis virescens
H. zea is really the true cotton bollworm (also known as the corn ear-worm
or the tomato fruitworm) and it attacks cotton, corn, tomatoes, tobacco,
beans, vetch, alfalfa, and many garden plants and flowers. H. virescens,
the tobacco budworm, causes similar damage and is controlled in a similar
manner. Both species together are called the "bollworm complex", and
their larval stages have no external differences that can be detected by
the naked eye. H. zea moths have light-brown wings mottled with dark-
brown spots; H. virescens moths have light-brown wings with diagonal
-------
olive bands. The life cycle of bollworms, which takes about 27 days to
complete, is illustrated below.
ADULT MOTHS
PUPAE
(nonfeeding)
10-12 days
LARVAE
(light green,
pink, brown,
and/or black)
Feeds average of
16 days; enters top
two inches of soil
to pupate.
EGGS Eggs oviposited
^ singly on cotton
3-5 days (preferably ter-
minals) ; average
1,000 eggs per
female oviposited
over 8-day period.
Bollworms overwinter as pupae in the top two to four inches of the soil in
or around cotton fields and other areas where alternate hosts occur. First-
generation bollworms utilize host plants such as clover, vetch, corn, and
tobacco. Some second-generation bollworms begin to infest cotton, but
economic levels of infestation on this crop usually occur with the third
generation when alternate hosts are no longer available. (Because first-
and second-generation larvae are usually suppressed by natural enemies,
early cotton crops are recommended.) Deep plowing and cultivating the
soil in the fall and winter are effective in exposing the pupae to natural
enemies and weather.
When worm infestation of cotton occurs, the moths feed on
plant exudates (i.e., nectaries) and females oviposit in the evening
hours. Terminal buds appear to be the preferred sites for oviposition
in cotton but lateral buds, square bracts, bolls, and bloom "tags" are
often used late in the season.
The bollworm and tobacco budworm larvae feed on the cotton
fruit and the principal damage is the loss of squares and bolls. Newly-
hatched larvae feed on lead buds and very small squares in the terminals,
later attacking the larger squares and making an entrance hole into each.
The injured square flares and drops from the plant. The large larvae
feed on squares, bolls, and on pollen in open flowers. They enter the
boll and sometimes consume the entire contents. When severe infestations
occur, the larvae may "top" the plants by devouring the tender terminals.
Generally, control measures for bollworms are not required
until mid- and late season. In many areas, natural enemies generally
keep populations in check until these predator populations become reduced
by the insecticides used to control other early-season pests, particularly
138
-------
boll weevils and the plant bug complex (Lygus spp.). When needed, in-
secticide treatment for bollworms begins when eggs (dome shaped, pin-
head size, white) are found and there are four or more worms per 100
terminals. Relatively effective control will result from applications
on small worms (less than one-half inch in length), but increased resis-
tance to presently-used insecticides in many instances has reduced the
effectiveness of these treatments. Repeat applications follow at four-
to-five-day intervals as needed. The principal materials used against
bollworms are toxaphene, carbaryl, methyl parathion, EPN, and azodrin.
Chlordimeform, with excellent ovicidal characteristics, is a relatively-
new insecticide with reported high efficiency in use against the boll-
worm.* Several of the above insecticides are often used in combinations
such as "Triple Kill", which is toxaphene, methyl parathion, and EPN.
Bollworms cause an estimated 4 percent annual loss in the
cotton crop; in 1973, this represented a loss of approximately $112 mil-
lion. However, many researchers feel that this figure is low because of
the increasing resistance of bollworms to conventional insecticides.
In any case, bollworms are definitely the second largest economic pest
on cotton (boll weevils are first) in terms of both yield loss and con-
trol costs. Because bollworms attack several other crops besides cotton,
they are probably the second most important set of economic insect pests
in U.S. agriculture with respect to the pesticide quantities required
for control.
Bollworms are a problem in all areas of the Cotton Belt,
except in the High Plains where they are a reoccurring threat. The upper
limit on acreage treated for these pests may be about 6.5 million acres.
However, probably no more than 50 percent of this acreage is treated for
bollworms specifically, i.e., the differentiation between treatments
for the boll weevil, spider mites, lygus, and other pests and specific
treatments for bollworms are usually impossible. Clear delineation of
the primary target insect is therefore not possible, but most growers
would probably claim that they are treating in mid- or late-season for the
worm complex with other pests controlled incidentally. The worm complex
often requires higher rates of specific pesticides for its control than
does the boll weevil and other pests.
Activity
Generally, viruses are host-specific, microbial agents that,
upon ingestion by the larval form of the host insect, cause a pathologi-
cal condition resulting in death of the larva. Typically, an incubation
period after ingestion is required before effects are observed; with
*Directly prior to the publication of this report, chlordimeform was vol-
untarily taken off the market for further toxicology testing.
139
-------
Heliothis spp. NPV, a two-to-fourteen-day lag occurs between application
(infection) and reduction of the target species to "acceptable" levels.
Reduction in larval feeding activity (i.e., damage to the plant) starts
slowly and builds as the infection spreads. It is necessary to apply
Heliothis spp. NPV in the early stages of infestation for a particular
generation of insects, i.e., before the larvae reach the fourth instar.
Fourth-instar larvae will survive to adulthood and be resistant to the
virus.
Standard spray application techniques and equipment, both
ground and aerial, can be used to apply Heliothis spp. NPV to cotton
(and other field crops). The recommended rate of application is one to
two pounds per acre (40 to 100 LE per acre) with applications repeated
every five to seven days. Because the virus is not very stable, handling
is critical; mishandling during storage (e.g., stored where temperatures
may reach 32° to 38°C) has resulted in some field trials conducted with
material at half-strength or less.
3. Base Case Analysis
Market Size—Heliothis' spp. is the only known insect group
affected by Heliothis spp. NPV. Thus, the maximum potential market size
for the virus is determined by the crops that are presently susceptible
to economic levels of Heliothis spp. damage. These crops are listed in
Table 22.
The table presents approximate domestic acreages for each crop
and the estimated share infested with Heliothis spp. at or beyond the
economic threshold. It was assumed that approximately five applications
of the viral pesticide would be required per season. The maximum poten-
tial market size is 27.8 million acre-treatments per year. Cotton is
the single largest component of that potential market, accounting for
more than 80 percent of potential sales.
However, even if Heliothis spp. NPV is fully registered, it
is unlikely to capture 100 percent of its potential market. The market
share ultimately achieved will depend on how it compares with other
registered products for control of Heliothis spp. The following impor-
tant perceived advantages and disadvantages of Heliothis spp. NPV
relative to toxaphene and organphosphates, the current pesticides most
frequently used to control Heliothis spp. , were derived from the panel
survey and personal interviews. (A number of other factors affect user
acceptance of pesticides, but they are generally the same for Heliothis
spp. NPV and its competitors. They include compatibility with other
pesticides, technical competence required of the user, duration of con-
trol, and price.)
Advantages Disadvantages
Pest specific Unstable
Nonresistant Slow acting
140
-------
Table 22
BASE CASE MAXIMUM POTENTIAL MARKET SIZE FOR HELIOTHIS SPP. NPV
Crop
Cotton
Total acres
(thousands)
12,000
Total
Treatable
Acres
(thousands)
4,500
Maximum Potential Market
Size at 5 Treatments Estimated
per Season Market
(thousand acre- Share
treatments per year) (percent)
22,500 40%
Estimated
Market Size
(thousand acre-
treatments per year)
9,000
Sweet corn
Tobacco
Tomatoes
600
1,000
450
600
350
100
3,000
1,050
500
20
20
20
600
210
100
Total
14,050
5,550
27,750
9,910
-------
Low residues Mode of action
Limited distribution system
From this list of advantages and disadvantages and from panel
members' assessments of market share, nominal estimates of the fraction
of acres that might ultimately be treated with a Heliothis spp. NPV pesti-
cide were made. These appear in Table 22 and constitute the base case
market share estimates for domestic Heliothis spp. NPV sales. The
maximum potential market size is therefore reduced from 27.8 million acre-
treatments per year to almost 10.0 million acre-treatments per year.
However, as indicated earlier, it will be difficult to obtain
meaningful patent protection for a Heliothis spp. NPV product under
current legislation. Thus, if there is a large market, other firms may
also begin producing the virus. In view of this competition potential,
the base case assumed that the hypothetical company will capture 50 per-
cent of the Heliothis spp. NPV market.
The maximum market size for the company has now been reduced
to five million acre-treatments per year. Since it is not likely that
this sales level will be reached during the first year of production,
or that it will continue at that maximum level indefinitely, a very
simple and perhaps ideal growth model has been assumed. Sales were in-
creased linearly over a seven year period to a maximum of five million
acre-treatments per year, held at that level for 14 years, and then
decreased linearly to zero over a ten year period assuming the evolution
of new, more effective controls.
For the base case, it was assumed that Heliothis spp. NPV would
be priced at 60 to 70 percent of the price of existing controls since the
latter control a widex variety of pests. The price of Heliothis spp.
NPV was set at $15 per acre per season, or $3 per acre per treatment
(assuming an application rate of five times per season).
The company making the commercialization decision will probably
not be the distributor; therefore the selling price was adjusted to re-
flect the price that the producer's customer (the distributor) pays.
Producers of similar products typically give the distributor a 40 percent
discount; therefore the price that the company receives was set at $1.80
per acre-treatment.
Production and Capital Costs—The base case assumes that
Heliothis spp. NPV is produced in vivo. While cannibalism among Helio-
this zea has been a major problem in mass rearing, necessitating that
the larvae be raised in individual containers, it was assumed that the
cannibalism problem will be overcome and that larvae can be produced at
a cost of lc per larva. It is expected that the equivalent of 60 larvae*
*See note a. on p. 147.
142
-------
will be needed to produce enough virus to treat one acre. Therefore,
the production cost through the raising of the larvae is equivalent to
about 60c per acre-treatment.*
Stability and shelf life have been two major problems limiting
the development of Heliothis spp. NPV. Efforts have been directed toward
these problems using sophisticated handling and packaging techniques, and
the base case assumed that they will be overcome at an additional cost,
equivalent to 60c per acre-treatment, for bioassay, formulation, pack-
aging, and handling. Total manufacturing costs were therefore estimated
at $1.20 per acre-treatment. Comparing these costs with the FOB plant
price of $1.80 per acre-treatment gives a margin of 60c per acre-
treatment . *
Capital costs for a plant capable of serving the estimated
five million acre-treatments per year market were assumed to be $6 mil-
lion.* This is less than the capital costs for a chemical plant serving
the same market size, because less sophisticated hardware is needed to
raise larvae. It was assumed that the plant capital would be spent over
a three year period and that there would be sufficient capacity in the
first year of full registration to meet the demands of that year. Working
capital requirements were assumed to equal two months of production costs.
Research and Development Costs—The panel survey indicated a
range of $1.4 to $10.1 million would be required for private R&D ex-
penditures for Heliothis spp. NPV; the average was $4.9 million. The
base case assumed that the company's research expenditures began in 1970
and that a full registration on cotton would be achieved in 1978 after a
$3 million R&D expenditure.*
Table 23 gives a breakdown of th° R&D expenditures by activity
and year. The isolation and identification activities are fairly minimal
because they will primarily be conducted by government and academia. It
was assumed (1) that three years of small scale field tests and two years
of large scale field tests will be conducted by the company, (2) that
toxicology expenditures would be the same as those for chemical pesticides,
and (3) that intensive formulation and process development efforts would
vary from two man-years per year in 1974 to five man-years per year in
1976 and to two man-years per year by 1978.
Cash Flow—As described earlier, the base case assumed that
present technical problems would be overcome; full registration would be
achieved in 1978; and that an ultimate sales volume of five million acre
treatments per year would be realized by the company. Table 24 gives
*See note a. on p. 147.
143
-------
Table 23
BASE CASE RESEARCH AND DEVELOPMENT COSTS
FOR HELIOTHIS SPP. NFV*
(Thousands of Dollars)
Activity
1970 1971 1972 1973 1974 1975 1976 1977 1978 Total
Isolation and identification 20
20
20
20
80
Field testing
50 65 100 250 250
715
Toxicology
100 175 300 300 300 150 1,325
Formulation and process
development
20 130 100 150 250 100 100 850
Registration
Total
30
30
20 20 40 300 340 550 800 650 280 3,000
See note a. on p. 147.
-------
the pre-tax cash flow that results from this scenario. Constant prices
and production costs were assumed, since inflation will affect them about
equally. The table shows that the venture would generate a $30 million
surplus of dollars received over dollars spent during its lifetime.
To more properly evaluate the project, future dollars must be
discounted to reflect the time value of money. The last two columns
discount the cash flows at eight and fifteen percent, respectively.
The sum of each column, the net present value at the respective discount
rate, can be considered the amount of present money that remains after
all expenditures on the project—including R&D, capital, and ongoing
expenses—have been recovered at the desired rate of return. Thus, given
a minimum acceptable rate of return, a net present value of zero indicates
that the project is just producing the acceptable rate of return.
The table shows that the pre-tax net present value of the proj-
ect is $10.9 million and $1.7 million at the respective discount rates
of eight and fifteen percent. Thus, if these discount rates constitute
acceptable rates of return, the project would be attractive.
4. Sensitivity Studies
The base case shows that Heliothis spp. NPV is an attractive
venture at an eight to fifteen percent discount rate. However, the
calculations were based on several assumptions that are subject to
variation. Therefore, before the hypothetical company would proceed with
the venture, it should examine the effect of changing base case assump-
tions. If a reasonable change in assumptions changed the project's
financial outlook from attractive to unattractive, then the decision
would be examined in greater depth before finalization.
Table 25 summarizes some of the more important sensitivity
studies that were performed on the base case. The table shows that sales
volume is one of the most important factors affecting project profitability.
Whereas the base case net present value at eight percent is $10.9 million
for a volume of five million acre-treatments per year, the net present
value increases to $25.8 million if the sales volume doubles. Decreasing
the market size to two million acre-treatments per year reduces the net
present value to $2.3 million, and reducing the sales volume to one mil-
lion acre-treatments per year decreases the net present value to a loss
oi; $0. 6 million.
Margin is also very important. The base case margin was assumed
to be $0.60 per acre-treatment. Decreasing the margin to $0.30 per acre-
treatment reduces the net present value to $1.4 million. There are
several factors such as price, manufacturing costs, and packaging costs
that might cause the margin to change, but any combination of changes
that lower the assumed margin by $0.30 per acre—treatment changes the
base project from a reasonably attractive venture to a marginally at-
tractive one, ^f all other variables remain constant.
145
-------
Table 24
BASE CASE PRE-TAX CASH FLOW FOR HELIOTHIS SPP. NPV VENTURE
Sales (3 $3.0 Production and Net Present Value
R&D per Acre- Distribution Capital Yearly (1974 dollars)
Year Costs Treatment Costs Costs Profit (§87. (gl5%
1970
- 20.0
20.0
27.2
- 35.0
1971
- 20.0
20.0
25.2
- 30.4
1972
- 40.0
40.0
46.7
- 52.9
1973
-300.0
- 300.0
- 324.0
-345.0
1974
-340.0
- 340.0
- 340.0
-340.0
1975
-550.0
- 400.0
- 950.0
- 879.6
-826.1
1976
-800.0
- 400.0
-1,200.0
-1,028.8
-907.4
1977
-650.0
- 800.0
-1,450.0
-1,151.1
-953.4
1978
-280.0
578.6
- 385.7
- 864.3
- 951.4
- 699.3
-544.0
1979
1,735.7
-1,157.1
-1,328.6
- 750.0
- 510.4
-372.9
1980
2,953.9
-1,969.3
- 935.4
49.3
31.1
21.3
1981
4,239,6
-2,826.4
- 942.9
470.4
274.4
176.8
1982
5,531.8
-3,687.9
- 543.6
1,300.4
702.5
425.1
1983
6,823.9
-4,549.3
- 543.6
1,731.1
866.0
492.1
1984
8,116.1
-5,410.7'
- 543.6
2,161.8
1,001.3
534.4
1985
8,829.6
-5,886.4
- 479.3
2,463.9
1,056.7
529.6
1986
8,964.6
-5,976.4
- 415.0
2,573.2
1,021.9
481.0
1987
9,038.6
-6,025.7
8.2
3,004.6
1,104.8
488.3
1988
9,045.0
-6,030.0
0.7
3,014.3
1,026.2
426.0
1989
9,045.0
-6,030.0
3,015.0
950.5
370.5
1990
9,945.0
-6,030.0
3,015.0
880.0
322.2
1991
9,045.0
-6,030.0
3,015.0
814.9
280.2
1992
9,045.0
-6,030.0
3,015.0
754.5
243.6
1993
9,045.0
-6,030.0
3,015.0
698.6
211.8
1994
9,045.0
-6,030.0
3,015.0
646.9
184.2
1995
9,045.0
-6,030.0
3,015.0
598.9
160.2
1996
9,045.0
-6,030.0
3,015.0
554.6
139.3
1997
9,045.0
-6,030.0
3,015.0
513.5
121.1
1998
9,045.0
-6,030.0
3,015.0
475.5
105.3
1999
8,640.0
-5,760.0
45.0
2,925.0
427.1
88.9
2000
7,830.0
-5,220.0
90.0
2,700.0
365.0
71.3
2001
6,977.2
-4,651.5
94.8
2,420.5
303.0
55.6
2002
6,077.2
-4,051.5
100.0
2,125.7
246.4
42.5
2003
5,172.7
-3,448.5
100.5
1,824.7
195.8
31.7
2004
4,268.2
-2,845.5
100.5
1,523.2
151.4
23.0
2005
3,363.7
-2,242.5
100.5
1,221.7
112.4
16.0
2006
2,459.2
-1,639.5
100.5
920.2
78.4
10.5
2007
1,554.7
-1,036.5
100.5
618.7
48.8
6.1
2008
650.2
- 433.5
100.5
317.2
23.2
2.7
2009
150.7
- 100.5
55.5
105.7
7.2
0.8
2010
56.2
- 37.5
10.5
29.2
1.8
0.2
2011
4.5
3.0
6.2
7.7
0.4
0.0
Total
57,637.5
10,901.5
1,655.5
146
-------
Table 25
SELECTED SINGLE-VARIABLE SENSITIVITY STUDIES
FOR HELIOTHIS SPP. NPV
Scenario
Market Size
10 million acre-treatments/yr
5 million acre-treatments/yr
2 million acre-treatments/yr
1 million acre-treatments/yr
Product Margina
$0.60 per acre-treatment
$0.30 per acre-treatment
$0.10 per acre-treatment
Registration
Obtained
Not obtained
R&D3
$3 million; 1978 registration
$3 million; 1982 registration
$8 million; 1982 registration
Building Capital3
$4 million
$6 million
$8 million
Net Present Value @ 8 Percent
(millions of dollars)
$25.8
10.9 (base case)
2.3
- 0.6
10.9 (base case)
1.4
¦ 5.2
10.9 (base case)
¦ 2.7
10.9 (base case)
6.7
3.4
12.5
10.9 (base case)
9.3
3Just prior to the publication of this study, new data were provided on these variables as
follows:
• Product Margin—$0.95 per acre treatment based on 25 larvae needed to produce
enough virus to treat one acre (40 LE per acre), instead of the assumed 60.
• R&D—$1.8 million
• Building Capital—$1.5-2 million.
All of these values are beyond the range considered in the sensitivity studies and proba-
bilistic analysis in this report, and the case in which they all occur approximates a net
present value of $25 million at eight percent and $4 million at fifteen percent. This
is obviously among the more attractive cases described herein. SRI was not able to verify
the accuracy of these data, but assuming they are correct they are based on proprietary
information not available to SRI, and/or information developed subsequent to the Insti-
tute's collection of basic data for this study in 1974 and early 1975. In either case,
this incidence of alternative data falls within the caveat emphasized in V "Limits of
Interpretation" concerning challenge to specific analyses according to alternative input
values or new data availability. Further, new data availability based on commercial or
semi-commercial experience does not significantly alter a basic element of this report,
which is focus upon alternative investment options by private industry at an evaluative
or developmental stage rather from a fully operational perspective.
147
-------
Failure to obtain registration would obviously make the project
unattractive. In terms of the base case, completing the assumed R&D
effort and then being unable to demonstrate efficacy or safety for reg-
istration purposes would result in a net present value loss of $2.7 mil-
lion.
Changes in the assumed R&D effort could also affect project
profitability. Delaying the registration date to 1982, but keeping R&D
expenditures constant decreases the base case net present value to $6.7
million. Delaying the registration date to 1982 and increasing the
amount that the company must spend on R&D to $8 million decreases the
net present value to $3.4 million.
Varying the building capital from $4 million to $8 million
(compared to the base case at $6 million) changes the net present value
from $12.5 to $9.3 million, respectively.
The preceding are all single variable sensitivity studies.
However, it is clear that the variables could also be jointly varied in
many combinations. The next section considers some joint sensitivities,
but from the preceding brief results it is clear that sales volume, pro-
duct margin, R&D expenditures, and the ability to obtain full registra-
tion are among the most critical factors that will affect the decision
to commercially develop Heliothis spp. NPV.
5. Probabilistic Analysis
The sensitivity studies showed that the project's attractive-
ness depends on the scenario that actually occurs. The next step of the
analysis was to assess the likelihood of the different scenarios in
order to evaluate the potential for different levels of profit and loss.
To simplify the analysis only the variables found in the sensitivity
studies to most critically affect the commercialization decision were con-
sidered.
Figure 17 outlines the basic elements of the decision to com-
mercially develop Heliothis spp. NPV. The company's first decision is
whether to fund the necessary R&D effort. If it decides to undertake
that effort, it faces the uncertainties of when the effort will be com-
pleted and how much it will cost. Furthermore, the company might not
be able to demonstrate the necessary efficacy and safety required for
registration (the research might finally show that the virus forms un-
acceptable mutations or that certain stability problems are unsolvable).
If the company undertakes the R&D effort and obtains registra-
tion, it then faces a second major decision: whether to actually market
the product. That decision will depend on how profitable the venture
appears as a result of the R&D effort. The profitability of the venture
depends directly on the sales volume and product margin. To more clearly
define the scenarios in Figure 17, the sales volume was subdivided into
two factors: the total Heliothis spp. NPV market size and the fraction
of that market that the company captures.
148
-------
Decision to Completion Company R&D SuccMsful Decision to
Fund R&D Date of R&D Costs ($) Registration Market
Total
Market
Size
Product
Margin
Fraction of
Market Captured
by Company
Fund
Market
Medium
Medium
Medium
Do Not Fund
Do Not Market
Medium
FIGURE 17 SCHEMATIC OF DECISION TO COMMERCIALIZE HELIOTHIS SPP. NPV
-------
Figure 17 presents a simplified version of the decisions a
company would make. For example, an actual company may be able to com-
mit its R&D costs in small increments, which would allow project ter-
mination earlier than assumed herein if unfavorable results were found.
Alternatively, the firms might elect to wait and see if another company
can successfully develop the product. However, the simple model in
Figure 17 is adequate for a first order assessment of the likelihood of
commercial feasibility.
To simplify the analysis the uncertain variables in Figure 17
were assigned a small number of values. For example, while the total
Heliothis spp. NPV market may range from zero to 50 million acre-
treatments per year, this analysis was limited to the effects of three
values: 2, 10, or 25 million acre-treatments per year. This keeps the
number of scenarios within manageable limits, while still preserving the
basic uncertainties of the venture. Figures 18 and 19 give the possible
values that were assumed for each variable.
Also given in Figures 18 and 19 are the likelihoods for each
of the variables assuming these possible values. The probability assign-
ments were made by SRI, based on the results of the panel survey and
personal interviews with experts in the field.
Company R&D
R&D Completion Costs Successful
Date (dollars) Registration
Yes
$3 Million
p ¦ 0.5
p - 0.4
No
1978
p « 0.5
p = 0.3
Yes
$6 Million
V
H
O
p = 0.6
No
p » 0.3
Yes
$4 Million
p - 0.7
p - 0.6
No
1982
p - 0.3
p - 0.7
Yes
$8 Million
00
o
s
a
p - 0.4
No
p - 0.2
FIGURE 18 PROBABILITY AND EVENT ASSIGNMENTS ON RESEARCH AND
DEVELOPMENT OUTCOMES AND ATTAINMENT OF REGISTRATION FOR
HELIOTHIS SPP. NPV
150
-------
Total Heliothis spp.
NPV Market Size
(Acre Treatments)
Product Margin Percent of Total Heliothis spp.
(Dollars per Acre NPV Market Captured
Treatment) by Company
40%
$0.60
p = 0.30
p = 0.25
15%
p = 0.70
50%
25 Million
$0.30
p = 0.50
p - 0.15
p = 0.50
20%
p = 0.50
60%
$0.10
p - 0.80
p = 0.25
20%
p = 0.20
50%
$0.60
p = 0.40
p = 0.25
20%
p = 0.60
50%
10 Million
$0.30
p = 0.50
p = 0.60
p » 0.50
25%
p * 0.50
50%
$0.10
p = 0.85
p = 0.25
33%
p - 0.15
100%
$0.60
p » 0.20
p - 0.25
50%
p - 0.80
100%
2 Million
$0.30
p - 0.50
p - 0.2S
p = 0.50
50%
p * 0.50
100%
$0.10
p - 1.00
p - 0.25
FIGURE 19 PROBABILITY AND EVENT ASSIGNMENTS FOR MARKET OUTCOMES FOR
HELIOTHIS SPP. NPV
151
-------
A company performing an actual analysis would make its own
probability assignments based on the information at its disposal. If it
had different information bases it might make different probability
assignments.
It is important to note that there are dependencies in both
the event and probability assignments. For example, if the R&D is com-
pleted in 1978, the assumed cost is either $3 million or $6 million.
However, if R&D is not completed until 1982, the effort will cost either
$4 million or $8 million. Similarly, the possible share of the Heliothis
spp. NPV market that the company might capture depends on the total market
size and product margin.
Some of the probability assignments depend on specific sce-
narios while others are assumed to be independent. For example, the
probability that the margin will be $0.60 per acre treatment is assumed
to be 25 percent, regardless of the market size or R&D expenditure.
Conversely, the chance of registration depends on the length of R&D ef-
fort: there is only a 50-50 chance of registration if R&D terminates
in 1978 at a company cost of $3 million, but if it continues until 1982
with an expenditure of $8 million, there is an 80 percent chance of
registration.
The possible scenarios in Figures 18 and 19 combine to pro-
duce 72 scenarios of the venture's final outcome. In the same manner
that the net present value for the base case was calculated, the net
present value for each of the other 71 scenarios can be determined.
Using probability theory, the likelihood of each scenario can also be
calculated, using the likelihood of the individual events.
Figure 20 defines these scenarios and presents the net present
value and likelihood for each. Whereas the net present value at eight
percent for the base case was found to be $10.9 million, the figure
shows that across the range of scenarios the net present value varies
between a loss of $8.4 million and a profit of $25.8 million.
The figure also shows that no single scenario is extremely
likely. The most-likely event is that the company will spend $4 million
on R&D and discover that, for some reason, it cannot register the product;
however, this scenario has only a 15 percent chance of occurring. If
the product is registered, the most-likely event is that It will be
registered in 1982 at an R&D cost of $4 million with a total Heliothis
spp. NPV market size of ten million acre-treatments per year and a mar-
gin of $0.30 per acre-treatment. This scenario results in a net present
value loss of $0.30 million.
The hypotehtical company's first decision is whether to support
the R&D effort necessary for developing Heliothis spp. NPV. That deci-
sion partly depends on anticipated profitability of the business after
the R&D effort is completed. The market size and margins are assumed
to be independent of the R&D findings. A separate calculation shows
that once registration Is achieved, the production and marketing
152
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
PAGE NOT
AVAILABLE
DIGITALLY
-------
components of the venture will always return an expected positive profit
on the investment in these post-registration activities (exclusive of
R&D costs). Therefore based on present information, if the product is
registered, the company will always elect to enter the market. The
possible profit and loss outcomes for the immediate decision of support-
ing R&D are listed in Figure 20.
Figure 21 summarizes all of these profit and loss possibilities
in a probability distribution. The graph was calculated by ordering the
rounded-off outcome values according to increasing profitability, and
then summing the likelihoods for all of the scenarios with the same
value. The graph repeats the net present value range of the venture
(from a loss of $8.4 million to a profit of $25.8 million) but reveals
that the chances of the $25.8 million outcome are infinitesimal compared
to those for some of the other outcomes. The most likely event is that
the project will lose $4 million to $6 million; there is a 50 percent
probability of this scenario. On the average, the venture would lose
$2.7 million.
Figure 22 gives the same information as Figure 21, in cumula-
tive form. (For ease of presentation the values were rounded off to
the nearest $2 million in Figure 21; they were not in Figure 22.)
Figure 22 gives the probability of the profit equaling or falling below
a specified amount; it shows explicitly that there is an 85 percent
chance of the project failing to make a positive profit at an eight
percent discount rate. If the discount rate were greater then there
would be an even greater chance of losing money. (A separate sensitiv-
ity study revealed an 88 percent chance of losing money at a 15 percent
discount rate.) There is little point in performing further risk
analyses of this venture, because any weighing of the negative outcomes
to reflect the company's aversion to loss would only show a higher level
of financial unattractiveness.
6. Problems To Be Overcome for Successful Commercialization
The preceding analysis, based on current levels of information,
indicates a low level of financial attractiveness for developing
Heliothis spp. NPV. Assuming that biological alternates to chemical
pesticides might be desirable, SRI investigated possible measures that
might make the ventures more attractive
Table 26 summarizes the critical attributes of the 22 scenarios
that resulted in a positive net value at eight percent, and shows the
basic events that must occur for the project to be attractive. Market
size and margin are very critical. In 14 of the 22 profitable scenarios,
the company's market size was five million or more acre-treatments per
year. In the four profitable cases where the market size was less than
four million acre-treatments per year, the margin was at the maximum
($0.60 per acre-treatment), and the R&D costs were at or near the minimum.
In none of the profitable cases was the margin as low as $0.10 per
157
-------
0.25
In
00
0.20
0.15
m
<
m
O
BC
a.
0.10
0.05
M
&
fl
jim.
-10
-2 0 2 4 6 8 10
NET PRESENT VALUE AT 8%
12 14 16 18
- millions of dollars
20 22 24 26
28
FIGURE 21 PROBABILITY DISTRIBUTION ON NET PRESENT VALUE, AT EIGHT PERCENT, OF HELIOTHIS SPP. NPV
VENTURE
-------
Table 26
CRITICAL ATTRIBUTES OF PROFITABLE HELIOTHIS SPP. NPV SCENARIOS
Scenario
Effective
Market Size
(millions of
acre-treatments)
Margin
(cents per
acre-treatment)
Present Value
of R&D at 8%
(millions of dollars)
1
2
3
4
7
10
4
13
5
5
60
60
30
30
60
$2.7
2.7
2.7
2.7
2.7
8
9
13
18
19
2
5
2
10
4
60
30
60
60
60
2.7
2.7
2.7
5.2
5.2
20
24
35
36
37
13
5
10
4
13
30
60
60
60
30
5.2
5.2
3.2
3.7
3.2
41
42
47
52
53
5
2
2
10
4
60
60
60
60
60
3.2
3.2
3.2
6.3
6.3
54
58
13
5
30
60
6.3
6.3
159
-------
Ill
-4-2 0 2 4 6 8 10 12 14 16 18
NET PRESENT VALUE AT 8% — millions of dollars
FIGURE 22 CUMULATIVE PROBABILITY DISTRIBUTION ON NET PRESENT VALUE. AT
EIGHT PERCENT, OF HELIOTHIS SPP. NPV VENTURE
acre-treatment, and in more than three-fifths of the cases the margin
was $0.60. In 14 of the 22 successful cases, R&D expenditures approxi-
mated $3 million; if those expenses exceeded $5 million, the project was
only attractive if the margin and market size were large.
The observation that a large market and margin and small R&D
costs are essential attributes of profitability are not unique to
Heliothis spp. NPV; they are conditions of almost any successful venture.
However, for Heliothis spp. NPV there is small likelihood that these
conditions will be realized, based on current levels of information.
The nomirjal company market size of five million acre-treatments
per year assumes that Heliothis spp. NPV will be used on 40 percent of
infested domestic cotton acres and that the company will capture 50 per-
cent of the market. If a meaningful proprietary position could be ob-
tained, the company would double its market size. Alternatively, the
firm could increase its market if the product were used on a larger
percentage of infested acres. (Very few sources expected Heliothis spp.
NPV to capture more than 50 percent of the infested cotton acres.)
About the only way this could happen would be if all other bollworm con-
trols were deregistered or become ineffective, and if no replacement
products were developed. Another way of increasing or ensuring the market
size, of course, would be through compulsory programs requiring the use
of Heliothis spp. NPV (e.g., through pest control districts or area-
wide management).
160
-------
Product margin is largely a function of production and handling
costs and the price that the market will bear. Production costs depend
on the process the company elects to use, but the government could
conceivable subsidize the development of more efficient production pro-
cesses; alternatively, public agencies might help provide cooling facil-
ities for storing the pesticide, since it appears that special facili-
ties may be required to keep the product at moderate temperatures.
The market price also depends on the relative price and ef-
ficacy of competitive controls. The only means to improve efficacy is
through additional research, which could be supported by government
funds. As long as registered competitive products exist, there is little
likelihood that Heliothls spp. NPV could command a significantly higher
price than that of these materials. Thus, if the production costs are
near the price that the market will bear, the only encouragement
government can offer appears to be price supports to maintain attractive
margins.
If government completely subsidized R&D efforts, more of the
scenarios would be profitable. Separate sensitivity analyses that
assumed the company's R&D costs were zero showed that the venture would
give an expected net present value of $1.8 million at an eight percent
discount rate.
Thus, if the product were fully developed by an outside agency,
the venture would be moderately attractive on an expected value basis.
However, as soon as the expected present value of R&D funds is added the
project becomes unattractive.
E. Survey Questionnaire
Presented in Figure 23 on the following pages is a copy of the
questionnaire sent to the panel of experts in the viral pesticide
survey.
161
-------
VIRAL PESTICIDES - CRITICAL FACTORS
IN SUCCESSFUL DEVELOPMENT
March 1975
The following viral complexes are among those being explored by researchers as potential pesticide products:
Agrotis, Peridroma, and related cutworms (NPV)
Argyro taenia velutinana (GV) (red-banded leaf roller)
Autograph* California virus (NPV)
Chilo suppressalit (GV)
Ephestia cautella (NPV + GV)
Heliothis complex viruses (NPV), e.g., armigera,
zea, virescens
Laspeyresia pomonalla (GV) (codling moth)
Malacosoma complex (NPV)
Mamestra brassicae (NPV)
Neodiprion complex (NPV)
Phthorimaea operculella (GV) (potato tuberworm)
Pieris (Mamestra) spp. (GV) (crucifer caterpillars)
Plusiinae viruses (NPV) e.g. Trichoplusia, Piusia,
and Pseudoplusia
Porthetria dispar (NPV)
Spodoptera complex viruses (NPV), e.g. littoralit.
exigua, frugiperda, litura, exempta
Which three of these (or others not listed) do you see as having the greatest ^commercial potential? *
1 (has best chance)
2 (has second best chance)
3. (has third best chance)
Do you think any others have reasonable short- to medium-term commercial potential? (Please list) _
For the viral complexes that DO NOT have this level of commercial potential (within the next 10 years, at
least), which ,of the following factors do you feel are the most limiting?
Factor Remarks
Product specificity, market size
Production problems
Storage problems
iconiinifd)
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES
162
-------
B. (continued}
Factor
Remarks
Distribution problems
Field-life problems
Sophistication required for use
Other problems relating to efficacy
Cost versus other pesticides
Absence of patent protection
Image in farmer's view
Image in public's view
Other (please explain)
C. In your view, why are these problems unlikely to be overcome within the next 10 years?
D. For each of the first three products listed on the preceding page that you identified as HAVING the greatest
commercial potential, please complete the following surveys.
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Continued)
163
-------
Product 1. *has '3est chance)
(fill in name from page 1)
The first three questions on this survey may be the most difficult — and there may, in fact, be no scientifically
justifiable basis for answers at this time. However, there certainly must be opinions, however tentative or fragile;
as implied in the cover letter to this survey, it is unlikely that the research being conducted on this and other
viral agents is being done in an absence of optimism about the future of these products as pesticides.
Please try to fill in as much as you can - take a deep look at your knowledge of the product in question, and
"jump." It is emphasized again that individual survey returns will not be separately divulged.
A. While you rated the viral agent identified above as having a high degree of commercial potential, do you
think it will actually become a commercial pesticide {e.g., registered and available for salel:
by 1980? by 1985?
If "no" to both dates, please explain why.
If "yes" to either date, please complete the following questions.
B. On which crop(s) or crop-pest completes) do you think this viral agent will have the greatest initial impact
as a commercial pesticide? Please list in order of anticipated registration.
1 . 3. -
2 4.
C. Please try to make even a "blue-sky" estimate of the percentage share of the national pesticide market* that
might be captured by this product for the crop or crop-pest complexes where you think it will be registered.
Crop or Crop-pest Complexes for Which
Registration Has Been Obtained
Estimated Percentage Share of
National Pesticide Market* Captured by;
1980
1985
1. .
2.
3. .
4.
'Defined at % of total number of vaatrntntt with any products for th* cop trot of a specific pest.
FIGURE 23 SAMPLE QUESTIONNAIRE—VfRAL PESTICIDES (Continued)
164
-------
If this product is successfully introduced by 1980 or 1985, who will be the customers?
Percent of Sales to in:
1980 1985
Growers
Pest management professionals
Government agencies
Industrial and institutional users
Home and garden users
(Remarks)
For the commercialization through first registration of the viral agent discussed here into a pesticide, what type
of R&D and commercial development funding do you "guesstimate" will be spent?
Percent Spent by:
Agriculture
Industry (e.g.,
Chemical co-ops, grower
Industry associations)
Production
Formulation
Storage, distribution
Efficacy, use in field —
Registration
Safety —
Other (specify) ——
TOTAL ——
Total
Expenditures Government,
($ millions) Academia
(Remarks)
T
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Continued)
165
-------
Product 2.
(fill in name from page 1)
(has second best chance)
The first three questions on this survey may be tfie most difficult — and there may, in fact, be no scientifically
justifiable basis for answers at this time. However, there certainly must be opinions, however tentative or fragile;
as implied in the cover letter to this survey, it is unlikely that the research being conducted on this and other
viral agents is being done in an absence of optimism about the future of these products as pesticides.
Please try to fill in as much as you can - take a deep look at your knowledge of the product in question, and
"jump." It is emphasized again that individual survey returns will not be separately divulged.
A. While you rated the viral agent identified above as having a high degree of commercial potential, do you
think it will actually become a commercial pesticide (e.g., registered and available for sale):
by 1980? by 1985?
If "no" to both dates, please explain why.
If "yes" to either date, please complete the following questions.
B. On which crop(s) or crop-pest complex(es) do you think this viral agent will have the greatest initial impact
as a commercial pesticide? Please list in order of anticipated registration.
1.
2.
3.
4.
C. Please try to make even a "blue-sky" estimate of the percentage share of the national pesticide market* that
might be captured by this product for the crop or crop-pest complexes where you think it will be registered.
Crop or Crop-pest Complexes for Which
Registration Has Been Obtained
Estimated Percentage Share of
National Pesticide Market* Captured by:
1980
1985
2.
•Defined at % of total number of treatment* with any products for the control of a tpecific pett.
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Continued)
166
-------
If this product is successfully introduced by 1980 or 1985, who will be the customers?
Percent of Sales to in:
1980 1985~
Growers
Pest management professionals
Government agencies
Industrial and institutional users
Home and garden users
(Remarks)
For the commercialization through first registration of the viral agent discussed here into a pesticide, what type
of R&D and commercial development funding do you "guesstimate" will be spent?
Percent Spent by:
Agriculture
Total Industry (e.g.,
Expenditures Government, Chemical co-ops, grower
($ millions) Academia Industry associations)
Production ____________
Formulation ____________
Storage, distribution
Efficacy, use in field
Registration
Safety
Other (specify) _________—'
TOTAL
(Remarks)
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Continued)
167
-------
Product 3.
(fill In name from page 1)
(has third best chance)
The first three questions on this survey may be the most difficult - and there may, in fact, be no scientifically
justifiable basis for answers at this time. However, there certainly must be opinions, however tentative or fragile;
as implied in the cover letter to this survey, it is unlikely that the research being conducted on this and other
viral agents is being done in an absence of optimism about the future of these products as pesticides.
Please try to fill in as much as you can - take a deep look at your knowledge of the product in question, and
"jump." It is emphasized again that individual survey returns will not be separately divulged.
While you rated the viral agent identified above as having a high degree of commercial potential, do you
think it will actually become a commercial pesticide (e.g., registered and available for sale):
by 1980? by 1985?
If "no" to both dates, please explain why.
If "yes" to either date, please complete the following questions.
B. On which crop(s) or crop-pest complex(es) do you think this viral agent will have the greatest initial impact
as a commercial pesticide? Please list in order of anticipated registration.
3.
4.
C. Please try to make even a "blue-sky" estimate of the percentage share of the national pesticide market* that
might be captured by this product for the crop or crop-pest complexes where you think it will be registered.
Crop or Crop-pest Complexes for Which
Registration Has Been Obtained
Estimated Percentage Share of
National Pesticide Market* Captured by:
1980
1985
3.
4.
'Defined at % of total number of treatments with any product! for the control of a specific pest.
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Continued)
168
-------
If this product is successfully introduced by 1980 or 1985, who will be the customers?
Percent of Sales to in:
1980 1985
Growers
Pest management professionals
Government agencies
Industrial and institutional users
Home and garden users
(Remarks}
For the qommercialization through first registration of the viral agent discussed here into a pesticide, what type
of R&D and commercial development funding do you "guesstimate" will be spent?
Percent Spent by:
Agriculture
Total Industry (e.g.,
Expenditures Government, Chemical co-ops, grower
($ millions) Academia Industry associations)
Production ________
Formulation
Storage, distribution
Efficacy, use in field
Registration __
Safety _______
Other (specify)
TOTAL
(Remarks)
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Continued)
169
-------
The following questions generally apply to the three viral agents that you have just made a rough commercial
analysis of.
A. What role do you feel the following types of private companies are likely to play in the commercial
development of these products? Please explain.
Small independents
(Remarks)
Medium-sized
specialized firms
(Remarks)
Major pesticide firms
(Remarks)
B. On pp. 1-2 of this survey are listed factors that have
commercialized as pesticides. Some others are:
Pest resistance to existing pesticides.
Unavailability of pesticides already deregistered
by EPA (e.g., DDT).
Unavailability of existing pesticides due to tight
supply (limited by capacity, raw materials, etc.).
Future deregistratiorj by EPA of pesticides
currently in use.
Changes in EPA registration requirements for viral
pesticides.
a bearing on whether or not viral agents are
Efficacy versus other pesticides.
Safety of viral agents (e.g., evolutionary or
mutational characteristics).
Production problems (e.g., production of constant
strains due to evolutionary or mutational
characteristics of agent and host).
Evolution of integrated pest management practices.
Changes in patent law affecting the current non-
proprietary characteristics of viral agents.
Using these factors as a reference point, why do you think the three viral agents that you feel have the
greatest commercial potential will succeed where others won't? Are there basic differences within the factors
delineated above — or are there other reasons? Please explain.
(continued)
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Continued)
170
-------
C. If you have any additional comments or summary remarks, please use the space below.
D. Now that you have completed this survey, have you changed your mind about your original estimates of
commercial potential for the three "best chance" viral agents you selected? Do you feel you would like to
go back and adjust some of your estimates to questions A, B, and C on pages 3, 5, and 7?
yes no
If yes, please do so without eliminating youi original estimates.
When you have completed this survey, please return to:
Thomas A. Blue
Manager, Agricultural Chemicals
Chemical Industries Center
Stanford Research Institute
Menlo Park, California 94025
Your assistance is deeply appreciated. When the final survey results are published, we will be glad to send you a
complimentary copy. Please indicate below where it should be sent":
'The contents of individual turytyl will not be divulged but w« do nnd to know who ratpondtd in caw we have additional
qutttioni and in ordtr to di«uibut« «urv«y r«ult«.
FIGURE 23 SAMPLE QUESTIONNAIRE—VIRAL PESTICIDES (Concluded)
171
-------
F. Selected References
The following is a list of some of the more pertinent literature
references relating to viruses as pesticides that were screened during
this report.
Andrews, G. L., et al., "Evaluation of Heliothis Nuclear Polyhedrosis
Virus in a Cottonseed Oil Bait for Control of Heliothis virescens
and H. zea on Cotton," Journal of Economic Entomology, Vol. 68,
pp. 87-90 (1973).
Anonymous, Biotrol VHZ, Technical Bulletin, Nutrilite Products, Inc.
(19 71).
Anonymous, Baculovirus Research Development and Registration Program,
proposed to the Ad Hoc University-EPA-USDA Subcommittee on New
Pest Control Materials, Washington, D.C. (1974).
Allen, G. E., B. G. Gregory, and J. R. Brazell, "Integration of the
Heliothis Nuclear Polyhedrosis Virus into a Biological Control
Program for Cotton," Journal of Economic Entomology, Vol. 59,
pp. 1333-1336 (1966).
Bullock, H. R., J. P. Hollingsworth, A. W. Hartstock, Jr., "Virulence
of Heliothis Nuclear Polyhedrosis Virus Exposed to Monochromatic
Violet Irradiation," Journal of Invertebrate Pathology. Vol. 16,
pp. 419-422 (1970).
Coaker, T. H., "Experiments with a Virus Disease of the Cotton Bollworm,
Heliothis armigera (Hbn)," Annals of Applied Biology. Vol. 46,
pp. 536-541 (1958).
Couch, Terry L., Factors which Influence Industrialization of Insect
Viruses, Abbott Laboratories, North Chicago, Illinois (1974).
David, W. A. L., "The Status of Viruses Pathogenic for Insects and Mites,"
Annual Review of Entomology, Vol. 20, pp. 97-117 (1975).
Djerassi, C., C. Shih-Coleman, and J. Drekman, "Insect Control of the
Future: Operational and Policy Aspects," Science, Vol. 186, pp.
596-607 (1974).
Dulmage, H. T., Responsibility for Producing and Developing Insect
Viruses: Individual User. Commercial, Public, presentation at
Symposium on the Specificity and Development of Insect Viruses for
Pest Control, Vllth Annual Meeting Society for Invertebrate Pathol-
ogy, Tempe, Arizona (June 16-21, 1974).
172
-------
Engler, R. , (ed.), "The Use of Viruses for the Control of Insect Pests
and Disease Vectors," WHO Technical Report Series No. 5 31, Report
of a Joint FAO/WHO meeting, Geneva, Switzerland (November 1972).
Falcon, L. A., Integrated Control of Cotton Pests in the Far West,
proceedings of Beltsville Cotton Production Research Conference.
Falcon, L. A., I. E. Gard, and J. Gallagher, Nuclear Polyhedrosis of
Heliothis: Field Evaluations on the Persistence and Effectiveness
of Improved Formulations in Cotton, report of the Division of
Entomology, University of California, Berkeley, California.
Falcon, L. A., A. Sorensen, and N. B. Akesson, "Aerosol Application of
Insect Pathogens," California Agriculture (1974).
Falcon, L. A., Some Recent Activities Concerning the Development,
Industrialization and Registration of Arthropod Viruses, report of
the Department of Entomological Sciences, University of California,
Berkeley, California (1974).
Fernandez, A. T., et al., "A Field Test Comparing Resistant Varieties
Plus Applications of Polyhedral Virus with Insecticides for Control
Heliothis spp. and Other Pests of Cotton," Journal of Economic
Entomology, Vol. 62, No. 1, pp. 173-177 (1969).
Greer, F,, C. M. Ignoffo, R. F. Anderson, "The First Viral Pesticide:
A Case History, Chem Tech, pp. 342-347 (June 1971).
Hink, W. F., P. V. Vail, "A Plague Assay for Tritration of Alfalfa
Looper Nuclear Polyhedrosis Virus in a Cabbage Looper (TN-368)
Cell Line," Journal of Invertebrate Pathology, Vol. 22, pp. 168-
174 (1973).
Hink, W. F., Propagation Insect Viruses in Tissue Culture, Meeting of
Society of Invertebrate Pathology (Abstract)(1974).
Ignoffo, C. M. , "Development of a Viral Insecticide: Concept to Com-
mercialization," Experimental Parasitology, Vol. 33, No. 2, pp.
380-406 (1973).
Ignoffo, C. M., "The Nuclear Polyhedrosis Virus of Heliothis zea (Boddie)
and Heliothis virescens (Fabricius), IV, Bioassay of Virus Activity,"
Journal of Invertebrate Pathology, Vol. 7, pp. 315-319 (1965).
Ignoffo, C. M., "The Nuclear Polyhedrosis Virus of Heliothis zea (Boddie)
an^ Heliothis virescens (Fabricius), I, Virus Propagation and its
Virulence," Journal of Invertebrate Pathology, Vol. 7, pp. 209-
216 (1965).
Klassen, W., "Alternative Methods of Pest Control," Part I, Agricultural
Programs of Hearings of the Subcommittee on Appropriations, House
of Representatives: 93rd Congress (February 1973).
173
-------
Marx, J. L., "Insect Control (II): Hormones and Viruses," Science, Vol.
181, pp. 833-835 (1973).
Smith, R. F., (ed.), "Considerations on the Safety of Certain Biological
Agents for Arthropod Control," WHO Bulletin 48, pp. 685-698 (1973).
University of California, Pest and Disease Control Program for Cotton,
California Agricultural Experiment Station Extension Service, 1974.
Vail, P. V., D. L. Jay, and W. F. Hink, "Replication and Infectivity of
the Nuclear Polyhedrosis Virus of the Alfalfa Looper Autographica
californica, Produced in Cells Grown in vitro," Journal of Inverte-
brate Pathology, Vol. 22, pp. 231-237 (1973).
Wells, F. E., and A. M. Heimpel, ''Replication of Insect Viruses in
Bacterial Hosts," Journal of Invertebrate Pathology, Vol 16,
pp. 301-304 (1970).
Williams. C. M., "Third-Generation Pesticides," Scientific American.
Vol. 217, p. 13 (1967).
174
-------
VIII PHEROMONES
A. Overview
Pheromones are natural substances emitted by organisms as a means
of communication. Pheromones relevant to pest control relate to mating
and aggregation. A large number of these substances have been isolated,
identified, and synthesized for specific insect pests. While many are
still at the laboratory experimental stage, several pheromones of eco-
nomically important pests are being used in field trials for both popu-
lation survey and suppression purposes. A few are approaching the level
of development required for registration as suppression agents (as
opposed to survey agents). While definitive conclusions on the commer-
cial viability of any specific pheromone system are difficult, SRl's
assessment of the likelihood of suppression programs employing phero-
mones appears in Table 27.
The following characteristics are considered the most significant
in assessing the market potential of pheromones for pest control.
Efficacy—Effectiveness has been demonstrated in laboratory and
limited field trials. Several pheromones are effective for survey use,
and some appear to be useful for the control of low to moderate pest
infestations. However, for high population densities these agents as
controls are usually ineffective when used alone. Technological advances
in pheromone formulation for sustained release at more uniform rates
should improve effectiveness by reducing the number of applications
required per season.
Product Specificity, Market Size, Incentives—Pheromones are rela-
tively specific for given insect species although occasional cross attrac-
tion can occur within a family. As with the viruses, the narrow spectrum
of activity may be environmentally favorable, but the market size for any
particular pheromone product is essentially limited to those crops
affected by single insect pests.
These compounds are natural products, which prevents the assignment
of composition of matter patents. However, patents can be obtained on
the basis of formulations or novelty of synthesis. Limits on patent-
ability and small market size appear to be constraints on the development
of these chemicals by large corporations. A few small organizations are
producing and/or formulating pheromones in moderate quantities (1 to
100 kilogram lots) for currently small survey and test markets. Most of
these firms could not manufacture the quantities required for large scale
use without significant expansion of production facilities.
175
-------
Table 27
RELATIVE LIKELIHOOD OF PEST SUPPRESSION PROGRAMS
USING PHEROMONES
(1980, 1985)
Relative
Likelihood
Pest 1980 1985
Pink bollworm
3
3
Stored product pests
3
3
Western pine beetle
2
3
Boll weevil
1
2
Codling moth
1
2
Leaf roller complex
1
2
Cabbage looper
1
2
Tussock moth
0
1
Gypsy moth
0
1
Spruce budworm
0
1
European corn borer
0
1
Southern pine beetle
0
1
*Relative likelihood: 0 = negligible;
«
1 = low; 2 = medium; 3 = high.
Production, Storage, Field Life Characteristics—Pheromones are
organic compounds of varying degrees of structural complexity, usually
containing ten to twenty carbon atoms. Frequently, active pheromones
are a mixture of two to four compounds with the ratios of individual
components being critical. Synthesis of individual components can be
very simple, with freedom from stereochemical complications, or quite
complex..
Pheromones are invariable liquids or syrups which are stable during
storage at ordinary temperatures; a few require dilution in inert organic
solvents, and/or storage at lower temperatures. Under field conditions,
however, pheromones may be affected by prolonged exposure to sunlight,
air, and high temperatures. Microencapsulated slow release formulations
176
-------
have frequently overcome these problems of storage and field Instability.
Slow release formulations may be affected by the volatility and viscosity
of the pheromone materials.
Toxicity—Pheromones contain only carbon, hydrogen, and oxygen, and
are biodegradable. Acute toxicity has not been observed under field use
conditions, although direct contact with the neat liquids in some cases
can produce irritation of sensitive tissues. Chronic toxicity has not
been determined for most of these compounds; this is compatible with
the nature of their chemical structures.
B. Background
Pest control techniques employing pheromones are among the newer,
relatively innovative methods. They involve the use of chemical agents
that elicit specific behavioral reactions from target organisms, par-
ticularly insects.
Typically, pheromone substances are used by insects in their
chemical communication systems for behavior patterns such as mating,
alarm, oviposition, aggregation, and trail following. These chemical
signals, and their receptor systems, are highly developed in many
insects and other arthropods. Theoretically, the appropriate manip-
ulation of these chemical agents, either in natural or synthetic form,
can be used to modify the behavior of pest organisms in order to achieve
a measure of control. The most commonly investigated chemical commu-
nication systems for pest control are those related to mating, involving
the so-called sex pheromones. Depending on the species of pest, either
the male or female emits a chemical signal that members of the opposite
sex use to locate the signaling or calling members; most control schemes
attempt to modify the behavior of signal-receiving members.
Two basic pest control strategies have evolved in the use of mating
or sex pheromones. The first employs pheromones to monitor or survey
the presence, population density, and location of a pest species.
Individuals are lured into traps where their numbers are recorded to
compile information on the status and dynamics of the population in a
given area. If the level of Infestation exceeds a predetermined level,
appropriate corrective measures (such as chemical pesticide application)
can be taken. Pheromones used in this manner are thus indirectly
involved in the control of a pest.
There' are two variants of the second basic strategy, which is aimed
directly at pest control by suppression. One involves the use of phero-
mone baited traps as described in the survey technique; however, greater
trap densities are employed in an attempt to reduce the pest population
to below-threshold levels. The other variant involves mass application
of a pheromone to a large area in an attempt to prevent aggregation or
to disrupt mating behavior; this assumes that when the surrounding air
space is saturated with a pheromone, the pest organism should have a
177
-------
lower probability of locating a mate because of such phenomena as phero-
mone chemoreceptor fatigue, adaptation, or habituation. This approach
attempts to reduce mating and thus the population density of subsequent
generations. It may also inhibit pests from aggregating in sufficient
numbers to effect significant damage.
Table 28 summarizes tfrese pest control strategies.
The following sections present:
• Profiles of major areas of pheromone research, based on
interviews.
• Decision analysis of a possible pheromone pesticide
venture (for suppression).
C. Individual Product Reviews
Since pheromones are being investigated in a number of specific
areas, SRI substituted direct inverviews with pheromone experts for the
blanket panel questionnaire survey approach used with bacteria and
viruses. With this approach it was appropriate to present findings by
specific markets or pest control areas.
1. Forest Pests
There are two main categories of forest pests—those insects
which affect the cambium and phloem of trees, and those which defoliate.
The major offenders of the former category are bark beetles of the family
Scolytidae, which are the primary cause of tree mortality in the western
and southern United States; associated losses are estimated at ten to
fifteen percent nationally for pine, fir, and spruce softwoods. The
major defoliating insects are the tussock moth, the gypsy moth, and the
spruce budworm. Defoliators primarily cause growth loss, and predispose
tree mortality from bark beetles and other wood-boring Insects. Forest
insects cause losses to the timber industry, and to the public, partic-
ularly in suburban and recreational areas.
Bark Beetles—The most important bark beetles are members of
the genus Dendroctonus. These include the western pine beetle (D. brevi-
comis), the southern pine beetle (D. frontalis), the Douglas fir beetle
(D. pseudotsugae), the mountain pine beetle (D. ponderosae), and the
spruce beetle 0D. rufipennis). A second group of pine bark beetles
belongs to the genus Ips affecting ponderosa and lodgepole pine (_I. pinl
and I_. paraconfusus) and southern pines Q_. grandicollis, _I. avals us,
and 1^. colligraphus) . A third genus of beetles, Scolytus, contains two
pests of economic significance, the Dutch elm disease vector (S^. multi-
striatus) and the white fir engraver (j5. ventrails) . A fourth genus,
causing degrade in lumber, is Trypodendron. The two-lined ambrosia
178
-------
Table 28
USE OF PHEROMONES IH INSECT CONTROL
Strategy
Suppression
Monitor-Survey Mass Trapping Mating/Aggregation Disruption
Basic concept Luring specie members to traps Attracting and trapping large Saturating the environment with
where they may be collected and numbers of specie members to a pheromone, rendering specie
counted, to determine population substantially reduce their members unable to aggregate or
levels and other peat dynamic population. locate potrniial mates, thereby
functions. reducing population levels of
subsequent genera t itins .
Pheromone purity
required
High* in order to compete with High; must compete with natural
natural pheromones In the pheromones as in monitor strat-
environment (a crude extract egy» but the large numbers of
may be sufficient for small sources required must nut lead to
scale use). mating disruption.
Pheromone need not. be of high
purity. Single compound of
multicomponent pheromone may be
sufficient to disrupt communi-
cation between potential mating
insects.
Pheromone
formulat ion
Slow release through containment by materials such as PE capsules,
rubber, sticky coatings, and open glass vials. For multicomponent
pheromones, each component may have to be prepared separately to
achieve combined release rates equivalent to those of insects.
Slow release as in traps, or
microencapsulated for spray
application. Spray technique
may need antioxidant to prevent
photoly tit degrada t ion.
Special equipment
required
Traps, with appropriate size, shape, color, and field placement.
Existing ULV spray equipment,
both ground and aerial, may be
used with little or no modification.
Developmental
stage of pest
for application
Adult. Adult.
Adult.
Specificity
Pheromones are relatively unique to each insect specie. However, closely related species may use the
same components in differing ratios. In trapping, precise formulation is required. In mating disrup-
tion, one component may disrupt several species.
Relationship to
population density
Nonlinear; traps must compete 1:1 with natural pheromone emitting
members of pest population. Trapping efficiency drops as popula-
tion density increases.
At high population densities,
physical proximity of potential
mating insects and other factors
can Interfere with disruption.
Infestation level
appropriate for
treatment
Useful at all levels. Maintenance of low level infes-
tations only.
Suppression of moderate infesta-
tion and maintenance of tow
levels of infestation.
Appropriate
application area
Well defined, small area such as an orchard or warehouse.
All areas.
Required frequency
of application
Traps must be checked peri- One or two applications per
odlcally. season.
One application per season for
some pests; others require appli-
cation every 2-4 weeks.
Latency
Function of follow-up control Reduces adult population and,
method. thus, succeeding generation's.
Reduces only succeeding genera-
tions* but may also immediately,
inhibit some crop damage.
Relationship to Chemicals nay be used to reduce heavy population density prior to pheromone treatment,
chemical pesticides
Chemicals can be used as follow- Chemicals can be used inside No use of chemicals,
up control method at levels indi- trap.
cated by trap census. The amount
of chemical pesticide may be
reduced by efficient timing of
operation.
Probability of
peat resistance
de volopment
Low; since only small numbers of
insects are dlroctly affected by
pheromone in traps. However,
pheromone formulation Is critical
Moderate to high; insects lured
to pheromone-balted traps are
removed from population. Phero-
mone formulation Is critical.
Moderate; more of population is
affected thereby increasing the
selection pressure. However,
considerable latitude in phero-
mone formulation lowers
probabilities.
Concentration
gradient ("edge
effect")
Not a problem with widely spaced survey traps; however, with
increased trap density, pest Insects from adjacent areas nay be
drawn to treated area.
Could attract. Insects from adja-
cent areas, causing increased
population levels in periphery.
Health risks
Extremel> low to low; only small
In traps.
quantities of pheromone are used
Yet to be determined; direct
application in the field rt»qulres
quantities greater than used in
traps.
Competition from \ Odors such as those from food sources and ovlposition sites may compete with the sex attractant odor,
other stimuli
179
-------
beetle (T. lineatum) causes pinholes in lumber, affecting hardwood and
softwood in the northern hemisphere.
In western U.S. forests, beetles emerge during late spring
(about May), and are active until late summer (through September).
Individuals range several miles, and newly emerging adults must aggre-
gate to both reproduce and cause significant tree damage. Female
Dendroctonus releases pheromones which attract males and other females,
promoting colonization in trees where mating and feeding occur. Colo-
nized trees rapidly lose their ability to defend themselves when they
become overwhelmingly infested; fungal pathogens that interrupt water
conduction to the foliage are introduced, and the trees soon die. Other
insects later invade the trees; they render pine unmerchantable after
one year and Douglas fir after two to three years. Larger, dominant
trees have been favorite targets of the western beetles. At lower popu-
lation densities, these pests exploit trees stressed by various factors
such as root disease, smog, flooding, drought, and lightning.
In the southern pine forests with warmer and more humid cli-
mates, beetles are active nearly all year. Trees of all sizes are
attacked, but the insects favor trees which have suffered some initial
distress from other factors. 'Lightning is a prime promoter of infesta-
tion; injured trees apparently emit volatile terpenes which attract
beetles. High rainfall and attendant poor drainage also promote stress
and invite pest infestation. Once a tree has been infested, the colony
later spreads to nearby trees.
Since bark beetles attack the bark rather than foliage, insec-
ticide cannof be applied efficiently by airborne spray. Spraying at
lower levels is effective for prophylaxis, but these applications are
limited to accessible areas. The use of small off-road vehicles for
low payload or hand-carried spraying ig not practical. In western
forests, except for small recreational and other selected areas, there
has been no major insecticide program against bark beetle since 1960.
In southern pine forests, lindane has been frequently applied with
questionable success on recent chronic infestations.
Good timber management, particularly the control of stocking
densities, appears to be the most effective current practical technique
for reducing bark beetle infestations and timber loss. In the West,
these insect problems have generally not caused enough economic impact
for industry to seek new counter measures, and lumber price adjustments
are apparently used to cover any significant losses.
Since beetles are attracted to large stands of mature and
over-mature trees, to trees kept in holding areas as part of lumber
price strategy, and to extensive areas of second growth and overstocked
stands 50 to 100 years old, the timber industry in California, Oregon,
and Washington is shortening the rotation time for cutting and refores-
tation. In addition, many mills have been equipped to process much
smaller trees for lumber and chips, which enables commercial thinning
of dense, overstocked stands.
180
-------
Significant progress has been made in the study of the use of
sex pheromones to control bark beetles. Pheromones have been isolated
and identified for all of the five Dendroctonus species mentioned
previously; D. brevicomis has been well studied, and some field trials
have been performed. The pheromone was isolated and identified as a
three-component mixture containing myrcene from the host (ponderosa pine),
frontalin from the male, and exobrevicomin from the female. Synthetic
production processes have been developed for frontalin and exobrevicomin,
and myrcene is commonly available.
The pheromone has been evaluated against the western pine
beetle in both survey and control situations. The control studies used
baited traps, baited trees, and aggregation interruption techniques;
surveys used baited traps.
Small scale field trials conducted by Wood and co-workers
during 1968-1970 demonstrated that '"stickum" coated traps, with
attached vials containing the evaporating pheromone, were particularly
effective in catching bark beetles. Comparison traps containing the
sticky substance, but no pheromone, were ineffective. Following these
observations, a large scale field trial was conducted in 1970 at Bass
Lake, California. This project was carried out over 20 square miles in
a recreational and timber forest area just south of Yosemite Park.
Baited traps were set in April to catch newly emerging adult beetles,
and approximately 400,000 were caught, plus an additional 190,000 in
survey traps. (Experimental unbaited traps were ineffective.) Tree
mortality was reduced about 70 percent during the test period. Aerial
photography, which has been ongoing, showed that significant reduction
of tree mortality extended for at least two years beyond the field
trials. Western pine beetle population surveys conducted for one year
after the treatment showed a substantial reduction.
A similar test utilizing the dead trapping method was conducted
at McCloud Flats in the Mount Shasta region of California during 1971-
1974. This site had ten times as many infested trees as the Bass Lake
site before treatment. Again, an excellent catch of beetles was achieved
(approximately four million) but effectiveness in saving trees did not
approach the Bass Lake results. Beetle-related tree mortality remained
essentially unchanged. The apparent cause of the differing test results
was a higher population of insects relative to the number of traps
placed in the McCloud Flats trials.
Another experiment was conducted with baited traps placed in
barren, volcanic areas adjacent to forest stands. Some traps placed
with imitation trees attracted higher numbers of beetles, suggesting
a visual component to the pheromone attraction process. The investi-
gators also used the condition of unforested areas adjoining the forest
to investigate the distance over which the attractants exerted their
action. This distance was found to be at least a quarter mile.
181
-------
Trap-out methods may be acceptable for recreational sites and
partially developed areas. However, in remote mountain areas traps
would have to be placed while snow is still present in order to be ready
for newly emerging beetles. There is an associated high cost due to the
inaccessibility of many such areas that could be treated. The inter-
ruption strategy is probably appropriate for these areas. Bark beetles
must aggregate to sustain reproduction and to produce tree damange.
Airborne application seems to be the most practical means to apply the
pheromone in large, inaccessible stands. The large payloads involved
would require high capacity aircraft.
Western pine beetle suppression trapping using slow release
pheromone formulations ideally requires only one application per summer;
about 0.25 pound of each of the three components is required for a
100 square mile area. Some trials have also been conducted with the
aggregation interruption method; on the basis of one application per
summer, the required quantities are 2.4 pounds per 100 square miles.
Recent prices for the pheromone components are: $50 per pound for
myrcene; $150 per pound for frontalin; and $3,000 per pound for exo-
brevicomin.
'Recent evidence indicates that the dextro-rotating isomer of
brevicomin is the active form. However, it is unlikely that optical
resolution of the racemic synthetic material would be economically
feasible. The high price of the pheromone causes the cost of trial
applications to be too great for all but major industry or the federal
government; practical evaluations with cost effective, successful results
will be required to convince industry to participate in the development
of these pheromones.
There ie no significant degree of commercialization yet for
bark beetle pheromone combinations, although some commercial organi-
zations have conducted preliminary marketability studies. Use and
formulation patents may be awarded, and Boyce Thompson Institute has
applied for such patents for frontalin and exobrevicomin. As with
nearly all pheromones, slow release formulations are preferred for
optimum efficiency. The fairly high volatility of the pheromone com-
pounds requires further developments in slow release technology to
provide reliable formulations. The volatility problem is accentuated
in western areas where application will probably occur at high alti-
tudes with temperatures above 32°C.
A two-phase slow release formulation has been developed by
Concerco Company, Seattle, Washington, and tested in the laboratory and
field for persistence. The formulation appears promising and can be
mass produced.
There is no storage problem with myrcene-frontalin-brevicomin
mixtures if an antioxidant is present to protect the myrcene. Stability
under field conditions (heat, air, sunlight) has not been reported, but
182
-------
gross deterioration has not been observed in exposed vials. Acute
toxicity is very low, producing only mild irritation at high concentra-
tion. No known chronic toxicity* studies have been performed.
The southern pine beetle is the primary pest in the southern
pine belts ranging from eastern Texas and Arkansas to the Atlantic Ocean,
and north through Tennessee and the Carolinas. The Ips complex beetles
and the black turpentine beetle are secondary pests which would cause
comparatively little damage if southern pine beetle populations were
selectively reduced. In 1974 Texas suffered an approximate loss of
three million board feet of timber due to beetles; similar losses are
believed to occur in other southern states. There is a large loss of
pulp and construction wood, and salvage and direct control are compli-
cated by the pattern of mortality and ownership. The cosmetic effect
of infestations in suburban forests also arouses public concern. The
USDA has selected the southern pine beetle as one of three insect pests
for major R&D attention, and it initiated a five-year program aimed at
the development control methods in fiscal 1975.
The pheromone of the southern pine beetle has not been fully
identified (as of the date of this research), but is known to contain
frontalin and a-pinene. A mixture that allows the conduct of field
tests has been compounded, but these components are not sufficiently
competitive with natural sources of attraction; further research aimed
at identification is under way. Investigators expect that frontalin and
a-pinene will be the major constituents of eventual pheromone formu-
lations, but they believe that volatility problems will delay the
development of a slow release formulation. However, frontalin and
a-pinene are not expensive materials, and Concerco's developments in
slow release formulations for the western pine beetle seem to indicate
good prospects for a low-cost interruption methodology for the southern
pine beetle. Population evaluation methodology comparable to that used
with the western pine beetle is being developed.
Some progress has been made in pheromone identification for
the genus Ips. Three terpene alcohols have been identified from
.1. paraconfusus: (+) cis-verbenol, (-)-2-methyl-6-methylene-7-octen-4-
ol (Ipsenol), and (+)-2-methyl-6-methylene-2,7-octadien-4-ol (Ipsdienol).
One or more of these compounds show cross attraction to other members
of the genus. No developmental pheromone work concerning these insects
appears to be in progress. However, practical synthetic procedures
have been developed for the last two alcohols listed above, while cis-
verbenol is readily available.
The pheromone of the Dutch elm vector (S^. multistriatus) was
recently identified as a three-component mixture. Two of the chemicals,
2,4-dimethyl-5-ethyl-6,8-dioxabyclo (3, 2, 1) octane (multistriatin)
and 4-methyl-3-heptanol, are produced by virgin female insects, while
the third compound, a-cubebene, is produced by the elm tree. The mixture
attracts both males and females to breed and feed upon host trees. The
actual elm disease is a fungus carried by the beetles and is transmitted
183
-------
to trees when the insects bore into crotches of branchlets that are two
to three years old. As with the Dendroctonus attacks, the transferred
fungi interfere with water transport in the trees. The insects' are
not attracted to healthy trees; they prefer unhealthy ones, usually
those which have already been infected with the disease. These weakened
trees are massively invaded and new broods are produced.
Dutch elm disease has historically been a problem in cities
in the eastern and midwestern United States, although it is now a
problem in California as well. Besides the aesthetics of damage to
these valued shade trees, cities and municipalities with infestations
incur considerable expense in applying pesticides* and removing dead
trees. The city of Detroit, for example, has spent up to $1,000,000
per year to protect its elm trees and about $10,000,000 per year for
removal of dead ones.
The adult beetles emerge in three peak periods during the
summer. Pheromone combinations are presently being used to monitor the
insects, and limited field trials have been conducted with the trap-out
control technique. Sticky traps with attached vials of evaporating
pheromone mixture are placed in the trees. A single trap uses 40 to
100 micrograms of each component per day or from 10 to 20 milligrams
per season. For maximum effect, several traps are required for each
tree, and the per-season amount is about 100 milligrams per tree or
about one pound for 5,000 trees. Data were not obtained for the maxi-
mum anticipated use. The field trials showed good catches of beetles,
but the number of trees actually saved was apparently not commensurate
with the numbers of trapped insects. No investigation of the aggregate
interruption method had been attempted as of the completion of this
research.
The production costs of the pheromone compounds for this pest
vector will probably be fairly high. The most complex compound (multi-
striatin) will be the most expensive and may be similar to exobrevicomin
in cost. Toxicity data have not yet been obtained, and no significant
degree of commercialization currently exists. The cubebene component
is somewhat unstable, although slow release formulations are being
developed which may reduce problems associated with instability.
*0ne of the primary insecticides used for control of these beetles has
been methoxychlor, which is used at a rate of approximately 25 pounds
per acre. This rate, higher than that for normal agricultural use,
in densely populated areas is a matter of environmental concern, espe-
cially in Detroit where drainage may enter Lake Michigan. In addition
to beetle controls, applications of chemicals (e.g., Benamyl®, Lignasan
BLF®) intended to control the fungal pathogen itself have been increasing.
184
-------
Gnathotrichus sulcatus, a relative of the ambrosia beetle
(Z* lineatum) has been investigated for its pheromones. A population
aggregation pheromone was recently isolated and identified as 6-methyl-
5-hepten-2-ol. Synthetic material is currently being evaluated in
British Columbia as a survey tool around sawmill sites and as a means
of preventing beetles from attacking green lumber after emerging from
infested logs.
Gypsy Moth—The gypsy moth, Porthetria dispar, is an imported
defoliating pest. Since its accidental introduction to New England in
1869, this pest has steadily spread southward and westward, preferring
oak but also defoliating many other hardwood species and conifers such
as spruce and pine. It has been reported as far west as Ohio and
Michigan, and is expected to range still further west and south.*
Gypsy moth infestations have been averaging nearly a million acres in
the northeast; there has been some effect on timber, but it mostly
causes undesirable aesthetic effects due to severe defoliation by vast
hordes of caterpillars. Insect populations build up over several years
prior to a dramatic population increase (outbreak) which lasts two to
three years. Hardwoods can tolerate one year of defoliation, but show
about 15 percent mortality after two consecutive years and 80 to 90 per-
cent after a third year.
The female moth does not normally fly, but secretes an attract-
ant to lure the males. The pheromone has been isolated and identified
as cis-7,8-epoxv-2-methyloctadecane (disparlure); it has been synthesized
and the resulting material has been found to attract males in laboratory
and field tests. Laboratory reared moths have shown more response to
synthetic attractant than field moths, possibly because there were no
competing natural female sources. The pheromone has been used for
survey purposes and has allowed investigators to detect the outer
limits of migration where few moths have yet appeared. The use of
disparlure for population estimation requires further development, e.g.,
computerization to calculate population dynamics.t
*
A gypsy moth citing in the San Jose area of northern California was
reported prior to the publication of this report.
+_ ,
Better population dynamics data are needed for this insect since low
population density changes may be important during the years between
major outbreaks. One problem with the gypsy moth is the approximate
one-half mile wind drift of webs of the newly emerging larvae; this
would dictate large area application of the disparlure pheromone at
least on the windward side to attain effective suppression.
185
-------
Mass trapping control has been attempted, but has not been
effective, probably because of the high insect population densities
during outbreaks. However, the method is still regarded as potentially
useful if various logistic problems related to trap numbers and place-
ment can be overcome. The large areas involved require aircraft
delivery. Traps have been dropped in grid patterns, but were not totally
effective on the ground. It is expected that properly placed traps
may be more effective, but manpower requirements for placement and
service appear to be prohibitive.
Mating interruption is currently under investigation and is
considered potentially successful. The pheromone is applied in a
microencapsulated, slow release formulation by aircraft spraying.
Encouraging results were obtained from field trials conducted in 1972
and 1973, but these tests were conducted against controlled population
densities in confined areas rather than with the high densities common
to epidemic infestations. Although improved slow release formulations
are expected to provide better results, serious outbreaks may not be
controlled by pheromones alone. It has been estimated that 93 percent
inhibition of mating is necessary to control gypsy moths, and inhibi-
tion with disparlure is presently less than 90 percent.
Disparlure has been produced in large scale by Chemical
Samples Company and Farchan Chemical, and microencapsulated by Pennwalt
Corporation and National Cash Register Company. The compound can be
conveniently synthesized on a large scale; however, extensive purifi-
cation is currently involved. It stores well as a neat liquid at room
temperature and microencapsulated material is at least as stable.
The current price is $115 per pound in neat form and $225 per pound
encapsulated. In the encapsulated form disparlure is applied at the
rate of 10 to 15 grams per acre with a material cost of $5 to $7 per
acre. Production costs can probably be reduced by about 50 percent
with larger scale manufacture.
Both acute and chronic toxicity of disparlure are quite low.
Spruce Budworm—The spruce budworm belongs to the family
Tortricidae, whose larvae feed on buds and newly forming foliage in
coniferous trees. These moths have a population cycle similar to the
gypsy moth where periodic outbreaks may follow several years of relative
inactivity. Outbreaks are exceedingly difficult to control because of
overwhelming numbers and the very large and often remote areas involved.
Conventional pesticides are used to control this pest by at least keep-
ing trees alive. A major complication in control has been the migration
distance of these moths which may range up to 30 kilometers per night
or 150 kilometers in their life-span, depending upon wind conditions.
Two economically important species are recognized: Christoneura
fumiferana appears in the eastern United States and Canada and feeds
upon balsam fir and white spruce forests; C. occidentalis appears in
the west and feeds on Douglas and white fir.
186
-------
A pheromone attractive to both fumiferaria and >C. occldentalis
has been isolated and identified as trans-ll-tetradecenal. Field trials
have substantiated this attraction, but minor amounts of other unidenti-
fied compounds may also be required. Survey trials conducted in east-
ern Canada in 1972 showed that single virgin females were at least as
effective for luring males to the traps. However, slow release formu-
lation technology should improve the results of the synthetic bait (a
previous problem was the erratic release rate of the aldehyde from
polyethylene vial stoppers). As with the gypsy moth, trapping for
control of the spruce budworm is impractical because of the high popu-
lation density during outbreaks, the inaccessibility of terrain, and
the vast acreages to be covered; high density would also cause poor
results with mating interruption techniques. A proper pheromone-based
strategy would require careful monitoring with pheromone traps during
nonepidemic years, followed by application for mating disruption when
evidence indicates a population buildup. Additional analogs such as
trans-ll-tetradecenyl acetate and the corresponding alcohol appear to
be mating inhibitors which might Also be used in control programs.
Either mating disruption method would require aircraft delivery over
large forest areas.
Since disruption testing has been carried out on a very
limited basis (and without sustained release formulations) it is not
yet possible to estimate the required amounts of pheromone or the costs
per acre for application. The material is not difficult to synthesize
and could be produced in suitable quantity, but its apparent high
volatility could make formulation more difficult. However, potential
instability from oxidation should be reduced by slow release formulation.
Douglas Fir Tussock Moth—Douglas fir tussock moth, Orgyia
pseudotsugata, causes severe defoliation of fir forests in western
North America. Annual population levels vary greatly with an irregular
cycle every few years. Populations can increase dramatically within
one season, reach very high levels and cause severe damage to fir
stands for one or two years and then subside (usually because of a
natural virus disease). This annual fluctuation in pest populations
has complicated assessments of potential control measures. Currently,
there are no means of controlling these outbreaks.*
Recently, Smith et al. reported the identification and synthe-
sis of the sex attractant pheromone for this moth as (Z)-6-heneicosen-
ll-one. Laboratory bioassays indicated that both stereoisomers are
active, but that the Z isomer is more active than the E. However, only
*
As footnoted earlier, the U.S. Forest Service received prior to the
publication of this report a registration for the use of the Orgyia
pseudotsugata nuclear polyhedrosis virus for the control of the
Douglas fir tussock moth.
187
-------
the E isomer was used in the initial field trials, because of an incorrect
initial assignment of E stereochemistry to the natural pheromone. If
subsequent field trials confirm the preliminary field data with the
incorrect isomer, the pheromone should be effective for surveying fir
forests for population buildups. Its use in other control strategies
has not yet been tested.
2. Agricultural Pests
Pink Bollworm—Pink bollworm (Pectinophora gossypiella) is a
worldwide cotton pest; in the United States, infestations have occurred
in cotton raised in Texas and arid southwestern areas at various times
over the past 40 years. However, these outbreaks have been either
eliminated or controlled by pesticides or crop management techniques.
The current problem in the Imperial Valley in southern California began
about eight years ago. The present area affected is 50,000 to 75,000
acres but observers fear that the insect will spread to approximately
one million acres of cotton in the San Joaquin Valley. Arizona has an
additional 300,000 acres which could also be affected.
During its larval stage, the pink bollworm feeds almost
exclusively on cotton. Treatment of the host crop with insecticides
has been an effective control, but additional problems, including some
that involve secondary pests (e.g., cabbage looper, beet armyworm,
spider mites) have resulted from spray programs. Insecticide treatments,
if continued too long, frequently kill natural predators of pink boll-
worm and the secondary pests. Recent annual control costs have been
about $90 per acre for ten to twelve applications of insecticide
(carbaryl, monocrotophos, or azinphosmethyl). The process is somewhat
dangerous since Applications are typically made at night when the adult
moths are active.
The sex attractant pheromone for pink bollworm has been iso-
lated and identified as a mixture of cis,cis and cis, trans-7,11-
hexadecadienyl acetate. In laboratory and field trials a 50:50 mixture
of the two isomers (gossyplure) produces an optimal effect compared to
other mixtures and to hexalure, a synthetic analog of the true pheromone.
While gossyplure-baited traps are currently used to monitor
moth population levels, the major anticipated use of the pheromone is
for mating disruption (mass trapping is not considered a practical
control strategy).
Field trials conducted on small acreages over the past several
years generally achieved a high level of mating suppression, even when
the less potent hexalure was used. The 1973 trial with hexalure reduced
the number of mated females by 80 percent. A larger 4,000 acre test
in the Coachella Valley area with gossyplure in 1974 caused a 75 percent
mating reduction. While these figures are impressive, over 80 percent
reduction may be necessary to achieve tru'e control. Improvements in
188
-------
formulation, application, and the use of check-control areas should
provide better performance data. In 1975 field trials were conducted
using microencapsulated gossyplure applied in ultra-low volume aerial
spray at a rate of one gram per liter of water per acre, with applica-
tion every two weeks or six times pet season. Check plots—untreated
and conventionally treated with chemical insecticides—were also*used
to assess the effectiveness of gossyplure.
At present the pheromone has been produced in about 50-kilogram
quantities by combining several batch processes. The material has been
manufactured by Chemical Samples Company and Story Chemical Company at
prices ranging from $1 to $5 per gram of neat material, depending on
the supplier and level of purity. Synthesis is reasonably difficult,
but it can probably be improved with new approaches, and the ultimate
cost may become less than $1 per gram. Less pure material has been
found quite acceptable for field application. Formulation and novel
synthesis patents could be obtained for this pheromone.
The material is a liquid and appears to be stable at room
temperature during storage. In the field, however, an antioxidant
must be present. The lower volatility does not suggest a problem for
compounding into slow release formulations. The expected formulated
cost is about $1.50 per gram, which extrapolates to a seasonal cost
of $35 per acre, including material and aircraft delivery.
At least one commercial organization is proceeding with
the production of a formulated pheromone for further field trials
and control use in the very near future. Toxicity testing is being
conducted by this organization. As with other pheromones, neither
acute nor chronic toxicity is anticipated for gossyplure. The pheromone
could be in actual use in about three years, assuming the granting of
an early registration.
Boll Weevil—The boll weevil, Anthonamus grandis (Boheman),
is .a major pest of cotton raised in the south and southeastern United
States. The weevil was introduced into the United States from Mexico
in 1892. It is a hardy insect with no significant natural predators
or parasites in U.S. cotton ecosystems and no other known hosts of
economic consequence. Population levels depend on cotton acreage and
the stage of plant development. Up to ten million acres may be affected
annually in the United States. Recent annual economic losses directly
associated with this pest have been estimated at about $260 million
(yield loss plus control costs), making the boll weevil, on a dollar
basis, the single most destructive insect In the United States.
Similar to the pink bollworm, control of primary and early
season pests with conventional pesticides often triggers severe second-
ary infestations, primarily Heliothis virescens and Heliothis zea.
(A prevalent early season pest in the southeast is the plant bug, Lygus
lineolaris. and the banded-wing whitefly, Trialeurodes abutilonea. is
becoming an increasing problem because of resistance to chemical
189
-------
pesticides.) Methyl parathion and toxaphene have been the most exten-
sively used insecticides, with recent average control costs ranging
from $0.53 per acre on the High Plains of Texas to $36 per acre in
southern California.
The attractant pheromone for the boll weevil has been isolated,
identified, and synthesized by Tumlinson et al. It is a four-compound
mixture of cis-2-isopropenyl-l-methylcyclobutaneethanol (I), cis-3,3-
dimethyl-A1,3-cyclohexaneethanol (II), cis-3,3-dimethyl-A1-q-cyclohexane-
acetaldehyde (III), and trans-3, 3-dime thy l-A^a-cyclohexaneacet aldehyde
(IV). The total mixture is commonly called grandlure. The components
have been mixed in varying ratios, but a 40:30:15:15 combination of
I-II-III-IV, respectively, is currently favored.
Grandlure acts as a female boll weevil sex attractant and also
as an aggregation attractant for both males and females, at different
stages of the growing season and insect development. Experimental evi-
dence to date suggests that grandlure has significant potential for
population survey, but doubtful potential below economic threshold as
a physical trapping control agent. No experiments with grandlure
sprayed on an entire field for mass interruption of normal mating patterns
have been attempted, primarily because of the material's cost. Such
confusion experiments, however, may be attempted using a mixture without
the most expensive compound (I).
Grandlure has been used with trap surveys to study weevil
populations and their dynamics. Boll weevil response and capture with
grandlure-baited traps is primarily influenced by (1) ratios of the four
components, (2) dosage and release rate, (3) trap size and color,
(4) the height of trap placement, (5) competition from naturally produced
insect and plant attractants, (6) temperature, wind, and humidity, and
(7) trap density. There are also indications of differential response
to the component ratios by geographic area.
The biggest drawback to traps has been the lack of an
accurate measure of trap efficiency (e.g., a trapping index). Correla-
tions between weevils captured and total field populations have not been
established. A number of studies have shown that although traps baited
with grandlure capture a significant number of weevils, many others are
not captured. The competition from other field stimuli, particularly
from the cotton plants themselves, is quite significant. It is pos-
sible that further research will show that other, minor components must
be present or that more precise ratios of components must be employed
to obtain a more competitive pheromone.
Current commercial producers of grandlure include the Farchan
Division of Story Chemical Company and the Chemical Samples Company.
Formulated products are produced by these companies and the Herculite
Corporation. The latter organization produces a slow release formulation
consisting of a three-layer plastic laminate containing the formulation
in the center matrix. The material comes in strip form and can be cut
190
-------
to desired lengths. The price for neat materials has recently ranged
between $27 and $30 per gram for compound I produced at the kilogram
level, and less than $20 per gram for the other compounds. The current
overall formulated price is $35 per gram. Large scale production may
decrease these prices. Per acre quantities required for survey use are
estimated at three milligrams per trap per service, requiring about
50 milligrams per acre per season or one kilogram per 20,000 acres.
Improvements in slow release formulation technology will probably lower
these requirements.
These compounds are liquids, with approximate boiling points
of 110° to 116°C, are soluble in organic solvents, and are immiscible with
water. The alcohols, I and II, are reasonably stable under normal condi-
tions, and can be stored neat under refrigeration for an indefinite
period. The aldehydes, III and IV, are unstable and should be kept in
solution (approximately ten percent in pentane) under refrigeration; they
are unstable in certain solvents such as methanol. Recent slow release
formulations containing all four components are reported (but not yet
proven) to have increased stability, allowing the material to be stored
for extended periods above 37°C.
The materials are acutely nontoxic, but chronic toxicity data
on mutagenicity, carcinogenicity, and teratogenicity are unavailable.
Grandlure is being considered for an experimental use permit (for uses
other than in traps), but may not .be ready for registration as a viable
agent. The USDA holds a public use patent on grandlure. No composition
of matter patents have been issued.
Orchard Pests—Several species of insects, all tortricid moths,
are economically important orchard pests. The major pests include the
codling moth (Laspeyresia pomonella), fruit tree leaf roller (Archips
argyrospilus), grape berry moth (Paralobesia viteana), lesser appleworm
(Grapholitha prunivora), oblique-banded leaf roller (Choristoneura
rosaceana), oriental fruit moth (Grapholitha molesta), red-banded leaf
roller (Argyrotaenia velutinana), summer fruit tortrix (Adoxophes orana),
and tufted apple budworm (Platynota idaeusales).
The insect pest problems of the several U.S. orchard crop
regions range from infestation by a single major pest to infestation by
several species. This section discusses only two insects for which a
sex attractant pheromone system has been tested as a control—the codling
moth and the leaf roller complex. Pheromones for the other orchard pests
are still in earlier stages of basic research.
The codling moth is a pest of apples, pears, walnuts, and other
deciduous fruit trees. The fruit is used by the moth as a source of food
and shelter during its larval development. Current control procedures
involve spraying the surface of the fruit with chemical insecticides
several times from fall until harvest. In the course of these spraying
programs for codling moth, beneficial organisms that help control other
191
-------
orchard pests (e.g., mites, aphids, pear psylla) are destroyed, neces-
itating the additional control of these secondary pests.
In several fruit producing areas, sex attractant pheromones
are being used on a trial basis to monitor population levels of the
codling moth to indicate when chemical insecticides can be most effec-
tively applied. Pest monitoring with pheromone baited traps in one area
of California during the 1969 through 1971 reduced the total cost of
pest control for both codling moth and secondary pests in nine test
orchards from an average of $75 per acre to an average of $35 per acre.
Cost reductions ranged from $15 to $89 per acre. (It should be noted that
a significant part of the value of these crops is related to exterior
appearance, so most growers try to protect against cosmetic damage through
the liberal application of chemical pesticides.)
The pheromone for codling moth, (Z)-2,(E)-6,7-methyl-3-propyl-
2,6-decaden-l-ol, is currently manufactured for use in baiting traps.
It is produced as either a wick or a slow release formulation and is
packaged in foil to prolong shelf life. The manufacturer claims that
refrigeration is not required for storage, although the pheromone should
be kept cool. One application of the slow release formulation is reported
to last the entire season.
Information on toxicology is not available since these tests
have not been completed.
The larvae of the red-banded leaf roller moth, like those of
the codling moth and other tortricid moths, are extremely destructive
orchard pest;s in the northeastern United States. Control of this insect
alone, however, has resulted in increased populations of normally minor
pests such as the oblique and fruit tree leaf rollers.
The sex attractant pheromone for the red-banded leaf roller
and the oblique leaf roller has been identified. (Z)-ll-tetradecenyl-l-ol
acetate (TDA) is the sex pheromone of both species, but the red-banded
leaf roller requires dodecylacetate (DDA) as a synergist. Using the
pheromone-synergist combination in mass trapping techniques at 40 traps
per acre, Roelofs et al. were able to obtain 98 percent control at low
population levels. However, the labor required to place and service
these traps reduced the economic advantage over control through pesticide
spraying.
Population levels are being monitored with pheromone baited
traps in some orchards. Individuals using the monitoring techniques
believe that additional research on the relationship between pest popu-
lation levels (as estimated by trap catches) and crop damage is required.
A second method of population estimation is needed to assess the effec-
tiveness of the pheromone traps.
Both mass trapping and monitoring techniques require the pre-
cise formulation of pheromone for the target pest species. As described
earlier, two species of leaf rollers respond to the same sex pheromone.
192
-------
However, the red-banded leaf roller requires DDA to synergize the phero-
mone TDA, while the oblique leaf roller's response is inhibited by DDA.
Therefore, if one pest is controlled, other methods must be used to
control the other. When saturating the atmosphere with pheromone to
disrupt mating, the presence of DDA is not critical. For these reasons,
a microencapsulated spray formulation for use in mating disruption of
two species of leaf rollers was initiated in 1975, with the microencapsu-
lated pheromone applied at five grams per acre by conventional spray
rigs.
European Corn Borer—The European corn borer, Ostrlnia
nubilalis (Hiibner) , is a significant pest of U.S. corn production. The
borer became a problem to U.S. agriculture in 1945, and by 1949 to 1950
it was affecting the midwestern corn producing area. The borer is
currently controlled by chemical insecticides (carbaryl and malathion)
applied in fairly large volumes.
The adult insects are moths (Lepidoptera:Pyralidae), but the
larval form is responsible for crop damage. Flight distance capability
is unknown, making it difficult to estimate the required application
area for pheromones. The pheromone was isolated and identified as a
mixture of geometrical isomers of 11-tetradecenyl acetate. It was later
found that moths from different geographical areas respond to varying
percentages of cis and trans isomers, and that these ratios are critical.
Moths in most geographical areas responded to mixtures of 90 to 95 per-
cent cis and 5 to 10 percent trans. However, some favored a nearly
reciprocal ratio, with 85 to 90 percent trans.
The pheromone has been synthesized and field tested, but the
trials have been discouraging. The moths tend to circumvent the attract-
ant, apparently by employing a number of alternate methods of mating
communication. Additional information on the mating behavior and popu-
lation dynamics of these moths is needed. The pheromone has not been
developed to the point where even survey use is practical. This will
require two to three more years of investigation and trial. If there
actually are significant alternate communication pathways for mating,
control by trap-out will probably not be viable, and a disruption strategy
should also encounter difficulty.
There are no unusual problems of production, formulation, or
toxicity with the European corn borer pheromone. It is readily synthe-
sized, and slow release formulations are becoming available. The materials
are stable in storage and if there is any photochemical or oxidative
degradation in the field, it would probably be overcome by microencapsu-
lation. The compounds are innocuous during application and as residues,
at least as far as acute toxicity is concerned.
193
-------
Stored Product Pests—Principal stored product insect pests are
beetles (Coleoptera:Dermestidae) and moths (Lepidoptera). Some members
of the Dermestidae have been investigated for pheromone activity, such
as the khapra beetle, Trogoderma granarium (Everts) and the related
T\ inclusum (LeConte), and the black carpet beetle, Attagenus megatoma.
Several of the Lepidoptera have been examined, including the almond moth,
Cadra cautella (Walker), the Indian meal moth, Plodia interpunctella
(Hiibner), the Mediterranean flour moth, Anagasta kuhniella (Zeller) , the
tobacco moth, Ephestia elutella (Hiibner), and the raisin moth, Cadra
figulilella (Gregson). The insects mostly feed on stored grain materials
and are serious problems in warehouses and the holds of ships. These
insects exist primarily in such captive surroundings, but they have some
flight capability.
Economic losses attributable to these pests can be great because
stored commodities are near marketing and production plus distribution
costs are involved. In some southern hemisphere countries serious infes-
tations often occur, especially at the domestic storage level. Chemical
pesticide control can be effective, but in the United States the presence
of insect or insect fragments in stored products for food use may lead
to rejection of entire lots; thus, even a small number of insects can
exert an economic impact, particularly if they multiply from undetected
low levels. In addition, pesticide residues are an obvious concern.
The application of pheromone trapping techniques for both pest
survey and control appears to be promising for stored agricultural
commodities. The insects release their pheromones during periods of
maximum light exposure. In storage areas they generally exist at low
population levels and often receive little exposure to light stimuli;
this reduces competition from naturally secreted pheromones. Flight
capability, although weak, could cause some attraction from the outside
to the storage facilities. However, traps could be externally placed
to attract insects out of these facilities. Mating disruption techniques
are not considered viable because of the necessity to thoroughly eliminate
the pests. In addition, the relatively large amounts of pheromones
required could create residue problems, i.e., besides their potential
human toxicity, residues could attract additional insects after redis-
tribution of commodities to new storage areas.
Substantial progress has been made in the development of phero-
mones for Dermestid beetles. The primary sex attractant for Trogoderma
inclusum has been isolated and identified as trans-14-methyl-cis-8-
hexadecen-l-ol. Other Trogoderma species are also attracted by this
compound. The black carpet beetle (Attagenus megatoma) attractant has
been identified as trans-3-cis-5-tetradecadienoic acid (megatomic acid).
Both of these attractants are functional as single compound pheromones,
so that the necessity for reproduction of complex natural mixes does
not arise. The attractants can be converyLently synthesized and are
produced by the Chemical Samples Company'.
194
-------
The Trogoderma attractant has been field tested in grain
storage areas with good results. Since the beetles are present in low
numbers, survey traps may also act as control traps, particularly when
they are used to capture survivors just after a chemical insecticide
treatment. The pheromone is generally used in a trap containing malathion
to kill the attracted insects. These traps are simple cardboard devices
which, including the pheromone and insecticide, currently cost about
30 cents each. Approximate trap requirements are 100 to 200 per ware-
house per week or $30 to $60 per week (depending on the size of the
warehouse). Additional experiments are in progress with pheromone baited
traps that contain an insect pathogen. The purpose is to infect an
entire insect colony through the attracted and infected males; a major
problem has been getting the males to leave the traps and return to the
colony to transmit the disease through the females.
The Dermestid attractants are oils that are convenient to syn-
thesize and apparently without stability problems. Toxicity data are not
yet available, but the chemical structures do not suggest problems.
Patents have not yet been filed either for synthetic processes, formu-
lations, or traps.
The pheromone activity of Lepidoptera has not been thoroughly
investigated. Cis-9-trans-12-tetradecadienyl acetate has been found to
be the principal attractant of the almond moth and Indian meal moth,
and it has also been found in the Mediterranean flour moth, the tobacco
moth, and the raisin moth. Traps baited with the pheromone have captured
significant numbers of the above species in storage facilities where all
were present at the same time. Such a wide spectrum of response to a
single pheromone is unusual, but is obviously advantageous for production,
application, and sales.
Cabbage Looper (Trichoplusla ni)—The larva of this moth is
a pest of a wide variety of economically important agricultural and
floricultural crops. The sex attractant pheromone of T_. ni.—cis-7-
dodecenyl acetate—was the first such pheromone identified for any
agricultural pest.
Most of the work with cabbage looper pheromones has been
directed toward mating disruption. Nocturnal release rates of about
4.1 milligrams per acre prodiiced about 90 percent disruption and about
41 milligrams per acre produced close to 100 percent disruption.
More research is required on mating disruption, since an
apparent lack of population density dependency has been noted in some
tests at different population levels of adult moths. Because adult
females appear to have a long flight range, tests should be conducted
on larger field test plots. The response of some closely related insect
pests (alfalfa looper, Autographa californica; soybean looper, Pseudo-
plusia incldens) and several other species—Heliothis zea (bollworm),
H. virescens (tobacco budworm), Spodoptera ornithogalli (yellow-striped
armyworm), and Pectinophora gossypiella (pink bollworm)—must be
195
-------
investigated. In earlier tests, mating of these insects was also dis-
rupted in the presence of the T\ nl pheromone.
Currently, relatively little research has been conducted on
this pheromone.
Other Agricultural Pests—Research on pheromone systems for
the control of other important agricultural pests is largely at the
basic level. Many pheromones are currently being identified and syn-
thesized. Those field experiments that have been undertaken have been
preliminary, and nothing conclusive has resulted. Most of these phero-
mones are three to five years from major field testing.
3. Human and Veterinary Pests
Musca domestica—The common housefly is a major vector of
typhoid, dysentery, diarrhea, cholera, yaws, trachoma, and many other
diseases. In addition, it serves as an intermediate host for roundworms
and tapeworms.
Carlson et al. isolated, identified, and synthesized a sex
attractant pheromone, muscalure (Z-9-tricosene), from female feces and
cuticle and demonstrated that it was an attractant to the male housefly.
In the field test reports to date muscalure increased the effectiveness
of various tested traps by three to twelve times. However, nearly
equal numbers of females and males were caught, suggesting that muscalure
may act as an aggregation pheromone rather than a male attractant.
A major problem with any housefly pheromone is that it must
compete with the varied and copious quantities of natural attractants
that local populations of flies have adpated to. In addition, M.
domestica has few antennal chemoreceptors and finds its way by a high
degree of exploratory activity in which it uses input from its visual,
gustatory, and olfactory senses. Therefore, the pheromone promises to
offer some aid in housefly control in situations where breeding sites
and sources of food have been reduced, but other, stronger measures
must be taken where this is not the case.
In early 1975, Zoecon Industries, & subsidiary of Zoecon
Corporation, began marketing "Muscamone" for use in traps. Zoecon syn-
thesizes the pure Z-9-tricosene and sells the technical material to
formulators.
Other Medical and Veterinary Pests—Pheromones for the lone
star tick, Amblyomma americana, and the brown dog tick, Rhipicephalus
sanguineus, have been isolated, identified, and synthesized, and are
being laboratory tested. The presence of pheromones for the tsetse fly,
Glossina spp., and the mosquito, Culiseta inornata, has been indicated
196
-------
but not yet identified. Pheromones have not yet been confirmed in most
species of mosquitos, fleas, and lice.
D. Decision Analysis of a Possible Pheromone Pesticide Venture
1. Introduction
This section presents an economic analysis of the commercial
feasibility of one pheromone product. The anlysis focuses on gossyplure,
an analog of the sex attractant pheromone for the pink bollworm.
Gossyplure was chosen since field trial data have been collected for
that product, its market potential is larger than for some other phero-
mones, and its use would not be expected to be dominated by government
agencies such as the U.S. Forest Service. This analysis presents the
implications of current levels of uncertainty on the ultimate profit-
ability of entering this business; however, a great deal of technical
and market data are still to be developed, and the analysis is by no
means a final assessment of the commercial feasibility of gossyplure.
2. Background on the Product
This information has already been presented on pp. 188-189.
3. Base Case
As with the preceding analyses of hypothetical B.t. and
Heliothis spp. NPV ventures, a single base case scenario under which
a gossyplure venture can be evaluated was initially developed. Sensi-
tivity studies determining the factors that most critically affect the
venture were then conducted, and probabilistic considerations developed
to describe the implications of uncertainty on the critical input factors
that determine the overall profitability of the venture.
Market Size—Since gossyplure is specific to the pink bollworm,
the maximum potential market size can be fairly readily determined. The
only major crop that is significantly damaged by this pest is cotton,
and the maximum potential market size is therefore limited to susceptible
cotton acreage. As described earlier, in the United States this cur-
rently consists of about 60,000 acres in the Coachella and Imperial
Valleys in California. (The pink bollworm has been effectively controlled
in Texas through cultural practices.)
California growers are concerned that the pink bollworm may
spread into the San Joaquin Valley, where about one million acres of
cotton are grown. However, possibly because of the distance involved
(approximately 200 miles), intervening mountains, or subtle climate
difference, the pink bollworm has not yet seriously infested the
197
-------
San Joaquin Valley. It is a major pest in a number of foreign cotton
producing countries such as Mexico and Egypt.
For base case purposes, only the Imperial and Coachella Valleys
were considered, and the maximum potential market size was accordingly
set at 60,000 acres. Alternative assumptions on the San Joaquin Valley
and foreign market potential are considered in later scenarios.
It was assumed that about 80 percent of the infested acres
would be treated with gossyplure if it is demonstrated to be effica-
cies, reducing the potential market size to about 50,000 acres. The
assumption seems reasonable since current pink bollworm insecticide
treatments are causing secondary pest infestations, and growers are
seeking alternate controls.
Since the market size is limited, it was further assumed that
no attempt would be made to circumvent the patents that will probably
be obtained for gossyplure. Thus, the entire potential market will be
captured by the one company that decides to enter the business. With
an application rate of one gram per acre per application, and six appli-
cations per year, the base cas'e market size for the hypothetical company
was determined to be about 300 kilograms of pheromone per year.
The sales volume was assumed to grow linearly from zero to
300 kilograms over a three year period, stay at that level for five
years, and then diminish to zero over the next ten years as the pest
becomes controlled or as alternative control measures evolve.
Production and Capital Costs—Because of the relatively small
quantities of material required annually, the hypothetical company
probably would not build a separate facility to produce the basic phero-
mone material. It was assumed that the company would instead contract
a chemical specialties firm to produce the neat material, and that it
would only be involved in the final formulation of the pheromone product.
Therefore, capital requirements should be quite small and almost negli-
gible if the company already has the space and formulation equipment
from other product lines. Working capital requirements were set at two
months' production costs.
Contract companies have recently been producing the basic
pheromone material for $1 to $5 per gram. The primary factor that
accounts for the price difference is product purity, but lower grade
product appears to be as efficacious as higher grade in the field.
Therefore, a production cost for the neat material of $1 per gram was
assumed.
Gossyplure is not extremely volatile and formulation costs
equal to 50 percent of the neat material costs, or 50 cents per gram,
were selected for the base case. Therefore, the total cost of producing
gossyplure, including contract production of the neat material and formu-
lation into final product form, was assumed to be $1.50 per gram.
198
-------
Selling Price—The selling price to users was difficult to
determine since there has been no formal marketing experience nor has
the efficacy of the product been finally established. However, some
assumptions were used to estimate a probable price range. The amount
that growers presently pay for pink bollworm control was determined and
used to develop a competitive price range for gossyplure. The principal
cotton pest difference between the Coachella and Imperial Valleys and the
San Joaquin Valley is the pink bollworm. Seasonal treatment costs for
all pests have recently averaged about $80 per acre in the Coachella
and Imperial Valleys and about $40 per acre in the San Joaquin Valley.
Therefore, as a first approximation, seasonal pink bollworm control costs
appear to be about $40 per acre. However, about three-eighths of this
cost is for application and it was assumed that growers currently pay
about $25 per acre per season for materials to control the pink bollworm.
The pheromone will be applied at a rate of one gram per acre
per season. Therefore, if a grower i^ currently willing to pay $25 per
acre per season for existing control materials, he should be willing
to pay $2.50 per gram for comparable control with the pheromone if it
were applied ten times per season (the current spray schedule) or $4.00
per gram if it were applied six times per year (the anticipated schedule
for gossyplure). For the base case, an intermediate selling price of
$3.50 per gram was assumed, of which the hypothetical company expects
to receive 85 percent and the distributor or reseller(s) the remaining
15 percent.
Research and Development Costs—Synthesis, scale-up, and small
field tests have been conducted on gossyplure; the remaining R&D expendi-
tures will principally be for large field testing and toxicology work.
Table 29 gives the R&D expenditures assumed for the base case. Syn-
thesis and scale-up costs are considerably lower than equivalent costs
for conventional pesticide products. Field tests, toxicology work,
and registration costs are approximately the same as for conventional
controls since all products must meet certain efficacy and safety stan-
dards. The table also shows that the hypothetical company will bear about
70 percent of the total R&D expenditures with university and government
groups conducting the synthesis research and some of the field testing
and toxicology work. The base case assumes that efficacy and safety
will be demonstrated, and that a registration will be obtained in 1977.
Cash Flow—The base case assumes that gossyplure will be reg-
istered after company R&D expenditures of $1.2 million; that the market
will be limited to the Imperial and Coachella Valleys; that the product
will sell for $3.50 per gram; that the company will spend $1.50 per gram
to produce it; and that negligible capital will be required for the
199
-------
Table 29
BASE CASE RESEARCH AND DEVELOPMENT COSTS
FOR GOSSYPLURE
Beginning Ending
Activity Year Year
Synthesis
1970
1973
Scale-up
1973
1974
Field tests
1973
1977
Toxicity
1974
1977
Registration
1976
1977
Total
Company Noncompany
Costs Costs Total Costs
(thousands (thousands (thousands
of dollars) of dollars) of dollars)
$ 0 $100 $ 100
30 0 30
400 200 600
750 250 1,000
30 0 30
$1,210 $550 $1,760
venture. Table 30 gives the pre-tax cash flow over time for this
scenario. Constant prices and costs have been assumed since inflation
will probably affect both equally.
The table shows that the net present value of the base case
is $1.3 million and $0.4 million for discount rates of eight and fifteen
percent, respectively. Thus, at either rate the project appears
attractive.
200
-------
Table 30
BASE CASE PRE-TAX CASH FLOW FOR GOSSYPLURE VENTURE
(Thousands of Dollars)
Year
R&D
Costs
Sales
@ $3.5
per
Gram
Production and
Distribution
Costs
Capital
Costs
Yearly
Profit
Net Present Value
(1974 dollars)
-------
4. Sensitivity Studies
The base case is not necessarily the most likely scenario,
however, and in this section the implications of changing some of the
base case assumptions are examined. Table 31 summarizes some of the
more important of these sensitivity relationships.
Table 31
SELECTED SINGLE-VARIABLE SENSITIVITY STUDIES FOR GOSSYPLURE
Scenario
Net Present Value
at 8 Percent
(millions of dollars')
Market Size
10,000 acres
50,000 acres
250,000 acres
500,000 acres
1,000,000 acres
0.6
1.3 (base case)
10.7
22.5
46.1
Selling Price
$2.00 per gram
$3.50 per gram
$5.00 per gram
0.8
1.3 (base case)
3.3
Registration
Obtained
Not obtained
1.3 (base case)
1.1
R&D
$1.0 million
$1.2 million
$1.8 million
1.5
1.3 (base case)
0.7
Building Capital
$ 0
$ 500,000
$1,000,000
1.3 (base case)
0.8
0.4
202
-------
As with the other potential pest controls analyzed in this study, the
table shows that the venture is very sensitive to market size. The
base case assumed a market of 50,000 acres (80 percent of the pink
bollworm infested acres of the Imperial and Coachella Valleys). It is
unlikely that the market would be smaller than this, because those acres
are already heavily infested with the pink bollworm, and growers are
searching for alternate controls. However, if the market size were only
10,000 acres, the venture would lose $0.6 million at eight percent.
Increasing the market to 500,000 acres, to cover the possibility of the
San Joaquin Valley becoming infested, increases the net present value
at eight percent to $22.5 million. A further market increase to one
million acres, to cover the possibility of both the San Joaquin Valley
and foreign countries becoming viable markets for the pheromone, raises
the venture's profitability to $46.1 million at eight percent, with all
other variables remaining constant.
However, markets as profitable as the latter two would probably
entice competitors, and no one company would be likely to capture 100 per-
cent. Reducing the hypothetical company's market share to 20 percent
of total market in each of these scenarios reduces the venture's profit-
ability to $3.8 million and $8.3 million, respectively.
Price is also an important factor, but the reasonable price
range is probably not as large as that for market size. Increasing the
price from the nominal value of $3.50 to $5.00 per gram increases the
net present value of the venture to $3.3 million at eight percent.
Decreasing the price to $2.00 per gram generates a loss of $0.8 million.
Similar changes in the net present value occur if the price is held
constant and the costs are changed by equivalent amounts.
Failure to obtain registration after R&D is completed results
in a net present value loss of $1.1 million at eight percent.
R&D expenditures for the base case totaled $1.2 million. It
is doubtful that actual company R&D expenditures would be significantly
lower, because most of the money has already been spent. However,
reducing the R&D expenditures by 20 percent increases the net present
value to $1.5 million at eight percent. Increasing the R&D expenditures
by 50 percent, to cover the possibility that more field and toxicology
tests will be required, decreases the project's profitability to $0.7 mil-
lion.
The base case assumed that no building capital would be re-
quired for the venture. Increasing the building capital needs to
$500,000 and $1,000,000 reduces the project's profitability to $0.8 mil-
lion and $0.4 million, respectively, at eight percent.
5. Probabilistic Analysis
The analysis of the gossyplure venture was conducted at a
modest level of detail, with highly aggregated variables. Even so,
203
-------
there were significant uncertainties on the values that those variables
might take, and as shown by the sensitivity studies, the ultimate profit-
ability of the venture is heavily dependent on the actual scenario of
events. In this section the implication of this uncertainty upon the
profitability of the gossyplure venture is reviewed.
Figure 24 summarizes the factors to be considered in this
probabilistic analysis. The company is faced with the decision of
whether to fund a million dollar R&D effort to develop gossyplure into
a commercial product. (While some of that R&D work has in fact already
taken place, this example assumes that the company must commit all of
the R&D funds immediately.) Even if the firm elects to undertake the
R&D effort, there is uncertainty about actual demonstration of efficacy
and safety. If it is not proven, the product will not be registered and
the company will lose its R&D expenditures.
If the product is registered, uncertainties remain as to the
ultimate market size and price. As discussed earlier, there are uncer-
tainties about the domestic market size, depending on whether the San
Joaquin Valley becomes infested with the pink bollworm. There are also
uncertainties about foreign markets, depending on whether foreign growers
elect to use the pheromone product. Since gossyplure has not been tested
in the market, there are also uncertainties about the price the company
will actually realize.
Novel and formulation patents may be obtained for gossyplure,
but if its market turns out to be large, there will be significant
incentive for petition to develop alternate formulations that circumvent
any such patents. Therefore, depending on the profitability of the
venture, there are uncertainties about the share of the total market
that the hypothetical company will capture.
Figures 25 to 27 summarize SRl's current evaluation of the
uncertainty associated with the various input factors. The uncertainties
are subjective probability assignments based on the information available
to SRI; as new information is developed the probability assignments and
associated findings may change. Many of the variables (e.g., domestic
market size) are continuous, but for reasons of simplicity they were
assigned a few discrete values, and the variables were encoded as dis-
crete distributions.
It should be noted that some of the probability assignments
are independent while others are dependent on other variables in the
analysis. For example, total market size and price were assumed to be
independent, but achievement of a registration is dependent on the amount
spent for R&D. Also, the market share that the company obtains depends
on the total market size and selling price.
204
-------
Decision to
Fund R&D
R&D Costs
Successful
Registration
Domestic
Market
Size
Foreign
Market
Size
Selling
Price
Market
Share
Fund
to
O
U1
Do Not Fund
Medium
Medium
Medium
Medium
Medium
J-Ou
FIGURE 24 SCHEMATIC OF GOSSYPLURE COMMERCIALIZATION DECISION
-------
R&D Costs
Successful Registration
Yes
0.85
$1.0 Million
p - 0.15
No
p = 0.15
Yes
0.7
$1.2 Million
0.50
No
0.3
Yes
0.6
$1.8 Million
0.35
No
0.4
FIGURE 25 PROBABILITY ASSIGNMENTS ON RESEARCH AND DEVELOPMENT
COSTS AND ATTAINMENT OF REGISTRATION FOR GOSSYPLURE
206
-------
Domestic Market
(Thousands of Acres)
Foreign Market
(Thousands of Acres)
Selling Price
($/Gram)
= 0.05
= 0.25
200
500
= 0.50
1,000
2.000
p = 0.15
p = 0.05
p = 0.25
$1.65
= 0.50
$2.00
$3.50
p = 0.20
$5.00
p = 0.05
= 0.90
50
p = 0.06
250
450
= 0.03
700
p = 0.01
FIGURE 26 PROBABILITY ASSIGNMENTS ON MARKET SIZE AND SELLING PRICE FOR GOSSYPLURE
-------
Percent of Market Captured
if Total Market Size >500,000
Acres and Sailing Price ^ $2.00/gm
Percent of Market Captured if
Total Market Size ^ 500,000 Acres
or Selling Price < $2.00/gm
0.5
100%
50%
0.3
25%
p - 0.2
0.2
100%
0.5
50%
20%
p » 0.3
FIGURE 27 PROBABILITY ASSIGNMENTS ON PERCENT OF GOSSYPLURE
MARKET CAPTURED
208
-------
0.40
0.32
0.28
0.24
0.20
0.16
0.12
0.08
0.04
ti—m
16 18
o L
-10
8
-6
-4
•2
0
2
6
4
8
10
20
22
24
26
12
14
NET PERCENT VALUE AT 8% INTEREST — millions of dollars
FIGURE 28 PROBABILITY DISTRIBUTION ON NET PRESENT VALUE, AT EIGHT PERCENT, OF GOSSYPLURE
VENTURE
-------
The different variables combine to produce 723 scenarios.
.The scenarios could be presented in a decision tree, such as that in
Figure 20 of Chapter VII, but its immense size precludes its presentation
in this report. The present value of each scenario was determined in the
same manner as the base case. Using the fundamentals of probability
theory the likelihood of each scenario was determined from the probability
assignments given in Figures 25 to 27.
Figure 28 shows the resultant probability distribution on pro-
fit at the eight percent discount rate. To facilitate the presentation,
the profit values were rounded to the nearest $2 million. The project
could lose as much as $8 million but there is only a 0.4 percent chance
of this loss. The venture could also earn as much as $25 million, but
there is only a 2 percent chance of this profit. On the average the
venture would return $2.7 million at eight percent.
Therefore, based on current state of information, the gossy-
plure venture appears modestly attractive. On the average the venture
would return $2.7 million, and there is a 20 percent chance that it
would return at least $2 million. The venture is modestly attractive
because there is an almost certain small market, and some potential for
a large market. Little plant 'capital is required, and the small market
is almost large enough to offset the R&D expenses, even if larger markets
never develop. However, the venture is not without risk; if registration
is not obtained, or if the markets and margins turn out to be very small,
it could lose money.
E. Selected References
The following is a list of some of the more pertinent literature
references relating to pheromones for 'pest control that were screened
during this report.
Bedard, W. D., R. M. Silverstein, and D. L. Wood, "Bark Beetle Pheromones,"
Science, Vol. 167, p. 1638 (1970).
Bedard, W. D., and D. L. Wood in "Pheromones," M. Birch ed., Frontiers
of Biology, Vol. 32, p. 441 (1974).
Berger, R. S., "Isolation, Identification, and Synthesis of the Sex
Attractant of the Cabbage Looper, Trichoplusia ni," Annals of the
Entomological Society of America, Vol. 59, p. 767 (1966).
Berger, R. S., Science, Vol. 177, p. 704 (1972).
Beroza, M., and E. F. Knipling, "Gypsy Moth Control with The Sex Attract-
ant Pheromones," Science, Vol. 177, p. 19 (1972)
210
-------
Bethell, R. S., et al., "Sex Pheromone Traps Determine Need for Codling
Moth Control in Apple and Pear Orchards, California Agriculture,
pp. 10-12, (May 19 72).
Bierl, B. A., M. Beroza, and C. W. Collier, Science, Vol. 170, p. 87
(1970).
-Brady, V. E., et al., Science, Vol. 171, p. 802 (1971).
Brown, T. M., and A. W. A. Brown, "Experimental Induction of Resistance
to a Juvenile Hormone Mimic," Journal of Economic Entomology,
Vol. 67, p. 799 (1974).
Burkholder, W. E., in "Pheromones," M. Birch, ed., Frontiers of Biology,
Vol. 32, p. 449 (1974).
Carlson, D. A., et al., Science, Vol. 174, No. 76, (1971).
Carlson, D. A., and M. Beroza, "Ovipo&ition by Brachymeria intermiedia
in the Presence of Disparlure," Environmental Entomology, Vol. 2,
p. 555 (1973).
Chow, Y. S., C. B. Wang, and L. R. Lin, "Identification of a Sex Phero-
mone of the Female Brown Dog Tick, Rhipicephalus sanguineus," Annals
of the Entomological Society of America, Vol. 68, p. 485 (1975)
Djerassi, C., C. Shih-Coleman, and J. Diekman, "Insect Control of the
Future: Operational and Policy Aspects," Science, Vol. 186,
pp. 596-607 (1974).
George, L. M., and L. M. McDonough, "Multiple Sex Pheromones of the
Codling Moth, Laspeyresia pomonella <%.)" Nature, Vol 239,
p. 109 (1972).
Hummel, H. E., et al., "Clarification of the Chemical Status of the
Pink Bollworm Sex Pheromone," Science, Vol. 181, p. 873 (1973).
Kaae, R. S., H. H. Shorey, and L. K. Gaston, "Pheromone Concentration
as a Mechanism for Reproductive Isolation Between Two Lepidopterous
Species," Science, Vol. 179, p. 487 (1973).
Kinzer, G. W., et al., "Bark Beetle Attractants Identification, Synthesis
and Field Bioassay of a New Compound Isolate from Dendroctonus,"
Nature, Vol. 221, 1969, p. 477 (1969).
Klassen, W., "Alternative Methods of Pest Control," Part 1, Agricultural
Programs of Hearings of the Subcommittee on Appropriations, House
of Representatives: 93rd Contress (February 1973).
211
-------
Kliewer, J. W. , et al., "Sex Pheromones and Mating Behavior of Culisita
inornata (Diptera:Culicidae), Annals of the Entomological Society
of America, Vol. 59, p. 530 (1966).
Klun, J. A., and T. A. Brindley, "Cis-II-tetradecenylacetate, a Sex
Stimulant of European Corn Borer," Journal of Economic Entomology,
Vol. 63, p. 779 (1970).
Langley, P. A., R. W. Pimley, and P. A. Carlson, "Sex Recognition
Pheromone in Tsetse Fly Glossina morsitans," Nature, Vol. 254,
p. 51 (1975).
Morgan, P. B., I. H. Gilbert, and R. L. Fye, "Evaluation of (Z)-0-
Tricosene for Attractancy for Musea domestica in the Field,"
Florida Entomology, Vol. 57, p. 136 (1974).
Mulla, M. S., H. Axelrod, and T. Ikeshoji, "Attractants for Synanthropic
Flies—Area-Wide Control of Hippelates collusor with Attractive
Baits, Journal of Economic Entomology, Vol. 67, p. 631 (1974).
Rodin, J. 0., et al., "Synthesis of Brevicomin, Principal Sex Attractant
of Western Pine Beetle," Journal of Chemical and Engineering Data,
Vol. 16, p. 380 (1971).
Roelofs, W. L. and H. Arn, "Sex Attractants of the Red-Banded Leaf
Roller Attractant," Nature, Vol. 219, p. 513 (1968).
Roelofs, W. L. and A. Comear, "Sex Pheromone Perception: Synergists
and Inhibitors for the Red-Banded Leaf Roller Attractant,"
Journal of Insect Physiology, Vol. 17, p. 435 (1971).
Roelofs, W. L., et al., "Sex Pheromone Trapping for Red-Banded Leaf
Roller Control: Theoretical and Actual," Journal of Economic
Entomology, Vol. 63, p. 1162 (1970).
Roelofs, W. L., and J. Tette, "Sex Pheromone of the Oblique-Banded
Leaf Roller," Nature, Vol. 226, p. 1172 (1970).
Sanders, C. J., in "Pheromones," M. Birch, ed., Frontiers of Biology.
Vol. 32, p. 436 (1974).
Schwalbe, C. P., et al., "Field Tests of Micro-Encapsulated Disparlure
for Suppression of Mating among Wild and Laboratory-Reared Gypsy
Moths,"" Environmental Entomology, Vol. 3, p. 589 (1974).
Shorey, H. H., and L. K. Gaston, in "Pheromones," M. Birch, ed.,
Frontiers of Biology, Vol. 32, p. 421 (1974).
Smith, R. G., G. E. Bateman, and G. D. Daves*, Jr., "Douglas-Fir Tussock
Moth: Sex Pheromone Identification and Synthesis," Science,
Vol. 188, p. 63 (1975).
212
-------
Tette, J. P., Development and Use of Sex Pheromortes for Pest Manage-
ment, a paper presented at 14th International Congress of
Entomology, Canberra, Australia (1972).
Trammel, K., in "Pheromones," M. Birch, ed., Frontiers of Biology,
Vol. 32, p. 416 (1974).
Tumlinson, J. H., et al., "Identification and Synthesis of the Four
Compounds Comprising the Boll Weevil Sex Attractant," Journal
of Organic Chemistry, Vol. 36, p. 2616 (1971).
Tumlinson, J. H., et al., "Boll Weevil Sex Attractant, Isolation Studies,"
Journal of Economic Entomology, Vol. 61, p. 470 (1968).
Weatherston, J., et al., "Studies of Physiologically Active Arthropod
Secretions: Sex Pheromone of the Eastern Spruce Budworm,
Choristoneura fumiferana (Lepidoptera:Tortricidae)Canadian
Entomology. Vol. 103, p. 1741 (1971).
Williams, C. M., "Third-Generation Pesticides," Scientific American,
Vol. 217, p. 13 (1967).
213
-------
IX "CONVENTIONAL" CHEMICAL PESTICIDES
A. Overview
To provide a framework for assessing the commercial potential of
new, innovative pesticide materials such as microbials and pheromones,
SRI reviewed the pattern of product development within the "conventional"
chemical pesticide industry. There were two specific aspects to this
investigation:
Interviews—A series of personal interviews were conducted with
R&D management employed by nine major U.S. pesticide companies. The
primary purpose of these interviews was to entertain an industry per-
spective on the directions of its research and the bases for its
research strategies, and to assess industry interest in new, innovative
pest control products, particularly those divergent from more tradi-
tional product lines.
Decision Analysis—A decision analysis investigation was made for
TH 6040 (Dimilin®) a "conventional" synthetic organic chemical with
proprietary characteristics but with an innovative mode of action—
interruption of the chitin formation process in certain insects.
The expected net present value of the TH 6040 venture was far more
attractive than those for Bacillus thuringiensis (B.t.), Heliothis spp.
NPV, and gossyplure; similarly, there was substantial risk involved.
The relevant opinions of pesticide company personnel included the
following:
• The classification of pesticide materials established by
the EPA compendium should be adhered to. Further sub-
division, if any, should be by mode of action—as for
pharmaceutical chemicals—or by relative mammalian tox-
icity.
• No special rules for registration or subsidies should be
established for any products, including the so-called
"new generation" pesticides. The consensus was that all
products or methods subject to regulation or support
should be treated according to the merits of the indi-
vidual situation, rather than by artificial product
classes.
215
-------
• Government should either discontinue research leading to
new products that would receive exclusive patents if
developed by industry, or develop a method for exploit-
ing public discoveries through selective licensing.
• A minimum patent life following registration of new
proprietary pesticides should be ensured.
• Administrative procedures should be sharpened and firm
guidelines provided and followed in registration proce-
dures in order to expedite these processes.
B. Industry Survey
A series of personal interviews were conducted with senior staff
in nine major pesticide companies to identify and evaluate the attitudes,
policies, and strategies of major U.S. firms that have contributed to
the discovery, development, production, and marketing of many of the
pesticides currently in use. This input was considered important
because the new products that will emerge from the chemical portion of
the industry during the next decade are either already under develop-
ment or will have characteristics shaped by the policymakers now estab-
lishing strategies in these companies (since development through
registration and market introduction typically takes seven to ten years).
The names and affiliations of individuals interviewed are included in
Appendix A.
Several distinct patterns emerged from this series of interviews
relating to (1) properties essential to the commercial success of new
pesticides, (2) approaches used to arrive at such successes, and
(3) difficulties which must be overcome in this process.
1. General Characteristics of Successful Pesticides
To achieve commercial success a new pesticide must fill
a need, recognized or unrecognized, so that a market can be developed.
It must adequately satisfy that need in order to compete with other
products with parallel market potential. It must be proven safe to
exposed persons and the environment so that it can be registered.
Finally, long run profits from sales must recover R&D costs, capital
costs, and operating expenses, and must provide a return to stock-
holders competitive with alternative uses of capital. Thus, a company
that wants to develop- a new pesticide must identify a need, find a
product which answers that need, determine whether that product is
safe, and finally decide whether it will achieve a return on investment
consistent with corporate goals.
216
-------
2. Industry Participation In Innovative Pesticides
With the exception of a few innovative synthetic chemical
pesticides such as ovicides, none of the nine companies was at the time
interviewed marketing innovative pesticide products. (There was a minor
exception, one company distributing B.t. under other producers' brand
names.) Furthermore, no company indicated any such products were under
development that would reach the marketplace before 1980. One company,
a previous producer of B. t. , was following this product's commercial
growth and could again become a producer and marketer.
However, all nine companies maintain a surveillance of research
efforts in these areas. Eight have engaged in research on such products,
four of which claimed to have discontinued this research.
3. Specificity of Pesticide Action
Specificity of action is often cited as a desired property of
a candidate pesticide. However, unless the specific activity is strong
and involves a major pest of a major crop, the market may be too small
to justify development costs. Growers have preferred broad-spectrum
products in order to avoid the costs of using several materials or
methods against multiple pest infestations. Problems of compatability
are also a consideration, when more than one formulated product is
combined in a spray tank. Furthermore, the chemical load of several
pesticides can possibly be more detrimental to the environment than
a single broad-spectrum chemical, assuming equal potency.
Some observers consider the importance of specificity to be
related to the conditions of use. For example, specificity is rela-
tively less important for soil insect control than for mosquito control,
where risks to nontarget species and the environment are generally
greater.
Specificity can also be a function of dosage. Many broad-
spectrum products show substantial specificity at low dosages, while
some specific products may exhibit broad-spectrum activity at high
dosages.
Industry claims that herbicides should have a broad-spectrum
effect since most crops have an infestation potential for up to eight
to ten weed species, and economic results can be achieved only if most
or all of these weeds are controlled. Thus, for herbicides in partic-
ular, specificity may not be advantageous, except for major monospecies
problems such as alligator weed in irrigation ditches.
217
-------
4. Pest Resistance
Industry leaders generally maintained that most innovative
pesticides are not likely to be more immune to the development of pest
resistance than "conventional" pesticide products.
5. Integrated Pest Management (IPM)
Industry members generally stated that IPM has merit in con-
cept, but that it may be receiving too much attention relative to
other worthy efforts. IPM in greater or lesser degree has been prac-
ticed for many years, especially in cotton culture, before becoming
recently organized and formalized. In the past, IPM involved extension
workers, large growers, pesticide industry personnel, and others. In
some situations, the concept appears logical and economically sound
to most observers. However, many industry members feel that the
number and interaction of variables in, most areas under study is so
complex that major difficulties will be encountered with training
needed professionals, with grower acceptance, and with the economic
effectiveness of widespread IPM adoption.
It is acknowledged that some state and federal extension work-
ers in IPM pilot programs feel that the pesticide industry is resistant
to IPM programs. On the other hand, many industry leaders feel that
IPM is a new catch phrase for an old practice, and that its objective
of reduced pesticide chemical use is not inconsistent with general
industry objectives of finding the most effective materials that can
be used infrequently and at low dosage rates for the economical control
of pests. In this context, industry members claim that the above
elements of good stewardship for pesticide products are already being
practiced by major companies (i.e., industry is aware of farmers'
growing consciousness of cost-effectiveness).
One goal of IPM is to reduce pesticide use to the minimums
required to maintain threshold levels. In this context, industry is
particularly concerned with any actions that would lead to a departure
from label directions for rates of application of registered products.
Industry feels that this could damage label integrity and contribute to
a diffusion and confusion of product liability when losses are claimed
by product users.
6. Research and Registration
Most of the industry leaders interviewed confirmed the reported
pattern of $7 million to $10 million per chemical for research costs,
and the seven- to ten-year span between discovery and commercial sales.
None expected that nonconventional products would be likely to depart
from this pattern. Most felt that total development costs for microbial
agents would be higher than for synthetic chemicals, particularly if
218
-------
costs for the necessary staff and equipment which most chemical companies
do not have are included. However, the production of microbial agents
by fermentation methods—where possible—could mean plant capital and
operating costs lower than those for chemical products that involve up
to ten sequential process steps.
As previously described, the development of pesticides with
narrow spectra can be discouraged by high R&D costs relative to limited
markets. The question has arisen about the possiblity of less stringent
registration requirements for these materials. Industry leaders oppose
any reduction of requirements on such arbitrary bases. There was wide-
spread feeling that there are so many unknowns associated with micro-
bials, especially viruses, that stringent registration requirements
should be retained until questions of safety are resolved. Concern over
the possible infectivity of viral agents to higher order animal species
was expressed by several industry members.
In the interviews, SRI attempted to assess the effect that
reducing or eliminating current EPA registration requirements might
have on R&D costs, assuming that companies would maintain their own
high-level standards for product safety. Individual estimates ranged
widely, suggesting that companies vary in their internal standards, or
that individual responses were not fully deliberated. Respondents
generally anticipated an average 15 percent reduction in cost and an
average 25 percent reduction in time required to commercialize a product
if there were no EPA requirements. They emphasized that the potential
time savings were important because of the impact on cash flows and the
early recovery of R&D costs, particularly if potential markets are large,
as discussed in subsection II-C-3, "Research and Development Costs,"
and II-C-5, "Capital Costs."
There was no indication that the companies contacted have
reduced their R&D efforts despite increased requirements for registration,
inflation, and increased costs associated with basic identification
activity for chemical pesticides. (Several contacts expressed concern
that increasing investment requirements were forcing smaller firms out
of new pesticide R&D.) Individuals expected new improved pesticide
products to continue emerging from the laboratories of major pesticide
companies. This implies that a continuous flow of new, environmentally
acceptable, and economically attractive products will substitute for
products that may be discontinued because of regulatory action or loss of
market position for other reasons. There may be a short-term gap in the
availability of controls for cutworm and wireworm in corn in the midwest,
but according to industry sources this problem and the problem of the
forest insects formerly controlled by DDT seem to be the only two sig-
nificant gaps. Most industry members accept the drive to cancel the
chlorinated hydrocarbon insecticides with demonstrated bioaccumulation
characteristics as an expected course of action by EPA. They object,
however, to inferences which lead the public to believe that the dis-
advantages of these chemicals may apply to pesticides generally, or to
219
-------
major groups or other classes of pesticides. Some industry members feel
that EPA is placing too much emphasis on relative cost and not enough on
efficacy as it evaluates and recommends substitutes for cancelled products.
Most industry members feel that EPA requirements for data on
the environmental fate of pesticide chemicals are excessive and unrea-
sonably applied to all products and uses.
7. Government Subsidies .
Since many innovative potential pesticides are narrow- rather
than broad-spectrum products, their development through registration for
small markets represents a problem. Since industry prefers to develop
broad-spectrum products, an overview of current industry attitudes and
policies with respect to the development and registration of pesticides
for minor uses was of interest.
In general, industry supports the government subsidy of pesticide
registration for minor uses. Most of the companies interviewed cooper-
ate -in the IR-4 Program, sponsored by the USDA to register products for
minor use in agriculture. This government subsidy seems to be both working
well and well received by industry. Companies acquire some data in
secondary screening which is a useful start for demonstrating efficacy
for registration; state and federal experiment stations acquire mo,re
useful data. Residue studies, also required for registration, are
conducted by Oregon State University. For proprietary products, the
company controlling the patent usually prepares and submits a petition
for registration based on both its own data and IR-4 data. The program
is coordinated by USDA through a small staff at Rutgers University.
However, while this program appears to be functioning well,
most companies feel that EPA should further ease registration require-
ments for minor crops and minor pests. One problem companies face in
this area is their potential product liability. Minor crops are usually
of high value, but represent a trivial market for pesticides; thus,
crop injury can lead to losses far exceeding potential chemical profits.
This problem might be resolved through some form of subsidy such as crop
insurance. Industry feels that grower associations and food processors
should assume more responsibility in this area, perhaps by increased
participation in new product registration and in the control of product
selection and application on minor crops.
Except for pesticides for minor crops, industry generally
considers government subsidy of new product discovery and development
unnecessary and sometimes harmful. Interviewed industry members felt
that government would make a greater contribution by focusing its
activities on crop production and basic research related to the biology
of pests and plants, rather than engaging in synthesis and screening
designed to discover patentable products. The National Agricultural
Chemicals Association placed a similar policy statement on record several
years ago and its individual members continue to express this attitude.
220
-------
Industry members perceive the pesticide industry as over-
regulated, and view potential additional subsidies as leading to further
regulation. Most individuals felt strongly that government should not
conduct research directed toward output that might be patentable if
developed in the private sector; industry tends to avoid work in fields
where public service patents, which cannot be exclusively licensed, are
likely to emerge. For example, the USDA recently announced the discov-
ery of two juvenile hormone mirafcs that could be incorporated in live-
stock and poultry feed for fly control; as commercial products they
would compete with Altosid® and TH 6040. Although companies have dis-
cussed the commercial development of these new products with USDA
personnel, apparently no commercial agreements have yet been reached.
8. Patentability
The legal monopoly provided by an issued patent provides one
of the strongest incentives for industry to develop coremercial products.
This becomes increasingly important as research costs rise and the time
required to achieve a registration increase. All companies contacted
considered patent coverage very important and, in many cases, essential
to pesticide development, as discussed in subsection II-C-1, "Market
Size."
Naturally occurring substances and microbes are not patent-
able as new compositions since they exist in nature, although new syn-
thetic mimics and new mutants of microbes may be patentable. Therefore,
potential pesticides based on these materials can be protected only by
method of manufacture or formulation patents. Patentability has also
increased in importance because of the new pesticides law which permits
a secondary registrant to have access, for a fee, to selected data sub-
mitted by the primary registrant.
In addition, because of the increased time required to regis-
ter new products, many industry members feel that the life of issued
pesticide patents should begin with the first registration.
9. Nomenclature
Industry contacts objected to the terms "third generation"
and "new generation" pesticides; they considered this nomenclature
arbitrary, misleading, and unwarrantedly implying superior performance
and improved safety for all products under its umbrella. Many indi-
viduals felt that this nomenclature may be contributing to improper
legislative and regulatory actions.
The term "third generation" pesticides was apparently first
coined (Williams, 1967) to describe a juvenile hormone mimic which
would be potentially useful as an insecticide. This chemical was
expected to differ from "second generation" products such as DDT—the
well known broad-spectrum insecticide to which about 100 species of
221
-------
insects have developed resistance—by being target species specific and
immune to the development of resistance.
Other terms have subsequently emerged to describe certain
"classes" of potential pesticide products or methods; included in this
nomenclature are terms such as "new generation," "biologically inte-
grated," "biorational," and "unconventional." These terms have expanded
Williams' initial narrow interpretation of a "third generation" insec-
ticide to encompass a wider variety of pest control methods and to
include some concept of product origin—for example, whether it occurs
naturally, or is of synthetic origin. Industry feels that the evolution
of such nomenclature has contributed to a serious semantic problem which
can be misleading to all concerned, including policymakers.
To evaluate this concern the next subsections review both the
approaches to discovering new pesticides, and the evolution of some mod-
ern pesticides. The last subsection presents an alternate approach to
nomenclature.
Approaches to Discovering New Pesticides—There are three
general approaches to the conception, synthesis, and characterization of
new pesticide chemicals.
(1) Random Screening Approach: This approach is an
essentially random biological screening of synthetic
or naturally occurring material from available sources.
It is sometimes referred to as a nondirected approach.
(2) Analog or Directed Approach: This approach consists
of synthesizing analogs of natural or synthetic chem-
icals that have a demonstrated biological activity of
interest. Analogs of natural products are often called
either mimics or antagonists, depending on whether the
biological activity exhibited by the natural product
is reinforced or inhibited by the synthesized chemical.
(3) Biochemically Rational Approach: This is investigation
based on knowledge or theory of vitally vulnerable life
processes of the pest, definition of the physical or
chemical characteristics or dimensions of the organs
and cells involved, and delineation of the pertinent
biochemical steps in the life processes.
Approaches (2) and (3) can be combined to design chemical
structures with particular kinds of biological activity.
Almost all breakthrough discoveries have resulted from random
screening, but a very high percentage of current commercial pesticides
emerged from some variant of the analog approach. The interview survey
reviealed a strong trend among the companies contacted toward biochemically
rational research strategies, sometimes involving basic biochemistry and
interactions at the molecular level.
222
-------
Nomenclature and Selected Pesticide Evolution—Many of today's
commercial insecticides act by inhibiting the enzyme acetylcholin-
esterase. Eserine (physotigmine) is a naturally occurring carbamate
which has long been known to cause constriction of the pupil, a test
animal response typical of organophosphate and carbamate insecticides.
In 1930 eserine was shown to inhibit acetylcholinesterase. Some 20 years
later, random screening revealed that 1-naphthyl methyl carbamate (carbaryl)
had strong insecticidal properties. Consideration of its chemical struc-
ture relative to eserine suggested that carbaryl might function as an
insecticide by inhibiting acetylcholinesterase. This hypothesis was
readily confirmed, and many other commercial carbamate insecticides for
agricultural and home use were developed through the analog approach.
Since the carbamates are analogs or mimics of a naturally occurring
product, they could be considered "third generation" products, although
they are usually identified as "second generation" or "conventional"
pesticide chemicals.
Synthetic pyrethroid products have a similar history. About
50 years ago the chemical structures of pyrethrin I and pyrethrin II
were elucidated by two Swiss chemists. Ten years later, two additional
insecticidal esters in pyrethrum flowers were identified and named
cinerin I and cinerin II. Using the analog approach with cinerin I as
a model, government scientists in the 1940s synthesized a mimic which
they named allethrin. This product was controlled with public service
patents in the United States and also with patents owned by the inven-
tors in many other countries, where it was more successful commercially
than in the United States.
In the 1960s, a new series of pyrethroids were announced by
a government scientist in England which eventually led to a product
called NRDC 143. Before this product, pyrethroids were largely used
for household fly control since they were unstable to light, moisture,
and air, and of low toxicity to mammals. Pyrethroids are believed to
act similarly to DDT, interfering with impulse transmission in the
nervous systems of insects. The more recently discovered NRDC 143 not
only has very high inherent activity against a variety of insects
(especially Lepidoptera), but also, in contrast to earlier pyrethroids,
exhibits strong residual effects. This property is expected to extend
the utility of pyrethroids to major use against lepidopterous larvae
affecting a number of important agricultural crops.
Although NRDC 143 was discovered by a government employee in
England, it is licensed to selected companies on a semi-exclusive basis,
thereby assuring wide market exposure and providing incentive for the
investment of development funds by the licensors. As a matter of
interest, pyrethrins are usually classified as "first generation"
pesticides, and the new synthetic mimic appears to have the same mode
of action as the naturally occurring products.
" Although products that interfere with basic insect growth
processes (such as Altosid® and TH 6040) inspired the term "third genera-
tion," plant growth regulators have been recognized for decades. Five
223
-------
types of plant growth regulators (hormones) are presently known. These
are auxines, gibberellins, ethylene, cytokinins, and abscislc acid; others
will probably be discovered in the future. Interaction among these five
types of hormones leads to a very large number of modes for affecting
a plant's growth. Also, a given hormone may influence more than one
cell process, and by different degrees. By analogy, it seems logical
to anticipate that basic biochemical research may lead to expanded
knowledge regarding the biochemistry of insects and fungi similar to
that which has occurred for plants other than fungi. In addition, there
are potential discoveries relating different classes of pesticides.
For example, a chemical which functions as an inhibitor of chitin syn-
thesis might be both insecticidal and fungicidal, since most insects
and some fungi contain chitin as a structural building block.
Alternate Nomenclature—This review and recent contacts with
industry leaders indicate that historical patterns of chemical pesticide
discovery will probably continue. Industry is concerned that this pat-
tern might be unnecessarily impeded by misleading or imprecise nomencla-
ture.
If alternative nomenclature is required to better classify
pesticide developments, the following system relating to mode of action
(Corbett, 1974) may be more suitable:
• Pesticides that interfere with respiration.
• Herbicides that interfere with photosynthesis.
• Insecticides that inhibit acetylcholinesterase.
• Neuroactive insecticides.
• Compounds that interfere with plant growth.
• Pesticides that are believed to inhibit biosynthetic
reactions.
• Pesticides whose mode of action is unknown or non-
specific.
• Compounds that interfere with insect growth and develop-
ment (category added by SRI).
C. Decision Analysis of a Possible Innovative Chemical Pesticide Venture
1. Introduction
To examine the commercial feasibility of developing an inno-
vative synthetic organic chemical pesticide, this section presents a
preliminary analysis of the decision to develop and market TH 6040
(Dimilin®), a chitin inhibitor. The analysis is based on nonproprietary
data and should therefore not be considered a final assessment of the
commercial feasibility of this product; it is intended to illustrate
the factors that influence the decision to develop a product.
224
-------
TH 6040 is a rather broad-spectrum pesticide that disrupts
the synthesis of chitin development in immature insects. It is primarily
effective on foliar feeding insects, since it must be ingested to take
effect. TH 6040 has been found to have activity against mosquitos,
flies, soybean looper, velvetbean caterpillar, cotton bollworm, alfalfa
weevil, beet armyworm, and tomato fruitworm. A possible' disadvantage
of the product is that feeding may continue for several days after
application, since all instars survive until their next molt.
The analysis begins by defining a base case as a point of
reference for evaluating changes in underlying assumptions. Subsequent
sensitivity studies determine the variables that most critically affect
the commercialization decision. Finally, a probabilistic analysis
considers the uncertainty of the venture's profitabilty.
2. Background on the Product
Review
TH 6040 is one of a large group of l-(2,6-disubstituted
benzoyl)-3-phenylureas discovered by Philips-Duphar B.V., The Netherlands.
It has been determined that TH 6040 interferes with the growth processes
of insects by inhibiting chitin synthesis and deposition in the endo-
cuticle; this interference leads to insect mortality by a rupturing
of a new delicate, malformed cuticle, or by starvation. Although TH 6040
has a rather wide range of insecticidal activity, sucking insects escape
its effect because primary activity results from ingestion. TH 6040 is
currently being developed for use as a pesticide in the United States
under license to the Thompson-Hayward Chemical Co. subsidiary of The
North American Philips Corporation.
• Chemical Name:
l-(4-chlorophenyl)-3-(2,4-difluorobenzoyl)-urea
• Structural Diagram:
CI
C-NH-C-NH
• Empirical Formula and Molecular Weight:
C14H9N2°2F2C1, M-W- 310*69
• Trade Names, Synonyms, Other Code Names:
Dimilin®; TH 6040; PH 604a; ENT 29054; OMS 1804; Largon
225
-------
• Stability;
1. In water: stable at pH 2-8, unstable at pH 12
In soil: 1/2 life = approximately 2 weeks.
2. In pure water: % deterioration
pH 6 or 8 1 day <2%
pH 12 1 day 12%
pH 12 7 days 76%
3. Thermal: less than 2% deterioration at 50°C for 1
week or 100°C for one day.
4. Photo: no decomposition in crystalline form on glass
plate when exposed to an MLU 300 W lamp at 15 cm for
24 hours. In solution at 0.13 ppm, 13% deterioration
when exposed to Xenon lamp for 92 hours.
• Melting point: 239°C
• Organic solubility: in g/100 ml of solvent
methyl pyrolidone 28
dimethyl sulfoxide 12
dimethyl formamide 12
acetone, cyclohexanone, dioxane, ethanol, isophorone,
methylene chloride, and xylene are all less than 2
• Volatility: essentially nil at 20°C
• Formulations:
Granular 0.5%
Wettable powder 25%
Liquid suspension 2 lb/gal
History of Development
In 1964, Philips-Duphar B.V., The Netherlands, set up a
directed program to develop chitin inhibitors into effective pesticides.
This program was successful in its early recognition of mode of action,
but was not otherwise productive and therefore abandoned in 1969. The
screen that had been developed was continued, however, and TH 6040's
insecticidal properties (it was originally designed to be an herbicide)
were subsequently discovered after a "routine" run-through. Further
testing indicated it was a product with promise and it was licensed for
development in the United States to the Thompson-Hayward Chemical Co.
subsidiary of the North American Philips Corporation.*
*Directly prior to the publication of this report, the Philips group in
The Netherlands was discussing the sale of nearly all its Philips-Duphar
pharmaceutical and pesticide operations with Schering AG, West Germany.
One published announcement of these discussions indicated that the
Amsterdam plant producing Dimilin®was being left out of the discussions
"for the time being" as its profit potential was "still being evaluated."
226
-------
Composition of matter and use patents were issued in the
United States in 1973 and other patents are pending. There is no
registration domestically but a provisional registration exists in
Europe; conversely, more field work has been conducted domestically
than overseas. Thompson-Hayward has applied for a temporary use per-
mit for mosquito control and expects to obtain an acceptance shortly;
no requests for tolerances have yet been made.* Reported results of
research completed or underway designed to comply with registration
requirements follow:
• A mammalian hormone test conducted by EPA was
found to be "negative;"
• Half-life in the soil is 2-3 weeks;
• There is several days' persistence in water;
• It does not appear to bio-magnify in food chains;
• It is not translocated in the plant, but has good
persistence on leaf surfaces;
• Plant and animal metabolism studies are in prog-
ress; all two-year feeding trials are to be com-
pleted by 1976.
Production
Specific details on the actual production route to TH 6040
are not publicly available. All output is by Philips-Duphar in The
Netherlands, and there appears to be some overlap in the process and/
or equipment used to manufacture dichlobenil (2,6-dichlorbenzo-nitrile)
at the same location. Presumably there is a common intermediate in
the production of dichlobenil and the 2,6-difluorobenzoyl fragment
of TH 6040.
Markets
The chitin inhibition mode of action of TH 6040 basically
defines the scope of its potential markets (it is a growth interruptor
but not a juvenile hormone). It is claimed to give "excellent control"
against foliar feeding insects after ingestion, but all instars sur-
vive until their next molt. Thus, considerable feeding can occur after
application, which may be a negative factor for some crops. In addi-
tion, TH 6040 provides no control against adult or sucking insects; if
these pests are a problem, TH 6040 must be applied in conjunction with
other controls.
*Directly prior to the publication of this report, Dimilin® had been
registered for restricted use on gypsy moth in the northeastern United
States, and had been cleared for experimental use on mosquitos, boll
weevil, and velvetbean caterpillar.
227
-------
Thompson-Hayward has indicated the following initial target
markets for TH 6040.
• Mosquitos
• Flies, including house, stable, farm, and face
flies
• Soybeans, primarily for the control of foliar
feeding Lepidoptera
• Peanuts
• Forest insects.
Although the product has been tested for control of other
insects and as a possible ovicide, the company's initial efforts have
been directed toward the above. Soybeans appear to be the most critical
market; Thompson-Hayward has stated that without a market of this size,
product development for other markets would be unlikely.
-*-• Soybeans. Insects feed on both the pods and foliage of
soybeans, and although these plants can tolerate consid-
erable foliar feeding with little loss of yield, at
certain stages of plant growth there can be reductions
in both yield quantity and quality. Historically, soy-
beans have been a relatively small market for insecticides,
but use of chemical insect controls has recently been
growing at a substantial rate as acreages have increased,
as the crop has been introduced into new growing regions,
and as the per-acre value of the commodity has escalated.
As reported in USDA pesticide surveys, only about 7%
(2.7 million acres) of total domestic soybean acreage
was treated with insecticides in 1966; by 1971, treated
acreage had increased to about 12% (5.1 million acres)
of the total. According to a recent SRI Chemical
Economics Handbook report on insecticides, about 2.5 mil-
lion pounds of carbamates, 1.7 million pounds of
chlorinated hydrocarbons, and 3.0 million pounds of
organophosphates were used for insect pest control in
soybeans in 1974. Many assume that insecticides will
continue to be used in increasing quantities on soybeans,
particularly if bean acreages continue to expand, and/or
if the per-acre values for this commodity escalate further,
thereby warranting increased protection.
The most widespread insect pests affecting soybeans have
been the Mexican bean beetle, Epilachna varivestis
(Mulsant); the green stink bug, Acrosternum hilare (Say);
and the bean leaf beetle, Cerotoma trifurcata (Foster).
Less widespread are the green cloverworm, Plathypena
scabra (Fabricius), and wireworms, Limonius dibitans
228
-------
(LeConte); there are also some 15 to 20 minor insect
and mite pests. In addition, as soybeans have increased
in prominence as a cash crop in southerly states, the
looper complex (including the soybean looper) and the
velvetbean caterpillar (Anticarsia gemmatalis) have
evolved as increasingly serious problems. It is toward
these latter pests the Thompson-Hayward appears to be
targeting the introduction of TH 6040 as a control agent.
(Current control of the looper complex and the velvetbean
caterpillar is generally achieved with carbaryl, methyl
parathion, parathion, and methomyl; the recent range in
per-acre material control costs for these products has
been $3 to $7.)
The company is not actively pursuing development of
TH 6040 for control of the other soybean pests mentioned
although limited tests have indicated that the product
controls the Mexican bean beetle at an active ingredient
application rate of 0.5 pounds per acre.
Other Markets. Mosquito larvae and pupae are susceptible
to TH 6040 in all stages (as opposed to Altosid® where
only the fourth larval instar is susceptible). There
did not appear to be any lack of susceptibility by
Culex pipiens quinquefasciatus larvae to TH 6040 after
twenty generations were treated with various concentra-
tions. Expected application rates would be relatively
low at about 0.025 pounds of active ingredient per acre.
Much of the mosquito market potentially available to
TH 6040 may be in areas where resistance has precluded
the use of many existing products (e.g., California).
Therefore, demand in these cases would not be in heavy
direct competition with a present array of control
products. In addition, the use of larval insecticides on
mosquitos in the South has not been regularly practiced.
Fly larvae are well controlled by TH 6040 in both feed-
through and topical applications, and Thompson-Hayward
has been considering a registration for topical appli-
cation fly control in feedlots.
Other markets that have been evaluated for TH 6040 include
forest insects (e.g., tussock moth, gypsy moth, spruce
budworm), peanut insects, rice water weevil, codling
moth, pear psylla, boll weevil, cotton leaf perforator,
and pink bollworm.
229
-------
3. Base Case
Market Size—The variety of pests susceptible to TH 6040
provides the product with a relatively large market size. Table 32
presents the principal crops and pests believed to be potential uses
for TH 6040. Soybeans offer the largest potential acreage, in both
domestic and foreign markets. Forested areas also offer large potential
acreage, particularly for budworm control in Canada. Cotton offers
important potential, but extensive field tests must still be performed
to determine the efficacy and application rates for the various pests.
The maximum potential treatable acreages for all crops total 22.1 million
and 42.6 million acres for domestic and foreign markets, respectively.*
Also given in the table are the application rates and percent-
age growth rates estimated for various markets. Most of these markets
are expected to remain relatively, constant over time, but the number of
treated soybean acres should increase as the demand for protein grows.
It was assumed that the gypsy moth market would probably increase as that
pest continues to spread geographically, and that the mosquito market
would shrink if that pest becomes better controlled. The total potential
market size in pounds of active ingredient is estimated from the acreages
and application rates.
A number of pesticides are currently used on the crop-pest
complexes listed in Table 32, so there is little likelihood that TH 6040
would capture 100 percent of any of these markets. To estimate the
share of each that TH 6040 will ultimately obtain, its relative advantages
and disadvantages were compared to the currently used controls. Table 33
summarizes the principal advantages and disadvantages of TH 6040 from
the standpoint of the soybean grower, relative to the principal pesti-
cides used for soybean looper and velvetbean caterpillar control.
TH 6040's principal advantages were considered to be probable
safety to humans, persistence, and a possibly superior ecological image.
On the other hand, TH 6040 can be slow acting, thereby permitting several
days of further damage potential after application; is ineffective on
certain pests, such as stink bugs and leaf beetles; and may be higher
priced than the alternative controls. All products appear about equally
efficacious on soybean loopers and velvetbean caterpillars; all appear
compatible for application with other pesticides; and if TH 6040 is
registered, there should be no availability or distribution problems.
Based on these relative advantages and disadvantages, SRI
estimated that TH 6040 would ultimately capture about eighteen and ten
percent of the domestic and foreign soybean markets, respectively. The
low foreign market estimate assumed lower marketing activity.
*The decision to include foreign market potential in this base case
derives from both the proprietary nature of the product and its
European development.
230
-------
Table 32
BASE CASE MAXIMUM POTENTIAL MARKET SIZE FOR TH 6040
Domestic Market
N5
CO
Total Estimated*
Treatable Growth Rate Peak Market
Foreign Market
Total Estimated
Treatable Grovth Sate Peak Market
Total Acres
Acres
Crop
Pest(a)
(percent
Share
Total Acres
Acres
(percent
Share
Market Size at First
Year of Registration,
Applications Pounds per Assuming Mo Growth*
(millions) (millions) per year) (percent) (millions) (millions) per year) (percent) per Season Application (millions of pounds)
Soybean Soybean looperj
Velvetbean j
caterpillar '
N/A Mosquito
N/A Fly
Forests Tussock moth
Gypsy moth
Spruce budvorm
Tobacco Budvorm
Pears Pear psylla
Cotton Bollvorm]
Budvorm 1
55.0
n.a.
n.a.
n.a.
o.a.
0.85
0.2
11.5
1.0
0.35
0.2
1.0
3.5
0.85
0.2
5.0
1.01
-1.0
0.0
0.0
2.0
0,0
0.0
0.0
0.0
181
40
10
10
20
20
5
20
40
n.a.
n.a.
n.a.
n.a.
91
70
10.0
1.0
0.35
0.2
1.0
20.0
0.5
0.2
7.5
2.OX
-1.0
0.0
0.0
2.0
0.0
0.0
0.0
0.0
lot
20
1
10
20
20
2
10
1.5
4.0
1.0
1.0
1.0
1.0
5.0
4.0
4.0
0.133
0.025
1.0
0.1
0.1
0.1
0.5
0.25
0.5
0.56
0.06
0.04
0.004
0.04
0.47
0.13
0.24
-1.26
n.a ¦ Not applicable.
*
This column is not sumned because not all crops are registered in the same year.
-------
Table 33
ADVANTAGES AND DISADVANTAGES OF TH 6040
COMPARED WITH OTHER PRINCIPAL PESTICIDES*
USED FOR SOYBEAN LOOPER AND VELVETBEAN CATERPILLAR
Advantage
Persistence
Safe for humans
Possible ecological
image
No Significant
Difference
Efficacy
Compatibility
Availability
Disadvantage
Slow acting
Price
Smaller spectrum
of target pests
*
Carbaryl, methyl parathion, phorate.
Table 32 includes base case estimates for market shares in
the other applications. Briefly, the mosquito market share was set at
40 percent because of the relative lack of efficacious competitors.
Shares in the other markets were set at lower levels because there were
more competitive controls, because TH 6040 is slow acting, and because
of potentially high delivery costs. (Different levels of market pene-
tration are considered later in the sensitivity studies.) It was
assumed that each market will develop linearly to peak level over five
years; that it will stay at peak level for six years; and that it will
diminish to zero in the following nine years.
The total market size in any one year was calculated by
multiplying the potential market size by the annual market share for
a given use and then summing. (Although first-year market sizes for
each crop are listed in Table 32, it is not appropriate to add them
because not all crops are registered in the same year.) A separate
calculation shows that the total market will ultimately reach a nominal
size of 3.8 million pounds of active ingredient per year (taking into
account the growth rates given in Table 32).
Production Costs—Detailed production cost data were not
available to SRI. Therefore, base case production costs were estimated
as shown in Table 34, based on general knowledge of the raw materials
and processes that will probably be used to produce TH 6040. These
estimates were also based on more detailed knowledge of somewhat similar
chemical processes, but they could be in error because of the lack of
232
-------
Table 34
Cost Category
Direct Operating Costs
BASE CASE PRODUCTION COST ESTIMATE FOR TH 6040
(Annual Capacity: One Million Pounds)
Basis or Unit Cost
Total Costs
Cents
per Pound
Thousands
of Dollars
per year
NJ
W
Labor
Operating
Maintenance
Control laboratory
Total labor
Materials
Raw and process
p-chloroaniline
Cyclohexanone
Chlorine
Potassium fluoride
Sodium cyanide
Ferric chloride
Phosgene
Solvent
Maintenance
Operating
Total materials
Utilities
Total Direct Operating Costs
Fixed Costs
Plant overhead
Taxes and insurance
G&A, sales, research
Depreciation
Interest on working capital
2.4 men/shift, $8.00 man-hr
3Z/yr of battery limits cost
20Z of operating labor
80c/lb
38c/lb
5c/lb
65c/lb
53c/lb
IOC/lb
19c/lb
n.a.
3Z/yr of battery limits cost
10Z of operating labor
Assume 2c/lb
80Z of total labor
2.7Z/yr of fixed capital
Assume 50c/lb
132/yr of fixed capital
8Z/yr on $600,000
28.00c
6.00
3.60
39.60$
42.67c
59.23
3.45
30.00
16.45
0.80
9.50
25.00
6.00
2.80
195.90c
2.00
237.50c
31.70c
6.00
50.00
30.00
4.80
280
60
56
$ 396
427
592
35
300
165
8
95
250
60
28
$1,961
20
$2,375
317
60
500
300
48
Total Production Cost
360.00c
$3,603
-------
available data on TH 6040's exact process steps. The fixed costs are
given for one year's production, assuming a one million pound per year
plant. For the base case, production and formulation costs total
$3.60 per pound.
Capital costs for a one million pound per year plant were
estimated to be $2.25 million, and it was assumed that it would take
two years to build. It was also assumed that if significantly more than
one million pounds of TH 6040 are required annually, additional one
million pound per year plants will be built in different locations.
Selling Price—The final selling price is difficult to estimate,
since the product has not yet been marketed. However, based on the
price of other proprietary chemical pesticides that have recently been
introduced into the market (e.g., methomyl), the base case selling
price was established at $16.00 per pound of active ingredient. At this
price, it will cost about $3.00 per acre per season to treat soybeans
for velvetbean caterpillar or soybean looper, a cost which is competitive
with other controls. Since the production cost estimates included no
allowance for distribution expenses, it was assumed that the producer
receives 62.5 percent of the $16 selling price, with the balance for
the reseller(s).
Research and Development Costs—Table 35 summarizes the R&D
costs that were nominally assumed for the base case. While many of
these expenditures for TH 6040 have already been incurred by the European
developer, estimates of total R&D expenditures were used to illustrate
the costs required to develop a "typical" synthetic organic chemical
pesticide.
Registration expenses are slightly higher than average,
because registrations will be sought for a wide variety of crop-pest
complexes. Since field tests have been completed for mosquitos, forest
insects, and soybeans, the base case assumed that use as a mosquito
control will be registered in 1976 and that registrations for soybeans
and forest insects will be obtained in 1977. The other uses were
assumed to be registered in 1980.
Cash Flow—The base case assumed that (1) TH 6040 will
eventually be registered for soybeans, mosquitos, flies, forest insects,
tobacco, pears, and cotton, (2) the active ingredient will cost $3.60
per pound to produce, (3) the retail price will be $16.00 per pound,
(4) plant capital f'or a one million pound per year plant will be
$2.25 million, and (5) R&D expenditures will total $7.4 million. Based
on these assumptions, the cash flow shown in Table 36 was calculated
for the venture over its life span. Constant costs and prices have
been used over the life of the project, assuming that inflation will
effect both factors about equally.
234
-------
TabTe 35
BASE CASE RESEARCH AND DEVELOPMENT COSTS FOR TH 6040
(Thousands of Dollars)
1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980
Synthesis and screening
875
875
Field tests
100
200
300
400
400
250
250
Toxicology tests
82
170
340
340
340
102
Formulation and chemical
development
82
170
340
340
510
170
49
Registration
_____
80
150
200
120
60
30
30
30
Total
875
875
264
540
1,060
1,230
1,450
642
359
30
30
30
-------
Table 36
BASE CASE PRE-TAX CASH FLOW FOR TH 6040 VENTURE
Net Present Value
R&D Sales at Production and Capital Yearly (1974 dollars)
Year
Cos ts
$16/lb
Distribution Costs
Costs
Profit
At
8 Percent
At 15 Percer
1969
$ -875.0
$ -375.0
$
-1,285.7
$-1, 759.9
1970
-875.0
-875.0
-1,190.4
-1,530.4
1971
-264.0
-264.0
-332.6
-401.5
1972
-540.0
-540.0
-629.9
-714.1
1973
-1,060.0
-1,060.0
-1,144.8
-1,219.0
1974
-1,230.0
$-1,125.0
-2,355.0
-2,355.0
-2,355.0
1975
-1,450.0
-1,125.0
-2,575.0
-2,384.3
-2,239.1
1976
-642.0
$ 58.8
$ -191.1
-31.9
-806.2
-691.2
-609.6
1977
-359.0
1,266.6
-604.2
-68.8
234.6
186.2
154.2
1978
-30.0
3,583.9
-1,396.8
-1,257.1
900.1
661.6
514.6
1979
-30.0
5,925.7
-2,197.7
-1,258.5
2,439.5
1,660.3
1,212.9
1980
-30.0
10,913. 7
-4,074.7
-1,437.8
5,371.2
3,384.8
2,322.1
1981
18,492,6
-6,666.8
-2,682.0
9,143.7
5,335.3
3,437.5
1982
24,917.4
-9,035.2
-1,519.7
14,362.4
7,759.6
4,695.1
1983
30,224.2
-11,021.3
-331.0
18,871.9
9,440.7
5,364.6
1984
35,531.9
-12,836.6
-302.6
22,392.7
10,372.2
5,535.1
1985
38,219.2
-13,755.7
-153.2
24,310.3
10,426.3
5,225.3
1986
38,286.1
-13,778.6
-3.8
24,503.7
9,730. 7
4,579.9
1987
38,324.6
-13,791.8
-2.2
24,530.6
9,019.9
3,986.9
1988
37,686.3
-13,573.5
36.4
24,149.2
8,221.9
3,413.0
1989
36,388.1
-13,129.5
74.0
23,332. 7
7,355.4
2,867.5
1990
35,075.2
-12,680.4
74.8
22,469.6
6,558.7
2,401.2
1991
32,290.9
-11,728.1
158. 7
20,721.5
5,600.4
1,925.6
1992
28,035.0
-10,101.5
2 71.1
18,204.5
4,555.7
1,471.0
1993
23,763.3
-8,640.5
243.5
15,366.3
3,560.6
1,079.7
1994
19,275.6
-7,003.1
272.9
12,745.5
2,734.5
778.8
1995
15,171.6
-5,531.0
245.3
9,886.0
1,963.9
525.2
1996
10,877.7
-4,062.4
244.8
7,060.1
1,298.6
326.0
1997
7,281.2
-2,661.3
233.5
4,853.4
826.6
195.0
1998
4,368.7
-1,665.2
166.0
2,869.6
452.5
100.2
1999
1,456.2
-669.1
277.5
1,064.7
155.5
32.3
Total
$300,033.5
$101,247.8
$41,315.3
236
-------
The table shows that TH 6040 is financially a very attractive
venture. At the eight percent discount rate the venture would return
a pre-tax net present value of $101.2 million over the life of the
project. Increasing the discount rate to 15 percent decreases the pre-
tax net present value, but the resultant $41.3 million value is still
quite attractive.
The $101.2 million net present value at eight percent may
appear high, but these calculations are on a pre-tax basis for a pro-
prietary product that was assumed in the base case scenario to capture
a rather large market. In the next section the implications of a
number of different assumptions on market size and other factors affect-
ing the profitability of the TH 6040 venture are examined.
4. Sensitivity Analysis
Table 37 summarizes some of the more important sensitivity
studies that were performed on the major factors in this analysis. The
sensitivity studies are reported as single variable studies, which can
be compared to determine which individual factor(s) appear to be most
critical.
The table shows that market size is extremely important.
Doubling the nominal market to 7.6 million pounds increases the net
present value at eight percent to $212.4 million; halving it decreases
the net present value to $45.9 million. Although this still represents
an attractive venture, profitability is considerably reduced.
Selling price is also critical. Lowering the selling price
to $13 per pound, all pther factors remaining constant, reduces the net
present value at eight percent to $64.6 million; an additional price
decrease to $10 per pound further lowers the net present value to $37.1
million. Similar effects occur if the production costs are changed.
For example, increasing the production costs from $3.60 to $5.00 per
pound decreases the pre-tax net present value to $75.1 million, all other
factors remaining constant.
Building capital and R&D costs, although significant front-
end expenses, do not as significantly affect the profitability of the
venture. The table shows that the venture still returns a net present
value of nearly $100 million at eight percent if either one of those
variables is at reasonably higher levels. These factors are compara-
tively unimportant because, with the market potential and margin so
large, onetime expenses are more than offset over the life of the
-project. However, failure to obtain a registration will result in a net
present value loss of $7.7 million at eight percent.
237
-------
Table 37
SELECTED SINGLE-VARIABLE SENSITIVITY STUDIES FOR TH 6040
Scenario
Market size
1.9 million lb/yr
3.8 million lb/yr
7.6 million lb/yr
Selling price
$10/lb
$13/lb
$16/lb
Production cost
$3.00/lb
$3.60/lb
$5.00/lb
Registration
Obtained
Not obtained
R&D
$5.0 million
$7.4 million
$10.0 million
Building capital
$1.5 million
$2.25 million
$5.0 million
Net Present Value
at 8 Percent
(millions of dollars)
$45.9
101.2 (base case)
212.4
37.1
64.6
101.2 (base case)
112.2
101.2 (base case)
75.1
101.2 (base case)
-7.7
103.8
101.2 (base case)
98.4
103.4
101.2 (base case)
99.1
5. Probabilistic Analysis
The preceding analysis of TH 6040 was based on nonproprietary
data and in many cases there was significant uncertainty about the values
of specific critical variables. This section discusses the effects of
these uncertainties on the overall profitability of the venture.
Figure 29 summarizes the major variables considered in this
probabilistic analysis.
Figures 30 and 31 summarize SRI's current uncertainty abcut
these variables. The probability assignments were subjectively determined
238
-------
Daemon to
Fund R&D
R&D Costs
Building
Capital
Production
Costs
Successful
Registration
Decision to
Market
Selling
Price
Fund
Do Not Fund
Market
&
Medium
Do Not Market
FIGURE 29 SCHEMATIC OF TH 6040 COMMERCIALIZATION DECISION
-------
R&D Com
Building Capital
Production Costs
0.20
$3.60 per lb
0.40
p - 0.40
0.15
$2.25 Million
0.50
p - 0.35
0.t5
$7.4 Million
p - 0.60
p - 0.25
FIGURE 30 PROBABILITY ASSIGNMENTS ON RESEARCH AND DEVELOPMENT COSTS.
BUILDING CAPITAL, AND PRODUCTION COSTS FOR TH 6040
240
-------
Successful
Registration
Selling Price
Market Size
PROBABILITY ASSIGNMENTS ON REGISTRATION, SELLING PRICE, AND
MARKET SIZE FOR TH 6040
FIGURE 31
241
-------
from relevant information at the Institute's disposal. Individuals with
access to proprietary information on TH 6040 might assign different
probabilities, but the assignments used in this study were judged to be
the best possible given the available level of information.
As with the preceding analyses for B.t., Heliothis spp. NPV,
and gossyplure, each variable in Figures 30 and 31 was assigned a limited
set of possible values for simplicity and to control the derivative
calculations. Each probability assignment was assumed to be independent
of the values of the other factors in the analysis.
The variables in Figures 30 and 31 combine to produce 690
scenarios of the final outcome of the venture. Each of these scenarios
was evaluated for its net present value, and related scenario probabil-
ities were determined by multiplying the probability assignments on
their respective input factors.
Figure 32 summarizes the overall pre-tax profitability of the
venture at eight percent in a probability distribution. To facilitate
the presentation, the values have been rounded to the nearest $5 million.
Based on present information, the venture appears quite attractive.
There is a seventy-nine percent chance that the company would at least
break even at an eight percent and a fifty percent chance that it would
return at least $45 million. The venture could lose up to $10 million
or return up to $240 million. However, the chances of this loss are
only 5.0 percent, and the chances of a return in excess of $200 million
are 1.5 percent. On the average, the venture would be expected to make
a pre-tax net present value of $54.5 million at eight percent.
Although the venture appears quite attractive on an expected
value basis, there is a 21 percent chance that the company would lose
some money, with the largest loss being $10 million. Therefore, the
venture's risk for companies with different assets was investigated.
As discussed in IV-B-4, "Probabilistic Analysis", degrees of risk can
be quantified, but these measures depend on both the profit lottery and
a company's attitude toward taking risk. This risk attitude generally
depends on the company's size as well as its business philosophy.
The exponential weighting factor described in IV-B-4 was
applied to the outcome measures, and the certain equivalent of the
venture was calculated.
Table 38 gives the certain equivalent of the TH 6040 venture
as a function of the exponential weighting parameter, 0. The table
shows that if O = 0, then the company is risk indifferent and the
certain equivalent is equal to the expected value of the venture, or
$54.5 million.
The table also shows that if a = 0.00000020080, the certain
equivalent of the venture becomes almost zero. Thus, if the company's
tolerance for risk was approximately reflected by the parameter
242
-------
ro
•F>
U)
>
(-
0.15
0.14
0.13
0.12
0.11
0.10
0.09
0.08
m
<
go
O 0.07
X
EL
0.06
0.05
0.04
0.03
0.02
0.01
I
J~L
m
-20 -10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180
NET PRESENT VALUE — millions of dollars
FIGURE 32 PROBABILITY DISTRIBUTION OF NET PRESENT VALUE, AT EIGHT PERCENT, OF TH 6040 VENTURE
-------
Table 38
CERTAIN EQUIVALENT OF TH 6040 VENTURE
AS A FUNCTION OF RISK COEFFICIENT, a
Certain Equivalent
a (millions of dollars)
0.0
$54.5
0.0000000010
53.1
0.00000020080
0.1
0.00000030070
-2.5
0.000001
-6.9
a = 0.00000020080, it would almost be indifferent to commercially devel-
oping TH 6040 (it would perceive the risk of losing money in the venture
as almost counteracting the potential gains). The literature on risk
preference (See IV-E, "Selected References") has shown that if an indi-
vidual is indifferent to a 50 percent chance of winning x and a 50 per-
cent chance of losing x/2, then his risk parameter, 0, is equal to 1/x.
Since 1/0.00000020080 is approximately equal to $5 million, a company
would find the TH 6040 venture too risky if it were unwilling to take a
50 percent chance at winning $5 million versus a 50 percent chance at
losing $2.5 million.
Viewed in this manner, the TH 6040 venture appears quite
different. Although the venture would return $54.5 million on the
average, there is a 21 percent chance of losing money with possible
losses as high as $10 million. The preceding calculations show that
if a company were unwilling to take a 50 percent chance at winning
$5 million and a 50 percent chance of losing $2.5 million, it would
find the TH 6040 venture unattractive. Most small and some modestly
sized companies probably could not sustain such losses. Therefore,
although TH 6040 appears to be an attractive venture on an expected
value basis, it would probably only be attractive for relatively large
companies.
244
-------
D. Selected References
The following are a few of the references screened that relate to
the Information in this chapter.
Corbett, J. R., The Biochemical Mode of Action of Pesticides, Academic
Press (1974).
Dewey, J. E. , "Pesticides for Minor Crops: A Major Dilemma," Weeds
Today, pp. 10-12 (Fall 1974).
Johnson, J. E., and E. H. Blair, "Cost, Time and Pesticide Safety,"
Chemical Technology, pp. 666-670 (1972).
USDA, Losses in Agriculture, Agricultural Handbook 291, Agricultural
Research Service (1965).
Williams, C. M., "Third-Generation Pesticides," Scientific American,
Vol. 217, p. 13 (1967).
245
-------
X OTHER AGENTS
A. Overview
There are a host of other agents and techniques that can be used
to control pests. These include pathogens, natural parasites and preda-
tors, sterilization techniques, sound and other energy waves, pest-
resistant crop strains, and so on. Many of these techniques were re-
viewed during the initial stages of this contract; for example, it was
determined that the use of sound and energy waves for weed control (by
broadcasting through soils on a pre-emergence basis to destroy weed
seeds) appears to be impractical because of high energy costs and the
variability of the soil as a media under actual field conditions.
In general, SRI concluded that the application of selected tech-
niques would continue on an evolutionary basis in agricultural, nonagri-
cultural, and public health programs. In many cases, the range of
activities is almost totally controlled by public agencies, and a more
detailed assessment of the critical factors associated with the con-
tinued development of such procedures could not be effectively evaluated
within the private industry commercial feasibility parameters used in
this study (although many of the most significant economic factors cited
are still likely to apply, particularly for agents or materials requiring
registration).
For background and reference, however, one such product was
briefly reviewed in the early stages of this contract, and the results
of this investigation follow. No decision analysis modeling was con-
ducted for this product.
B. Colletotrichum Gloeosporioides, A Fungal Pathogen of Northern
Jointvetch
1. Introduction
The leguminous weed, northern jointvetch (Aeschynomene
virginica [L.] B.S.P.) is a troublesome pest in rice fields in
Arkansas, Mississippi, and Louisiana, causing losses in rice yield and
quality estimated at up to $2.7 million annually. Several chemical
herbicides, but chi*efly the phenoxy herbicide 2,4,5-T, are currently
used for its control.
For the past seven years research has been conducted at the
University of Arkansas (Fayetteville) and the University's Rice Branch
Experiment Station (Stuttgart) to evaluate the weed control potential
247
-------
of the endemic fungal parasite, Colletotrichum gloeosporioides, the
causal agent of an anthracnoce disease on northern jointvetch. Results
of this study have demonstrated that a high degree of control can be
obtained by inundative field inoculations with spores of the pathogen
produced via fermentation processes.
In April, 1974, the University of Arkansas applied to the EPA
for an experimental use permit with waiver of fee, and requested an
exemption from tolerance in order to initiate large-scale field testing.
This request was granted. Additional research in mammalian toxicity,
biomagnification potential, spore stability, pathogen susceptibility to
fungicides, and quality control of the product (spores) was also planned.
2. Summary
Control of northern jointvetch by the mycoherbicide C.
gloeosporioides in the three-state area would require an estimated
3,125kg of inoculum annually in the form of wet spore paste and could
displace the annual use of approximately 50,000 to 60,000 pounds of
2,4,5-T. Gross annual sales4of the mycoherbicide in this area could
amount to $500,000. Displacement of 2,4,5-T to the extent estimated
could be offset by an increase in the use of other chemicals such as
2,4-D and silvex for necessary broadleaf weed control, however. It is
expected that the practice of "spot spraying" would increase at the ex-
pense of aerial application of chemical herbicides.
Commerciability of the mycoherbicide is favored by its antici-
pated cost'competitiveness, the high level of northern jointvetch control
achievable, host specificity, convenience of application, absence of
timing requirements, absence of drift problems, absence of phytotoxic
and mammalian toxicity effects, safety to the environment, and favor-
able grower attitudes. Commercialization by private industry is inhib-
ited, however, by potential problems with product (spore) shelf life,
by an apparent inability to obtain product or process patent protection,
and, particularly, by the small apparent market size.
As of this writing, the mycoherbicide was still in the devel-
opmental stage. All developmental research has been conducted by
research personnel connected with the University of Arkansas. The
university had received an experimental use permit and exemption from
tolerance for large-scale field testing from the EPA. A large commer-
cial producer of pharmaceuticals was cooperating with the university and
had produced one 26-kg batch of wet spore paste for experimental use by
university researchers. This firm was not participating in field test-
ing and had no agreement with the university in regard to possible com-
mercial production or marketing. The mycoherbicide was currently being
produced in experimental quantities only.
"k
Directly prior to the publication of this report, it was announced that
possibilities for the commercial production of this mycoherbicide were
being explored by the Upjohn Company.
248
-------
It is unlikely that a meaningful patent opportunity exists for
the mycoherbicide, since the fungus is an unpatentable natural entity,
and because the method of production via fermentation is commonplace.
Any fermentation process patent that might be obtained could probably be
easily broken.
3. Description and Properties .
Name
Colletotrichum gloeosporioldes (Penz.) Sacc. forma species
aeschynomene. A fungus.
History and Distribution
The fungus was first found in 1969 on northern jointvetch in
experimental rice plots at Stuttgart, Arkansas, during the course of
studies on competition between weeds and rice. It was subsequently
shown to be endemic on this host in thirty-one rice producing counties
in Arkansas, and is probably endemic throughout all rice producing
counties in the Mississippi River Delta. Neither the fungus nor its
host is present in California rice fields. Although the fungus may
be frequently observed in nature, the disease exerts little, if any,
natural control pressure on the host weed because of the low volume of
viable spores produced in nature. However, when inundative inoculum
levels are artifically provided by the spray application of large
quantities of laboratory-produced spores, satisfactory levels northern
jointvetch control have been achieved.
Mycological Description
The fungus was initially identified as a new species of
Gloeosporium. Since this genus is no longer considered as a viable
taxon, the fungus was placed in the valid and closely-related genus,
Colletotrichum, and was established as the species gloeosporioldes,
form specialis aeschynomene. This was based upon evidence that the
fungus is infective only on hosts within the genus Aeschynomene. The
genus Colletotrichum is within the form family Melanconiaceae, form
order Melanconiales, and form class Deuteromycetes (Fungi Imperfectly),
whose members conventionally reproduce only by means of asexual spores
(conidia). Although little is currently known about the life cycle of
C. gloeospot'ioides, it has been presumed to represent an asexual or
imperfect stage of the genus Glomerella. The sexual stage of this
species, however, has not been observed.
Toxicology
Spore suspensions used as inoculum are not phytotoxic to
plants that are immune to infection.
Preliminary mammalian toxicological studies have been conducted
at the Department of Animal Science, University of Arkansas, where
249
-------
weanling rats and rabbits were subjected to various treatments with
spore suspensions or spore pastes. The following tests were performed.
• Acute Oral. No toxic effects were observed when the spores
of the pathogen were administered orally to weanling rats.
A dose equivalent to 1500 ml per 70 kg man (1500 ml will
be applied to a 100-acre field) was administered by stomach
tube to each of 10 weanling rats and observations were made
for 10 days. The test animals exhibited no symptoms of
discomfort, and rate of weight gain was not affected. His-
tologic examination of the alimentary track, heart, kidneys,
lungs, spleen, pancreas, and liver revealed no abnormalities.
• Eye Irritation. Eye irritation was not noted in four rabbits
following a 3-drop dose of an aqueous suspension of 2 million
spores per ml. Observations were made daily for one week.
• Dermal Sensitivity. Spore paste was applied to the shaved
skin of four rabbits and observations were made daily for
one week. Slight reddening and thickening of the skin
occurred in three of the four rabbits. This reaction was
apparent at 24 hours and slowly subsided in three to four
days. Histologic examination revealed congestion and a
mild lymphocytic infiltration.
• Inhalation. Four weanling rats were placed in a closed
four-liter container and 5 ml of an aqueous suspension con-
taining 8 million spores per ml was atomized into the con-
tainer. Animals were weighed weekly and observed for 14
days. Post mortem revealed no abnormalities of lung or
other tissues. Isolations were made of the macerated lung
in an effort to recover the fungus. No Colletotrichurn were
recovered from the cultures.
• Intraperitoneal and Subcutaneous Inoculation. Five weanling
rats were injected subcutaneously with 1.5 million spores
in an aqueous suspension. Five other animals were given a
similar dose intraperitoneally. Experimental animals and
controls were observed daily and weighed each week.
Necropsies were performed at four weeks postinoculation.
No symptoms or lesions were observed. Histologic studies
are incomplete.
• Mycotoxin Production. The fungus was cultured on double-
autoclaved brown rice in 500 ml Erlenmeyer flasks for 3
weeks at room temperature. The molded rice was then dried,
ground, and mixed with a standard rat ration to make a final
ration concentration of 25 percent molded rice. This ration
was fed ad libitum to 10 weanling tats for 10 consecutive
days. Weights of control and test animals were similar and
no signs of distress or disease were observed. Necropsies
were performed and no gross ^esions were apparent. Histologic
studies are incomplete.
250
-------
Further toxicological tests are planned.
Host Range
Pathogenicity tests in the field and the greenhouse have
indicated that the pathogen is only virulent to northern jointvetch.
A low order of infection has been achieved on Indian jointvetch
(Aeschynomene indica L.) samples collected in Arkansas, India, Rhodesia,
and Australia, but A. faliata from Brazil was immune to infection. All
other species tested were immune to infection. Table 39 lists those
species immune to infection under field conditions; infection of the
test plants was not observed, though northern jointvetch grown as a
control in every tenth row of the test plots was killed by the treatment
(2 million spores per ml at 40 gallons per acre). Yields from 10
varieties of soybeans in field plots sprayed at the rate of 10 gallons
per acre with a spore suspension containing 3 million spores per ml
were not significantly different than yields from nonsprayed plots.
Three spray treatments applied—at pre-bloom, bloom, and post-bloom
stages of plant maturity.
It may be concluded from current information that virulence
of Colletotrichum gloeosporioides is sufficiently specific to northern
jointvetch to preclude unacceptable risk to adjacent, desirable plant
species.
Symptomology and Etiology
Single-celled, orange-colored, ovoid or oblong condidia,
ranging in size from 14 to 17 by 4 to 5.7 y are produced on a disc-
shaped or cushion-like subepidermal acervulus from simple conidiophores.
The spores accumulate in a slimy orange mass on the acervular surface
of the host epidermis, and may be spread by splashing (rain or irriga-
tion water) or by insects. Typical anthracnose lesions are produced on
the northern jointvetch host, largely on stems and frequently in the
leaf axil region following inundative inoculations. Lesions resulting
from natural infection are found largely along the water line on stems.
Individual lesions are typically purple in color, within which the
orange spore masses may easily be seen by the naked eye. Lesions are
from 1.0 to 3.5 cm in length, and frequently coalesce to girdle the
entire stem. Under natural conditions the rate of disease spread is
low, never appearing in appreciable intensity until late in the growing
season. Following inoculation, a period of 4 to 7 days is required for
the appearance of disease symptoms, the first of which appears as a
small, water-soaked spot. With normal disease development, death of the
infected plant ususally occurs within 2 to 5 weeks. Infected plants
which are not killed are severely stunted, produce few seeds, and do
not actively compete with rice. The pathogen is favored by warm tempera-
tures (28° to 30°C) and humid conditions. Overwintering of the fungus
occurs in infested plant debris in or on the soil.
251
-------
Table 39
CROP, ORNAMENTAL, AND,WEED PLANTS THAT HAVE BEEN OBSERVED
TO BE IMMUNE TO INFECTION WITH COLLETOTRICHUM
GLOEOSPORIOIDES UNDER FIELD CONDITIONS
Common Name
Latin Name
Commercial Variety
CROPS
Alfalfa
Bean (green)
Bean (lima)
Beet
Cabbage
Cantaloupe
Carrot
Clover (red)
Collard
Medicago sativa L.
Phaseolus vulgaris L.
Phaseolus limensis
Macf.
Beta vulgaris L.
Brassica oleracea
L., var. capitata
Cucumis melo L.,
var. reticulatus
Naud.
Daucus carota
L., var. sativa D.C.
Trifollum repens L.
Brassica oleracea L.,
var. acephala
Buffalo
Cody
Delta
Keiser Water
Tolerant
Lahontan
Ranger
Vernal
Victoria
Burpee Stringless
Green Pod
Golden Wax
Kentucky Wonder Pole
Romano Pole
White Hale Runner
Henderson's Bush
King of the Pole
Ruby Queen
Stonehead
Round Delecti
Honeydew
Crensaw
Casaba Golden
Danvers
Yates
252
-------
Common Name
Table 39 (continued)
Latin Name
CROPS (continued)
Commercial Variety
Corn (field)
Corn (sweet)
Cotton
Cucumber
Eggplant
Gourd
Lespedeza
Lupine
Mangel
Mustard
Oats
Okra
Onion
Zea mays L.
Zea saccharata Sturtev
Gossypium hirsutum L.
Cucumis sativus L.
Solanum melogena L.
Cucurbita sp.
Lespedeza stipulacea
Maxim.
Lupinus albus L.
Beta vulgaris L.,
var. macrorhiza
Brassica sp.
Avena sativa L.
Hibiscus esculentus L.
Allium cepa L.
McNair 6904
Stull 911 Sx
Tenn. 501 R
Texas 30A
Pennington (silage)
F-M Cross
Brycot 4
Coker 310
Delcot 277
Lockett 4789
Pennington HyBee
Quapaw (Ark 61-28)
Stoneville 213
Stripper 31-D-21-2
Tamcot Sp 37
West India Gherkin
White Wonder
Model
Black Beauty
Ball Mixture
326-10
Summit
Kobe
Mammoth Long Red
Southern Giant
Curly
Compact
Arkwin
Clemson Spineless
Dwarf Long Pod
Yellow Globe
253
-------
Common Name
Table 39 (continued)
Latin Name
CROPS (continued)
Commercial Variety
Pea
Pepper
Pumpkin
Radish
Rice
Sorghum
Soybean
Pisum sativum L.
Capsicum frutescens L.
Cucurbita peop L.
Raphanus sativa L.
Oryza sativa L.
Sorghum bicolor L.
Moench.
Glycine max L.
Little Marvel
Alderman
Lady Cowpea
California Wonder
Cinderella
Icicle
9654
Dawn
Nato
Nortai
Nova 66
Saturn
Star Bonnet
Acco R-1093
AKS 614
AKS 653
Funks G-393
Funks G-522
Leafmaster 43
(silage)
McNair 546
Arksoy
Bragg
D62-7812
Dare
Davis
Hale-3
Hill
Hood
Lee
Lee 68
Mack
McNair 600
McNair 800
Ogden
PI-82 264
PI-230 979
Pickett-71
Ransom
Semmes
York
254
-------
Common Name
Table 39 (Continued)
Latin Name
CROPS (continued)
Commercial Variety
Spinach
Squash
Sugarbeet
Swiss Chard
Tomato
Turnip
Watermelon
Wheat
Spinacia oleracea L,
Cucurblta sp,
Beta vulgaris L.
Beta vulgaris L.,
var. cicla Moq.
Lycopersicon
esculentum Mill.
Brassica rapa L.
Citrulus vulgaris
Schrad.
Triticum aestivum L,
Asgrow 424
Dixie Market
King of Denmark
Northland
Virginia Blight
Resistant
Buttercup
Blue Hubbard
Improved Water
Hubbard
Sweet Meat
N.K. U.S. strain
Fordhook Giant
Traveler 193
Marion
Crawford
Purple Top
White Globe
Shogoin
Charleston Gray
Summerfield
Monor
Omar
Purple Straw
ORNAMENTALS
Marigold
Sweet Alyssum
Zinnia
Tagetes sp.
Lobularia maritima
Zinnia elegans Jacq,
Burpee's Best Double
Dwarf
Giant Dahlia Mixed
255
-------
Common Name
Table 39 (Concluded)
Latin Name
WEEDS
Arrowhead
Barnyardgrass
Blunt Splkerush
Broadleaf Slgnalgrass
Duck Salad
Eclipta
False Nutsedge
Gooseweed
Hemp Sesbanla
Purple Ammannia
Redstem
Rice Flatsedge
Sprangletop
Waterhyssop
Sagittaria sp.
Eclinochloa crus-galli L.
Beauv.
Eleocharis obtusa (Willd.)
Schutes
Brachiaria platyphylla (Griseb.)
Nash
Heteranthura limosa (Sw.) Willd.
Eclipta alba L. Hassk.
Cyperus strigosus L.
Sphenoclea zeylanica Gaertin.
Sesbanla exaltata (Ruf.) Cory
Ammannia cocclnea Rottb.
Ammannia auriculata Willd.
Cyperus iria L.
Leptochloa panicoides (Presl.)
Hitchc.
Bacopa rotuudifolia (Michx.)
Wettst.
Winged Waterprimrose
Jussiaea decurrens (Walt.) DC
256
-------
Mechanism of Action
The mechanism through which control is effected is biological
infection, resulting from spores germinating on the host surfaces and
followed by parasitism of the infected host tissues by the established
pathogen. Little is known about the actual biochemistry of parasitism
by this particular fungal parasite, except that it leads to the deterio-
ration, collapse, and death of the infected host tissues, eventually
resulting in the death of the entire plant. Observations made during
early disease stages suggest that a toxin may play a role in pathoge-
nicity. Control of the susceptible host is achieved through concentra-
tions much in excess of those present under natural conditions.
Efficacy Data
Controlled inoculation field experiments have been conducted
over the past seven years at Stuttgart. In 1970, small field plots
containing northern jointvetch were inoculated with spore suspensions
containing 2, 3, and 6 million conidia/ml. Inoculations at these three
levels were conducted at three different plant heights—10, 25, and 89
cm. Test results are given in Table 40.
Table 40
CONTROL OF NORTHERN JOINTVETCH WITH COLLETOTRICHUM GL0E0SP0RI0IDES
IN SMALL FIELD PLOTS, STUTTGART, ARKANSAS, 1970
Percent Control at;
Plant Height
(cm)
2 million
spores/ml
3 million
spores/ml
6 million
spores/ml
Average
10
100%
86%
86%
91%
25
86
86
72
81
58
86
72
57
72
Average
91
81
72
In 1971, the effect of infection on stand and weight of
northern jointvetch was studied. Inoculum was applied to plants 36 and
66 cm in height at the rate of about 33 gal/acre (2.6 million spores/
ml). Reductions in stand and dry weight of sprayed plots were compared
with nonsprayed controls; test resutls are shown in Table 41. Stand
data were determined 40 days after treatment for plants that averaged
36 cm in height at the time of inoculation, and 27 days after inocula-
tion of plants averaging 66 cm in height. Dry weight' data were deter-
mined for weeds harvested 53 days after the 36-cm height treatment, and
40 days following the 66-cm height treatment.
257
-------
Table 41
EFFECT OF INOCULATION OF NORTHERN JOINTVETCH IN SMALL FIELD PLOTS WITH
SPORES OF COLLETOTRICHUM GLOEOSPORIOIDES, STUTTGART, ARKANSAS, 1971
Weed Stand
Average Weed Height at
Time of Treatment (cm)
36
66
Check
Percent
Plants/m^ Reduction
0.03
0.01
5.08
99%
98
Weed Dry Weight
Percent
g/m^ Reduction
1.2
1.0
160.0
99%
99
During 1972, large field plots were treated with aerially ap-
plied inoculations at the rate of 10 gal/acre with a spore suspension
containing 2 million spores per ml. The weeds in seed rice fields were
sprayed when they were 30 to 45 cm tall. Control was 47% complete by
the 16th day following treatment, 84% complete by the 33rd day, and 99%
complete 46 days after inoculation. Plants not killed by the treatment
did not grow appreciably after inoculation. In 1973, four large fields
between 10 and 40 acres in size were inoculated by ae.rial applications
of spore suspensions (10 gal/acre, 2 million spores/ml); test results
are shown in Table 42. Data from the 1974 tests were not available at
the time this report was prepared.
Table 42
AERIAL APPLICATION OF COLLETOTRICHUM GLOEOSPORIOIDES SPORES
IN LARGE FIELDS, STUTTGART AND CARLISLE AREAS, 1971
Field
No.
Acres
Treated
Stage and Height of Plant
Field
Condition
Percent
Control
1
10
Seedling, 10-15 cm
Flooded
95%
2
10
Seedling, 10-15 cm
Dry
95
3
20
Juvenile, 40-50 cm
Flooded
100
4
40
Flowering, 110-160 cm
Flooded
99
From the results of these tests, it appears that a satisfactory
level of northern jointvetch control is possible in the field using
inundative inoculations with spores of _C. gloeosporioides. Moreover,
the natural spread of the fungus from field inoculations was claimed by
University of Arkansas personnel to be sufficient to control weeds
emerging after treatment.
258
-------
4. Production
Isolation
The pathogen is easily isolated from diseased tissue, and it
sporulates profusely in a pure culture on potato dextrose agar, lima
bean agar, or liquid nutrient media. Optimum growth occurs at 28 to
30°C; no measurable growth occurs at either 10 or 40°C. Stock cultures
of the fungus can be maintained for an indefinite period in sterile
soil cultures.
Mass culture
Mass production of spores for large-scale field use is amenable
to large batch fermentation methods, using aerated liquid media. High
spore yields have been obtained on a modified Richard's solution con-
taining 50g sucrose, lOg KNO3, 5g KH2PO4, 2.5g MgS04*7H20, 0.02g FeCl3,
150 ml "V-8" juice, and distilled water to make 1 liter. The pH of
the medium is adjusted to 6.0 with 50 NaOH. Brewer's yeast or other
supplements may be substituted for the "V-8" juice component. Following
fermentation, spores are separated from the growth medium by filtration
and centrifuging, yielding a spore paste which may be resuspended in
water for field application. Alternative methods for spore harvesting
include freeze drying or spray drying, the latter being faster and less
expensive. Mass production of spores had not yet been carried beyond
the experimental stage; the largest batch produced as of this writing
was 26kg of spore paste from about 400 liters of "beer" fermented for
48 hours. However, because of the straightforward nature of the process,
no insurmountable obstacles to large-scale production were anticipated.
Spore stability
Questions remain about the stability of spores, and to suitable
methods of spore preservation following their harvest from the culture
medium. Wet spore paste can be stored for only 2 to 4 days at 40°F
before contamination by airborne bacteria begins to sharply reduce spore
viability. Freeze-dried spores can be stored indefinitely without ap-
preciable loss of viability. No information was available on the lon-
gevity of spray-dried spores. Considerable research is definitely needed
in this critical area.
Formulation and application
Formulations of spore material used in experimental field
inoculations consisted of simple water suspensions of condidia. Spore
concentrations in standard stock suspensions are determined via spore
counts using a haemocytometer. Stock suspensions are added to water in
the spray tank to deliver the desired final spore concentration. Concen-
trations of 1 to 2 million spores per ml applied at the rate of 10 gal/
acre have provided satisfactory control levels. The two-million spores/
ml level represents about 25 grams of wet spore paste per acre. Field
259
-------
applications may be made with conventional ground or aerial spraying
equipment, and should be done at dusk to avoid excessive drying of the
inoculum.
Costs of production
No cost information had yet been developed for spore production.
In view of the straightforward fermentation techniques expected to be
involved, it is doubtful that production costs should inhibit commer-
ciability. Costs for nutrient medium ingredients alone would probably
not exceed 6c per liter. Assuming that 20 litres of "beer" are required
to produce 1 kg of wet spore paste, nutrient medium costs would be ap-
proximately $1.20 per kg wet spore paste produced, or 3c per acre treated
(25g of wet spore paste). Cost of culture medium ingredients, therefore,
do not appear prohibitive. Additional costs would be incurred for labor,
equipment, power, spore harvesting, packaging, storage, and distribution.
At this time, however, it would appear unlikely that total production
costs would exceed $1.00 per acre treated. Allowing for a suitable
margin of profit for this relatively minor market, the cost of the
mycoherbicide to the grower should be competitive with that for current
herbicide controls. Grower costs for 2,4,5-T, for instance, have been
about $4 to $5 per treated acre ($2.85 to $4.25 per lb). However,
definitive cost estimates for the Colletotrichum mycoherbicide must await
the results of further research.*
5. Market and Use Parameters
Northern Jointvetch
Northern jointvetch is an annual legume that reproduces by
seeds. It is a significant pest in cultivated rice. The common name
derives from the fact that the mature seed pod easily segments into 4
to 10 nearly square "joints". The weed is also called "sensitive joint-
vetch" and "bashfulweed" because of a shrinking response of leaflets to
touch (thus the generic name, Aeschynomene, Greek for "ashamed"). Other
common names are "curly indigo" and "coffeeweed".
The species is native to the eastern seaboard south from New
Jersey to the southeastern states, the rice-growing states of Arkansas,
Mississippi, Louisiana, and Texas, and to Mexico and the Caribbean.
It is not present in the California rice-growing region. It is favored
by habitats subject to periodic inundation, and therefore appears in
*Directly prior to the publication of this report, it was announced that
a method of mass production for this mycoherbicide had been developed by
USDA-ARS weed scientists in conjunction with plant pathologists at the
University of Arkansas, and that possible commercialization was being
explored with the Upjohn Company.
260
-------
rice fields and along irrigation canals, levees, and ditch banks. This
weed may also be found in soybean fields planted in rotation with rice,
although it is not a major pest in soybeans.
The plant is erect with a branching stem, and attains a height
of 1.5 m. The compound leaves bear 25 to 50 linear leaflets that fold
at night or when touched. The infloresence bears 1 to 6 yellow flowers.
The jointed seed pods are 2.5 to 7.6 cm long at maturity, and the kidney-
shaped seeds may remain viable for several years in the soil.
Northern jointvetch is but one of several weeds troublesome to
the southern rice grower. Hurst ot aL. list 32 genera of weeds that
are considered significant pests in Arkansas rice fields. Northern
jointvetch is perhaps the most troublesome species of the group, however.
Effect of the Weed on Rice
In the three-state rice-growing area of Arkansas, Mississippi,
and Louisiana, approximately 125,000 acres of the 800,000 acres in rice
production have been infested with northern jointvetch. Presence of the
weed reduces both rice yield and rice quality. Populations as low as
2.7 plants per square meter cause yield reductions if allowed to persist
through the growing season. Yield losses are largely related to shading
of the young rice plants by the taller weeds, which is most evident 8 to
.10 weeks after seedlings emerge. Seeds of northern jointvetch are dif-
ficult to remove from rice during milling. Since the tolerance for
these seeds in packaged rice is zero, rice grade is reduced if these
seeds are present. Losses to Arkansas rough rice production due to
northern jointvetch in 1973 were estimated to be $2.7 million, which
includes a 2% yield reduction and a 4% reduction in rice quality (grade)
value.
Current Controls: Problems and Costs
Chemical herbicides are currently used to control northern
jointvetch. Phenoxy herbicides such as 2,4,5-T, 2,4-D and silvex are
used against broadleaved species, including northern jointvetch.
Propanil is used against grassy weeds such as barnyardgrass and is often
applied early in the season prior to the period of rice development that
is safe for phenoxy herbicide application. The most effective of these
herbicides against jointvetch is 2,4,5-T, since it kills the weeds
within two weeks following application at the rate of 1.0 to 1.5 lbs
per acre.
The use of chemical herbicides rarely results in complete
control of seasonal stands of northern jointvetch. This is largely due
to the lack of residual action (weeds emerging after treatment are not
affected). The average level of control has been about 75%.
If not applied at the proper time, phenoxy herbicides can
seriously injure developing rice plants. These herbicides can be applied
261
-------
only during a 7-to-10-day period coincident with internode elongation;
applications should be terminated when the first internode is approxi-
mately 1.3 cm in length. If the herbicide is applied before this
period the rice panicle can be prevented from emerging because of
abnormal development of the surrounding boot leaves. If applied after
panicle emergence the panical may not fill properly due to injury by the
herbicide, resulting in yield losses up to 30%. The timing of this
"safe period" varies between varieties. Unfavorable weather conditions
may prevent the grower from spraying during the safe period, in which
case the field remains untreated.
Improper timing of phenoxy herbicide applications may also
increase the susceptibility of rice to a brown leaf spot fungus disease
caused by Helminthosporium oryzae.
Crops susceptible to phenoxy herbicides (particularly cotton
and soybeans) that grow adjacent to treated rice fields may often be
damaged by these herbicide sprays drifting from the target area. The
phenoxy herbicides typically injure such broadleaved plants, and propanil
has also been shown to injure soybeans. Because of this risk, rice
farmers often either refrain from spraying and suffer the consequent
losses from weed infestations, or spray only the center portion of rice
fields, in order to leave an unsprayed border area sufficiently wide to
capture drift. Because of the drift problem, which is intensified by
the relatively small size of the average rice field (40 to 60 acres), it
has been estimated that only about 75% of the rice acreage infested with
northern jointvetch was actually treated with 2,4,5-T in the three-state
area in 1974.
Northern jointvetch is traditionally controlled with phenoxy
herbicides, principally 2,4,5-T, applied once in the growing season at
rates of 0.5 to 1.5 lbs per acre. In 1974, an estimated 90,000 acres
were treated in the three-state area, requiring an estimated 90,000 lbs
of 2,4,5-T. Custom rates for aerial application of herbicides averaged
about $7.50 per acre, varying with the amount applied. Of this cost
about $4.00 was the average cost for the herbicide; the remainder was
for operating expenses, liability, and the custom applicator's profit.
Therefore, approximately $675,000 was spent annually for 2,4,5-T applica-
tion in rice, of which about $360,000 was the material cost.
Field Application and Relative Utility of the Mycoherbicide
The recommended rate of application of the mycoherbicide' is
10 gallons per acre of spore suspension (in water) containing a concen-
tration of 1 to 2 million spores per ml. Twenty-five grams of wet spore
paste would be sufficient to provide the inoculum for an acre at a con-
centration of approximately 2 million spores per ml. This quantity of
spores represents about 0.008 lbs of dry matter. Application is most
effective when made near dusk.
262
-------
The timing of the application is not as critical as it is with
phenoxy herbicides, as long as treatments are applied early enough in the
season to prevent retardation of .young rice plants by shading. Weed
control is most effective when the weeds are relatively young*; however,
99% control has been obtained when the plants were 66 cm tall. Field
observations indicate that treatments at very early stages of weed de-
velopment are less effective because insufficient leaf area is exposed
to receive the inoculum, or because the weed plants are protected by the
rice canopy.
It is apparent from available research and observations that
the use of C^. gloeosporioides as a mycoherbicide against northern joint-
vetch would have several advantages over the use of 2,4,5-T:
• The mycoherbicide provides a 98 to 100% level of control,
due in part to the residual effectiveness afforded by the
inundative levels of the fungus inoculum applied.
• It produces no phytotoxic effects on rice.
• It does not increase the susceptibility of rice to other
pathogens.
• Drift is not a problem, since the fungus Is avirulent on
crop plants or ornamentals that might be within the drift
pattern. Therefore, a greater number of acres infested
with the target weed could be treated.
• There is little restriction on the timing of applications
relative to a specific growth stage of the rice. Applica-
tions could therefore be made at greater convenience to
growers.
• The pathogen is apparently environmentally safe and leaves
no residues harmful to wildlife or soil flora.
• It is apparently nontoxic to mammals, including man.
• The quantity of material applied amounts to only 0.008 lbs
per acre (ca. 3.6 grams), which is about 1/125 the amount
of active ingredient applied in 2,4,5-T treatments.
• Grower attitudes appear to be favorable toward the use of
this control. Since the target weed is a serious rice
pest and conventional controls involve a level of risk
which is often unacceptable, growers involved in the
experimental field applications of the mycoherbicide have
been favorably Impressed and are anxious to have the pathogen
regularly available for use.
Some of these advantages are offset by disadvantages, however.
One prime functional disadvantage is that the mycoherbicide is only ef-
fective against northern jointvetch, while 2,4,5-T controls several
weed species with a single application. Therefore, in fields where a
spectra of weeds are economic problems, application of the mycoherbicide—
despite Its effectiveness on northern jointvetch—could represent an
263
-------
added cost to the grower. (No information had been developed on the
compatibility of the mycoherbicide with chemical herbicides.)
Three other factors also appear to detract from the viability
of the mycoherbicide as a commercial product: the problem of spore
stability and shelf life, the lack of patentability, and the small
market size. Considerable research is needed to determine suitable meth-
¦p
ods for packaging and storing the spores for extended periods while
maintaining spore viability and preventing contamination by degrading
microorganisms. An 18-month shelf life is considered a requirement by
industry, although it would seem that a 6-to-9-month storage life should
be acceptable in this case.
The lack of patentability and the small market size will remain
serious obstacles to the commercialization this biological entity. It
would appear that little opportunity exists to develop a process patent,
because of the ease with which spores of the pathogen can be produced
via conventional fermentation techniques.
With reference to market size, about 125,000 acres of the ap-
proximate 800,000 total acres of rice grown within the three-state area
are estimated to be infested with northern jointvetch (only 90,000 of
which are currently treated with 2,4,5-T). This indicates a potential
annual market for about 3,125kg of wet spore paste (25g/acre). At a
grower cost for the inoculum of $4.00 per acre, this potential market
represents annual sales of $500,000, which are not large by industry
standards. A sizeable portion of the estimated 90,000 lbs of 2,4,5-T
used annually in rice could be displaced by the mycoherbicide, possibly
as much as 60,000 lbs per year. However, complete displacement would
be precluded by the role of 2,4,5-T in controlling weed species other
than northern jointvetch. (A partial displacement of 2,4,5-T could be
accompanied by an increase in the use of 2,4-D and silvex for necessary
broadleaf weed control, but increased use of these herbicides should be
less than the anticipated decrease in use of 2,4,5-T.) "Spot spraying",
the practice of manual application of chemical herbicides in localized
trouble spots within fields, could increase if the mycoherbicide were
available. If so, problems related to the less-troublesome broadleaf
weed species could be handled without the risks imposed by aerial ap-
plications of chemical herbicides.
C. Selected References
The following references, plus personal communication with
indicated experts listed in Appendix A, provides the primary data base
for this chapter.
Daniel, J. T., G. E. Templeton, R. J. Smith, Jr., and W. T. Fox,
"Biological Control of Northern Jointvetch in Rice with an Endemic Fungal
Disease", Weed Science, Vol. 21, pp. 303-307 (1973).
264
-------
Templeton, G. E. , Application for Experimental Use Permit for Field
Testing and Endemic Fungus Colletotrichum gloeosporioides f¦ sp.
aeschynomene as an Experimental Mycoherbicide, to U.S. Environmental
Protection Agency (1974).
Smith, R. J., Jr., J. T. Daniel, W. T. Fox, and G. E. Templeton,
"Distribution in Arkansas of a Fungus Disease Used for Biocontrol of
Northern Jointvetch in Rice", Plant Disease Reporter, Vol. 57, pp. 695-697
(1973).
Barnett, H. L., and B. B. Hunter, Illustrated Genera of Imperfect Fungi,
Third Edition, Burgess Publishing Company, Minneapolis, Minnesota,
p. 200 (1972).
Hurst, H. R., B. A. Huey, and R. J. Smith, Jr., Weeds of Arkansas Rice
Fields, Cooperative Extension Service, University of Arkansas and USDA,
Fayetteville, Arkansas (1973).
Smith, R. J., Jr., and W. D. Shaw, Weeds and Their Control in Rice
Production", USDA Agricultural Handbook 292, U.S. Government Printing
Office, Washington, D.C. (1968).
Smith, R. J., Jr. and G. E. Templeton, "Response of Rice to Phenoxy
Herbicides and Brown Leaf Spot", Arkansas Farm Research, Vol. 17 (3),
p. 4 (1968).
Smith, R. J., Jr., and C. E. Caviness, "Propanil Injury on Soybean
Varieties", Arkansas Farm Research, July-August (1973).
265
-------
Appendix A
PROJECT CONTACTS
The following is a complete list of people who graciously contri-
buted to this project, including panel survey members and persons inter-
viewed. Each individual's affiliation and location is as of the time of
contact during the project; several of these people are currently in
different positions.
Name, Affiliation, Location
Adkisson, Perry
Texas A&M University
College Station, Texas
Alder, Edwin F.
Lilly Research Laboratories
Eli Lilly and Company
Indianapolis, Indiana
Algot, John
P. 0. Box 7797
Fresno, California
Andrilenas, Paul A.
National Economic Analysis Division
USDA - ERS
Washington, D.C.
Antognini, Joseph
Zoecon Corporation
Palo Alto, California
Apple, James
University of Wisconsin
Madison, Wisconsin
Arthur, Wayne
Ciba-Geigy Corporation
Agricultural Division
Greensboro, North Carolina
Pest Control Information Area
Cotton pest control
General, chemicals
Bacteria
Economics, statistics
Insect growth regulators,
pheromones
Corn pest control
General, chemicals
267
-------
Name, Affiliation, Location
Pest Control Information Area
Backman, Paul
Auburn University
Department of Botany and Micro-
biology
Auburn, Alabama
Barras, Stan
USFS
Pineville, Louisiana
Bay, E. C.
Chairman, Department of Entomology
University of Maryland
College Park, Maryland
Bayley, Ned
Office of Science and Education
US DA
Washington, D.C.
Bedard, William
U.S. Forest Service
Berkeley, California
Bell, William J.
Department of Physiology, Cell
Biology and Entomology
University of Kansas
Lawrence, Kansas
Peanut pest control
Forest insects, pheromones
Bacteria
Bacteria
Forest insects, pheromones
Cockroaches, pheromones
Bentley, Michael
Department of Chemistry
University of Maine
Orono, Maine
Mosquitos, oviposition
Birch, Martin
Department of Entomology
University of California
Davis, California
Blackman, Cyril
Clemson University
Blackville, South Carolina
Forest insects, pink bollworm,
pheromones
Soybean pest control
Blickenstaff, Carl
USDA - ARS
Entomology Laboratory
Kimberly, Idaho
Sugarbeet pest control
268
-------
Name, Affiliation, Location
Bode, William M.
Pennsylvania State University
Fruit Research Laboratory
Biglerville, Pennsylvania
Bogard, T. D.
Technical Advisory Service
McLaughlin Gormley King Company
Minneapolis, Minnesota
Boyer, W. P.
USDA Extension Service
University of Arkansas
Fayetteville, Arkansas
Brazzel, J. R.
Methods Development Staff
USDA - APHIS
Hyattsville, Maryland
Bronduik, William J.
North Dakota State Department of
Agriculture
Bismarck, North Dakota
Pest Control Information Area
Viruses
Viruses
Soybean pest control
General, pheromones, viruses
Alfalfa pest control
Buchenar, George Wheat pest control
South Dakota State University
Brookings, South Dakota
Burkholder, Wendell Stored product pests,
Department of Entomology pheromones
University of Wisconsin
Madison, Wisconsin
Bunting, Henry TH 6040
Thompson-Hayward Chemical Company
Kansas City, Kansas
Burgess, H. D. Bacteria
Agricultural Research Council
Glasshouse Crops Research Institute
Littlehampton, Sussex, England
Burton, Vernon E.
Department of Entomology
University of California
Davis, California
Viruses
269
-------
Name, Affiliation, Location
Cameron, Alan
Department of Entomology
Pennsylvania State University
University Park, Pennsylvania
Campan, Edward J.
Lilly Research Laboratories
Eli Lilly and Company
Greenfield, Indiana
Campbell, William
North Carolina State University
Raleigh, North Carolina
Carde, Ring
Department of Entomology
Cornell University
Geneva, New York
Chandler, Herbert
Yolo County Agricultural
Commissioner
Woodland, California
Chanan, Henry
Story Chemical Corporation
Farchan Division
Willoughby, Ohio
Chauthani, Abdul R.
Entomological Research Department
Nutrilite Products, Inc.
Lakeview, California
Clower, Dan F.
Department of Entomology
Louisiana State University
Baton Rouge, Louisiana
Corn, Thomas E.
Fresno County Agricultural
Commissioner
Fresno, California
Coster, Jack E.
School of Forestry
Steven F. Austin University
Nacagdoches, Texas
Pest Control Information Area
Gypsy moth, pheromones
General, chemicals
Peanut pest control
Orchard insects, pheromones
Tomato pest control
Pheromones
Viruses
General, viruses
Citrus pest control
Forest insects, pheromones
270
-------
Name, Affiliation, Location Pest Control Information Area
Couch, Terry L. Viruses
Research Entomologist
Abbott Laboratories
North Orange, Illinois
Craig, George B., Jr. Mosquitos
Biology Department
University of Notre Dame
Notre Dame, Indiana
Crawford, J. L. Corn pest control
Cooperative Extension Service
University of Georgia
Tifton, Georgia
Creighton, C. S. Bacteria
USDA - ARS
Vegetable Insects Laboratory
Charleston, South Carolina
Crovetti, Aldo
Ag-Vet Products Division
Abbott Laboratories
North Chicago, Illinois
Cuthbert, Ray
USDA - USFS
Delaware, Ohio
Daniel, James T.
Department of Plant Pathology
University of Arkansas
Fayetteville, Arkansas
Bacteria
Dutch elm disease, bark
beetles, pheromones
Mycoherbicides in rice
Davich, Ted Pheromones
Boll Weevil Research Laboratory
Mississippi State University
Mississippi State, Mississippi
Day, Boysie General, herbicides
University of California
Berkeley, California
DeAtley, L. S. TH 6040, general
Thompson-Hayward Chemical Company
Kansas City, Kansas
271
-------
Name, Affiliation, Location
deBarjac, Huguette
Institute Pasteur
Paris, France
DeCico, Robert
Thompson-Hayward Chemical
Company-
Kansas City, Kansas
Deems, Robert E.
Agricultural Division
American Cyanamid Company
Princeton, New Jersey
Deer, James
Texas Agricultural Extension
Service
Weslaco, Texas
Diekman, John D.
Zoecon Corporation
Palo Alto, California
Drummond, Paul
Agricultural Chemical Division
FMC Corporation
Middleport, New York
DuImage, Howard T.
USDA - ARS
Brownsville, Texas
Easton, Gene D.
Washington State University
Prosser, Washington
Emmick, Thomas L.
Lilly Research Laboratories
Eli Lilly and Company
Greenfield, Indiana
Engelheimer, Larry ,F.
Farmland Industries, Inc.
Kansas City, Missouri
Pest Control Information Area
Bacteria
TH 6040
General, chemicals
Cole crop pest control
Insect growth regulators
General, chemicals
Viruses, bacteria
Potato pest control
General, chemicals
Pest management
272
-------
Name, Affiliation, Location
Pest Control Information Area
Falcon, Louis A. General, viruses
Division of Entomology-Parisitology
University of California
Berkeley, California
Fast, Paul G. Bacteria
Insect Pathology Research Institute
Sault Ste. Marie, Ontario, Canada
Faust, R. M. Bacteria
USDA - ARS
Plant Protection Institute
Beltsville, Maryland
Fox, Don TH 6040
Thompson-Hayward Chemical
Company
Kansas City, Kansas
Frazier, J. S. Lygus bug, boll weevil,
Department of Entomology pheromones
Mississippi State University
Mississippi State, Mississippi
Frisbie, Ray General, economics, pheromones
Texas Agricultural Extension
Service
College Station, Texas
Fuller, Thomas Weeds in rice
California Department of
Agriculture
Sacramento, California
Gard, Ivan
University of California
Berkeley, California
Viruses
Gaston, Lyle K.
Department of Entomology
University of California
Riverside, California
Pink bollworm, cabbage looper,
pheromones
Gates, Robert L. General, chemicals
Agricultural Chemical Division
FMC Corporation
Middleport, New York
273
-------
Name, Affiliation, Location
Pest Control Information Area
Goring, Cleve A. I. General, chemicals
Ag-Organics Department
Dow Chemical U.S.A.
Midland, Michigan
Greenlee, Kenneth Pheromones
Chemical Samples Company
Columbus, Ohio
Gregory, W. W., Jr. Corn and tobacco pest control
Agricultural Science Center
University of Kentucky
Lexington, Kentucky
Grimble, D. G. Bacteria
School of Forestry
State University of New York
Syracuse, New York
Grimes, Walter H. General, chemicals, viruses
Chemagro Agricultural Chemicals
Division
Mobay Chemical Corporation
Kansas City, Missouri
Gueldner, Richard G. Pheromones
Boll Weevil Research Laboratory
USDA - ARS
Mississippi State, Mississippi
Gutierrez, A. P. Pest management
Division of Biological Control
University of California
Albany, California
Gyrisco, George G. Corn pest control
Department of Entomology
Cornell University
Ithaca, New York
Hardee, D. D. Pheromones
Story Chemical Corporation
Starkville, Mississippi
Harper, James D. Viruses
Department of Zoology-Entomology
Auburn University
Auburn, Alabama
274
-------
Name, Affiliation, Location Pest Control Information Area
Harwood, Robert Wheat pest control
Washington State University
Pullman, Washington
Hedin, Paul Pheromones
Boll Weevil Research Laboratory
USDA - ARS
Mississippi State, Mississippi
Hefner, R. E. General, chemicals
Ag-Organics Department
Dow Chemical U.S.A.
Midland, Michigan
Henning, Robert J. Peanut pest control
Cooperative Extension Service
University of Georgia
Tifton, Georgia
Hill, Kenneth
Agricultural Division
American Cyanamid Company
Princeton, New Jersey
Hink, W. F.
Department of Entomology
Ohio State University
Columbus, Ohio
Hoelscher, Clifford
Agricultural Extension Service
Stephenville, Texas
Hoger, H.
Minnesota Department of Agriculture
St. Paul, Minnesota
Hoyt, S. C.
Tree Fruit Research Center
Washington State University
Wenatchee, Washington
Hunt, Thomas Corn pest control
Survey Entomologist
North Carolina State University
Raleigh, North Carolina
General, chemicals
General, viruses
Peanut pest control
Corn pest control
Bacteria
275
-------
Name, Affiliation, Location
Pest Control Information Area
Ignoffo, C. M. Viruses, bacteria
Biological Control of Insects
Research
USDA - ARS
Columbia, Missouri
Jaques, R. P. Viruses
Research Station
Harrow, Ontario, Canada
Johnson, Albert Tobacco pest control
Pee Dee Experiment Station
Florence, South Carolina
Kaya, H. K. Bacteria
Connecticut Agricultural
Experimental Station
New Haven, Connecticut
Keith, David L. Corn pest control
Department of Entomology
University of Nebraska
Lincoln, Nebraska
Kennedy, Donald General, viruses
Biology Department
Stanford University
Stanford, California
Klassen, W. General
National Program Staff
USDA - ARS
Beltsville, Maryland
Klingman, Glenn C. General, chemicals
Greenfield Laboratories
Eli Lilly and Company
Greenfield, Indiana
Klun, J. A. European corn borer,
USDA, Corn Borer Investigations pheromones
Ankeny, Iowa
Krister, Charles J. General, chemicals
Biochemicals Department
E. I. DuPont de Nemours & Company,
Inc.
Wilmington, Delaware
276
-------
Name, Affiliation, Location
Kuhlman, Donald
University of Illinois
Urbana, Illinois
Laird, Marshall
Research Unit on Vector Pathology
Memorial University of Newfoundland
St. John's, Newfoundland, Canada
Lawrence, Kathryn A.
Midwest Research Institute
Kansas City, Missouri
Leonard, David
Department of Entomology
University of Maine
Orono, Maine
Lewis, F. B.
Forest Insect & Disease Laboratory
USDA - USFS
Hamden, Connecticut
Libby, John
University of Wisconsin
Madison, Wisconsin
Lloyd, F. P,
Boll Weevil Research Laboratory
USDA - ARS
Mississippi State, Mississippi
Lofgren, John
University of Minnesota
St. Paul, Minnesota
Lowen, Warren
Biochemicals Department
E. I. DuPont de Nemours & Company,
Inc.
Wilmington, Delaware
Lukefahr, M. J.
USDA - ARS
Brownsville, Texas
Maxwell, Fowden G.
Department of Entomology
Mississippi State University
Mississippi State, Mississippi
Pest Control Information Area
Corn pest control
Public health pest control
Bacteria
Gypsy moth, parasites
Viruses
Potato pest control
Pheromones
General
General, chemicals
Viruses
Viruses
277
-------
Name, Affiliation, Location
MacSwan, Iain
Oregon State University
Corvallis, Oregon
Mathe, Donald
Montana State University
Bozeman, Montana
McBride, Dean
North Dakota State University
Fargo, North Dakota
McClaughlin, Roy
USDA Boll Weevil Research
Laboratory
Mississippi State University
Mississippi State, Mississippi
Mello, Matthew
Santa Cruz County Agricultural
Commissioner
Watsonville, California
Pest Control Information Area
Apple pest control
Wheat pest control
Sugarbeet pest control
Cotton pest control
Cole crop pest control
Metterhouse, W. Bacteria
New Jersey Department of
Agriculture
Trenton, New Jersey
Morris, 0. N. Viruses
Chemical Control Research Institute
Canada Department of Environment
Ottawa, Ontario, Canada
Morrison, Kenneth Wheat pest control
Washington State University
Pullman, Washington
Newsom, Dale General, pheromones, viru
Department of Entomology
Louisiana State University
Baton Rouge, Louisiana
Nicholson, John Pest management
IPM consultant
Shafter, California
278
-------
Name, Affiliation, Location Pest Control Information Area
Noon, Zenas B., Jr. Viruses
Sandoz, Inc.
Homestead, Florida
Norgarrd, Richard Economics
Gianinni Foundation
University of California
Berkeley, California
Norris, J. R. Bacteria
Shell Research Limited
Borden Micorbiological Laboratories
Sittingbourne, Kent, England
Payne, Thomas L.
Department of Entomology
Texas A&M University
College Station, Texas
Peacock, John
U.S. Forest Service
Delaware, Ohio
Southern pine beetle,
pheromones
Dutch elm disease, bark
beetles, pheromones
Peters, Leroy Sorghum pest control
South Central Experimental Station
Clay Center, Nebraska
Phillips, William Research, viruses
EPA, Criteria and Evaluation
Division, Office of Pesticide
Programs
Washington, D.C.
Pittman, Gary
Boyce Thompson Institute
Grass Valley, California
Pitts, C. N.
Department of Entomology
Kansas State University
Manhattan, Kansas
Forest insects, pheromones
Face fly, pheromones
Potter, Howard Wheat pest control
Michigan State University
East Lansing, Michigan
279
-------
Name, Affiliation, Location Pest Control Information Area
Pulliam, Jean Research, viruses
EPA, Criteria and Evaluation
Division, Office of Pesticide
Programs
Washington, D.C.
Rabb, Robert L.
Department of Entomology
North Carolina State University
Raleigh, North Carolina
Re tan, Arthur
Washington State University
Pullman, Washington
Ridgway, Richard L.
USDA - ARS
College Station, Texas
Viruses
Apple pest control
General, pheromones, viruses
Roberts, James Peanut pest control
Virginia Polytechnic Institute and
State University
Blacksburg, Virginia
Robertson, Robert Tobacco pest control
North Carolina State University
Raleigh, North Carolina
Roelofs, Wendell L. Orchard insect pests,
Department of Entomology pheromones
Cornell University
Geneva, New York
Rogoff, Martin Bacteria, viruses
EPA, Registration Division
Office of Pesticide Programs
Washington, D.C.
Roselle, Robert Wheat pest control
University of Nebraska
Cooperative Extension Service
Lincoln, Nebraska
Rudd, L. A., Jr. Viruses, pest management
California Agricultural Services
Kerman, California
280
-------
Name, Affiliation, Location
Pest Control Information Area
Schneider, Charles
Michigan State University
East Lansing, Michigan
Schuldt, Paul H.
Agricultural Division
Diamond Shamrock Chemical Company
Painesville, Ohio
Sugarbeet pest control
General, chemicals
Semtner, Paul
Southern Piedmont Research and
Continuing Education Center
Virginia Polytechnic Institute and
State University
Blackstone, Virginia
Seymour, Lon L.
Sandoz, Inc.
Homestead, Florida
Tobacco pest control
Bacteria
Shadbolt, C. Allen
Thompson-Hayward Chemical
Company
Kansas City, Kansas
Shieh, P.
Sandoz, Inc.
Homestead, Florida
Bacteria
Viruses
Shorey, H. H.
Department of Entomology
University of California
Riverside, California
Shurtleff, Malcolm C.
Plant Pathology
University of Illinois
Urbana-Champaign, Illinois
Silverstein, R. M.
Department of Chemistry
Syracuse University
Syracuse, New York
Slife, Frederick
Department of Agronomy
University of Illinois
Urbana, Illinois
Pink bollworm, pheromones,
cabbage looper
Corn pest control
Pheromones
Herbicides
281
-------
Name, Affiliation, Location
Pest Control Information Area
Smirnoff, W.A.
Luarentian Forest Research Centre
Environment Canada
Quebec, Quebec, Canada
Bacteria
Smith, Nicholus
Michigan State University
East Lansing, Michigan
Corn, wheat pest control
Smith, Ray J.
USDA - ARS
Rice Branch Experiment Station
Stuttgart, Arkansas
Mycoherbicides in rice
Snow, John
Louisiana State University
Baton Rouge, Louisiana
Cotton pest control
Sokolov, Phillip G.
Department of Biological Sciences
University of Maryland
Catonsville, Maryland
Circadian rhythms
Soper, Richard S.
Bacteria
USDA - ARS
New England Plant, Soil, and Water
Laboratory
University of Maine
Orono, Maine
Sorensen, Anne Cotton pest control
University of California
Berkeley, California
Southwood, T. R. E. Viruses
Department of Zoology and Applied
Entomology
Imperial College
London, England
Stairs, Gordon R. Viruses
Department of Entomology
Ohio State University
Columbus, Ohio
Stark, R. W. Bacteria
College of Forestry, Wildlife, and
Range Science
University of Idaho
Moscow, Idaho
282
-------
Name, Affiliation, Location Pest Control Information Area
Stetson, G. G. General, chemicals
Chemagro Agricultural Chemicals
Division
Mobay Chemical Corporation
Kansas City, Missouri
Stevenson, Walter Tomato pest control
Purdue University
West Lafayette, Indiana
Stewart, Fred D. Viruses
Western Cotton Research Laboratory
USDA - ARS
Phoenix, Arizona
Sweat, Mike
Farmland Industries, Inc.
Kansas City, Missouri
Swisher, Ely
Rohm and Haas Company
Philadelphia, Pennsylvania
Templeton, George E.
Department of Plant Pathology
University of Arkansas
Fayetteville, Arkansas
Tettle, James
Department of Entomology
New York State Agriculture
Experimental Station
Geneva, New York
Thompson, C. G.
Forestry Sciences Laboratory, USFS
Corvallis, Oregon
Tinsley, T. W. Viruses
Unit of Invertebrate Virology
Commonwealth Forestry Institute
Oxford, England
Tumlinson, James Pheromones
Insect Attractants and Basic
Biology Laboratory
USDA - ARS
Gainesville, Florida
Pest management
General, chemicals
Mycoherbicides in rice
Orchard insect pests,
pheromones
Viruses
283
-------
Name) Affiliation, Location
Tynes, James
Louisiana State University
Extension Service
Baton Rouge, Louisiana
Pest Control Information Area
Cotton pest control
Vail, P.
Western Cotton Research Laboratory
USDA - ARS
Phoenix, Arizona
Van den Bosch, Robert
Division of Biological Control
University of California
Albany, California
Van Tassell, B.
Nutrilite Products, Inc.
Buena Park, California
Walker, Kenneth
USDA - ARS
Washington, D.C.
Watanabe, Warren H.
Rohm and Haas Company
Spring House, Pennsylvania
Wells, Frank E.
Midwest Research Institute
Kansas City, Missouri
Westall, E. G.
Lakeview Operations
Nutrilite Products, Inc.
Lakeview, California
Viruses
General, viruses
Viruses, bacteria
General, minor pesticide uses
General, chemicals
Viruses
Viruses
Wiley, Vincent Pheromones
Chemical Samples Company
Columbus, Ohio
Wilson, H.
Department of Entomology
Cornell University
Geneva, New York
Orchard insect pests,
pheromones
Wong, Dr. General, chemicals
Agricultural Division
American Cyanamid Company
Princeton, New Jersey
284
-------
Name, Affiliation, Location
Pest Control Information Area
Wood, David L.
Department of Entomology
University of California
Berkeley, California
Yearin, W. C.
Department of Entomology
University of Arkansas
Fayetteville, Arkansas
Yendol, William G.
Pesticide Research Laboratory
Pennsylvania State University
University Park, Pennsylvania
York, Alan C.
Department of Entomology
Purdue University
West Lafayette, Indiana
Young, Seth
Department of Entomology
University of Arkansas
Fayetteville, Arkansas
Forest insects, pheronomes
Viruses
Viruses
Tomato pest control
Viruses
285
-------
Appendix B
CROP-PEST SURVEY
The main body of this report assesses the commercial feasibility of
new innovative pesticides and the probable rate of introduction of such
materials to the marketplace from the developer/supplier point of view.
As an alternative to this perspective, an extensive questionnaire survey
of over 170 specialists familiar with crop-pest problems was conducted,
covering 15 crops in 34 states. Figure B-l at the end of this Appendix
presents a sample copy of the questionnaire survey.
Unfortunately, the results of this' survey were not highly conclusive.
Only 40 percent of those surveyed responded, and even fewer could provide
any input directly relevant to the future of specific innovative products.
However, the completed questionnaires contained valuable opinions about
anticipated trends in certain crop markets for pest control. A summari-
zation of some of these opinions suitable for generalization, plus com-
pilations of responses for specific crop areas, are presented in the
following.
A- Selected General Findings
Chapter I, "Summary and Conclusions," incorporates information
gathered through the survey. This section considers general findings on
acreage, crop diseases, and insect pests, and presents some detail on
specific crop-pest complex areas.
1. Acreage
Respondents generally expected acreage for most of the 15 crop
(which represented 82 percent of the total 326 million harvested U.S. acres
in 1974) to increase by 1985. However, it is improbable that acreages
will expand simultaneously in view of the limited availability of quality
reserve land in this country. Weather, irrigation, double cropping,
economics, world food needs, and pressure for available land were listed
as critical factors influencing expected acreage levels for any crop
within a given growing season. For example, respondents indicated such
potentially influential factors as (1) glandless cotton varieties, if
commercial developed, would be expected to substantially increase cotton
acreage, (2) the release of allotments on peanut acreages, combined with
an expected increase in world protein demand would cause an increase in
acreage, (3) higher density planting of apple orchards, and higher
yielding varieties of cole crops would result in little if any increase
in acreage.
287
-------
2. Control of Selected Crop Diseases
Stem rust and leaf rust on wheat, and foliar disease of corn
were identified as diseases with little or no alternative controls if
EBDC fungicides are deregistered. A number of experts anticipated
increasing soil disease problems as a result of cultural practices
involving stubble mulching and minimum tillage. Respondents expected
these and other diseases to be controlled with systemic fungicides and
via resistant crop varieties by 1985. They expected increased commercial
use of systemics, although expensive, because of the fewer number of
applications required at reduced rates. In addition, systemic rather
than foliar fungicides would be preferred for crops requiring irrigation.
Almost no respondents expected the so-called "new generation" products
to be used for the control of plant diseases.
3. Control of Selected Insect Pests
Wireworm control on corn and wheat, and white grub control on
corn were identified as the crop-pest problems most seriously lacking
alternative controls if present insecticides were deregistered. Many
respondents expected that (1) the development of insect resistance to
current insecticides, and (2) public concern with the safety aspects of
their use, would be the basic factors influencing future pesticide devel-
opment. Most respondents expected future insect control to operate
increasingly within a framework of integrated pest management (IPM)
techniques.
Respondents made too few projections on the future levels of pest
control iexpected with developmental (unregistered) products (classes B
through H in the ^questionnaire) to provide statistical validity. At
best, these responses indicated that such products are generally
recognized candidates for the future. Table B-l summarizes the esti-
mates received of the level of control expected as a percentage of the
average cost of control per treated acre in 1985 (for all except regis-
tered "second generation" products).
B. Summary of Questionnaire Responses for Specific Crops
1. Corn
A total of twenty-four questionnaires were mailed, and replies
were received from ten respondents, half of whom elected not to partici-
pate in the survey. The five responses were considered insufficient to
provide statistically valid data on anticipated trends. However, there
were two significant findings—an overlap of the pests considered to be
current and future pest control problems, and the pests identified as
lacking alternative control if selected pesticides are deregistered.
The overlap in pests included: foliar disease, wireworms, and white
grubs.
288
-------
Table B-l
RESPONDENTS EXPECTATIONS OF 1985 LEVELS OF PEST CONTROL BY PESTICIDE CLASS FOR SELECTED CROPS AND PESTS
(Percentage of Average Total Annual Control Cost per Treated Acre)3
Pesticide Class
Crap
00
v0
Class C
Black cutvorm, European corn
borer and armyvorro: 57. of
total annual control
Bo11worm, budworm, fleahopper,
and lygus: 50% of total an-
nual control (pyrethroids and
ovicides)
Class D
Wheat
Soybeans
Tobacco
Peanuts
Sorghum
Sugarbeets
Tomatoes
Cole crops
Potatoes
Apples
Aphids:
control
100% of total annual
Corn earworm: 45% of total
annual control
Budworm, hornworm, and cut-
worms: 10% of total annual
control
All imaature insect forms:
35*40% of total annual con-
trol (TH 6040)
Aphids: 5% of total annual
control
Aphids: 10Z of total annual
control; Wireworm: 20Z of
total annual control;
Colorado potato beetle: 10Z
of total annual control
Aphids, San Jose scale: 50Z
of total annual control;
Codling moth; 30Z of total
annual control; McDaniel
site: 40Z of total annual
control
Black cutworm, European corn
borer, and armyworm: 15% of
total annual control
Bollworm, budworm, and loopers:
25% of total annual control
Grasshoppers: 2% of total
annual control (viruses)
Budworm, hornworm, cutworm, and
fle&beetles: 20% of total annual
control (B. thuringiensis)
Burrowing bug: 25% of total an-
nual control
Two-spotted spider mite: 50%
of total annual control
Class E
Class F
Lepidoptera: 25% of
total annual control
Corn leaf aphid:
57» of total annual
control
Grasshoppers: 20% of
total annual control
(protozoa)
Class H
Budworm, hornworm:
4% of total annual
control
Lesser cornstalk
borer: 10% of total
annual control
Sorghum greenbug:
95% of total
annual control
Lepidoptera: 10% of total annual
control
Lepidoptera: 30% of total annual
control (B. thuringiens is: NPV)
Lepidoptera: 15% of total annual
control (NPV)
Leafhopper: 90% of
total annual control
Root maggot: 5% of
total annual control
Mites: 25% of total
annual control
Scale: 50% of total
annual control
Mites: 40% of total
annual control
*Class C:
Class D:
Class E:
Novel new chemical products, e.g., juvenile
hormones and analogs* chitln inhibitors,
ovicides, ecdysones.
Microbials, e.g., viruses, bacteria.
Pheroaones.
Class H;
Controlled release of
natural parasites and
predators.
Other.
(Classes B and G were not reported)*
t>Tt»ese data are expressed as the percent of the total annual cost (all methods) of controlling
the pest in question that was expected from the Indicated class of "innovative" products.
-------
In addition to these pests, one expert anticipated corn root-
worm resistance to organophosphates and carbamates by 1985, and expected
that a new product for these pests would be needed- In the area of
disease control, one expert emphasized that there is no practical treat-
ment for diseases such as stalk rot, smut, and leaf diseases. Resistant
varieties and cultural' practices are the only means employed to suppress
these diseases.
Selected data on anticipated disease and insect pest problems
in three states appear in Tables B-2 and B-3.
2. Cotton
Of the fourteen questionnaires sent out, four responses were
received, with one respondent electing not to participate in the survey.
The low response rate did not justify statistical analysis of the opin-
ions received. The respondents did, however, specifically mention the
following factors:
• Any practical development of glandless cotton vari-
ties would stimulate increases in cotton acreage.
• Microbial control of bollworms, budworms, and loopers
was expected to represent 25% of total market for
these pests by 1985.
• Effective control of boll rot is needed.
3. Wheat
Ten of the eighteen experts surveyed responded, two of whom
elected not to participate. Respondents' projections of planted acreage
for 1980 and 1985 indicated no significant change from 1974 acreage.
Specific comments pertinent to acreage were:
(1) Saline seep may decrease acreage; continuous crop-
ping may increase acreage in parts of Montana.
(2) Anticipated world demand will convert pasture to
wheat acreage in South Dakota.
(3) Irrigation possibilities may cause replacement of
wheat by specialty crops in Washington.
The disease organisms for which treatment on the largest num-
ber of acres was anticipated included smut (identified either as smut or
flag smut) and rust (identified as rust, stem rust, leaf rust, or stripe
rust). One respondent anticipated an increase in disease problems with
adoption of stubble mulch and minimum tillage cultural practices.
290
-------
Table B-2
CORN DISEASES IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Acres
(thousands)
1974
1980
1985
Disease / Organi sin Stated Infested Treated Infested Treated Infested Treated
Foliar disease
Seed decay
Seedling blight
Smut
Stalk rot
Stalk rot
Root rot
Michigan 852-1065
Minnesota 400
Michigan
Many
Minnesota 69
Minnesota 2300
Michigan
2130
0
0
Most
0
0
0
800-1000
300
Many
50
2000
2000
0
n. a.
Most
n. a.
n.a.
n.a.
760-950
300
Many
50
1800
1900
0
n. a.
Most
n. a.
n. a.
n. a.
n.a. = not available.
~
Number of respondents = 5.
^Data for only the states for which completed questionnaires were obtained.
-------
Table B-3
CORN INSECT PESTS IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Acres
(thousands)
1974
1980
1985
Estimated
Control
Cost
(dollars per
treated acre)
Pest
Stated
Infested
Treated
Infested
Treated
Infested
Treated
1974
1980
1985
Armyworm
Wisconsin
2
2
10
10
10
10
n. a.
n.a.
n. a.
True annyvora
Illinois
20
15
20
15
20
15
$2.50
$5.00
$5.00
False arm/worm
Illinois
140
115
70
50
70
50
3.00
6.00
6,00
Corn flea
Illinois
90
85
75
65
75
65
2.00
4.00
5.00
beetles
Corn leaf aphid
Wisconsin
5
5
5
5
5
5
n. a.
n.a.
n. a.
Illinois
220
200
300
250 "
300
250
2.00
4.00
5.00
Corn rootwonn
Wisconsin
1,900
1,900
2,000
2,000
2,000
2,000
n. a.
n. a.
n. a.
Illinois
2,000
6,000
2,000
4,000
2,500
4,000
3.20
6.00
7. 50
Minnesota
2,375
1,560
2,400
1,600
2,450
1,650
3.80
3.80
3.80
Cutworms
Wisconsin
20
8
20
10
20
10
n.a.
n.a.
n. a.
Illinois
70
60
150
120
150
120
5.00
8.00
8.00
Minnesota
50
32
50
35
50
35
0.08
0.08
0.08
European corn
Wisconsin
10
10
10
10
10
10
n. a.
n. a.
n. a.
borer
Illinois
170
130
200
150
200
150
3,00
6.00
7.00
Minnesota
47
16
45
12
45
10
3.80
3.80
3.80
White grubs
Wisconsin
20
5
20
5
20
5
n. a.
n.a.
n. a.
Wireworm
Wisconsin
500
20
500
50
500
50
n. a.
n.a.
n. a.
Illinois
3
3
3
3
3
3
2.25
4.50
5.00
Minnesota
75
49
70
50
70
50
0.12
0.12
0. 12
n.a. " not available.
*
Number of respondents = 5.
^Data for only the states for which completed questionnaires were obtained.
-------
Respondents anticipated the following methods to control all
of the identified diseases at the indicated percent of the total annual
average cost of control per treated acre:*
(1) Class A (registered "second generation" pesticides):
control at 80 to 100 percent in 1974 and 1980; and
60 to 100 percent by 1985.
(2) Class B (new chemical products of the same general
type as those in Class A): no control in 1974; 20
to 30 percent control in 1980; and 40 to 50 percent
control by 1985.
(3) Class H ("other"—in this case, resistant wheat
strains): control at 10 to 20 percent in 1974, 10
to 100 percent in 1980, and 40 to 100 percent by
1985 (with 100 percent use for control of rust).
No significant information on anticipated average cost of con-
trol was included with any of the estimates.
Specific comments on the type of control anticipated and the
factors affecting development of pesticide products included:
(1) Growers lack recognition of disease problems and
loss.
(2) Private chemical companies search for products to
control foliar disease instead of the more diffi-
cult problems such as root rot.
The respondents expected the following insect problems to
require the most attention:
(1) Wireworms, requiring control in 1974, 1980, and
1985 on 1,508 to 1,515 thousand of the 2,500
thousand infested acres covered by responses.
(2) Grasshoppers, requiring treatment of 385 thousand
of the 1,200 thousand infested acres covered by
responses in 1974; 320 thousand of the 1,140
thousand in 1980; and 325 thousand out of 1,140
thousand acres in 1985.
(3) Cutworms (including army cutworm and pale western
cutworm), requiring treatment on 101 thousand of
221 thousand acres covered by responses in 1974,
and on 107 thousand of 210 thousand acres in 1980
and 1985.
*These data are expressed as the percent of the total annual cost (all
methods) of controlling the pest(s) in question that was expected from
the indicated class of products.
293
-------
The cost per treated acre for control of these primary insect
pests from 1974 to 1985 in terms of 1974 dollars was expected to increase
from $3.00 to $5.00 for cutworms, from $0.50 to $2.00 for wireworms, and
from $2.00 to $5.00 for grasshoppers.
Respondents anticipated the following methods, at the indicated
percent of total average annual cost of control (all methods) per treated
acre, to control all of the identified insects:
(1) Class A (registered "second generation" pesticides) :
90 to 100 percent of control in 1974; 50 to 100 per-
cent by 1980; and 40 to 100 percent by 1985.
(2) Class B (new chemical products of the same general
type as in Class A ) : 90 to 100 percent control in
1974 (grasshoppers); 10 to 100 percent control by
1980 and 1985 (most frequently 10 to 30 percent for
control of greenbug and wireworm).
(3) Class C (novel new chemical products): no control
use in 1974; 10 to 50 percent by 1980 (aphid, wire-
worm, and greenbug); and 10 to 70 percent by 1985
(wireworm, greenbug, and grasshoppers).
(4) Class D (microbials): no control use in 1974; 5 to
30 percent by 1980 and 1985 (armyworm, cutworm;
virus control of grasshoppers).
(5) Class F (controlled release of natural parasites and
predators): no control use in 1974; 10 to 20 percent
by 1980; and 70 to 80 percent use by 1985 for grass-
hopper control through protozoans.
(6) Class H ("other," represented by resistant wheat
varieties): 90 to 100 percent control for use
against wheat stem sawfly, 1974 through 1985; 10
percent control by 1980, and 40 percent by 1985 for
wheat stem maggot; 50 percent control by 1985 for
greenbug.
Specific comments on the use of particular types of control
products included:
(1) Effective control of wheat stem maggot is needed.
(2) Controlled release of sterile insects is too costly.
(3) Controlled release of natural predators and para-
sites is ineffective.
(4) The pale western cutworm, grasshoppers, and the
wheat stem sawfly probably do not have pheromones.
Selected data on anticipated disease and insect pest problems
in six states appear in Tables B-4 and B*-5.
294
-------
Table B-4
WHEAT DISEASES IDENTIFIED BY SURVEX RESPONDENTS
1974, 1980, 1985
Acres
(thousands)
1974
ro
so
ui
Disease/Organism
Cephalosporin* strip*
Cercosporella foot rot
Cotnn bunt
Dwarf bunt
Flag aaut
Leaf rot
Leaf rust
Nematodes
Root rots
Rust
Rust, aaut, leaf diseases
Septoria
Ssuts
Snow mold
Stes rust
Stripe rust
Wheat spindle streak
mosaic virus
Wheat streak mosaic
Yellow dwarf
Yellow leaf blotch
Washington
Montana
Washington
Washington
Washington
Washington
Washington
South Dakota
South Dakota
Michigan
Montana
Michigan
South Dakota
Montana
Washington
South Dakota
Wash ingtoa
Michigan
Washington
Washington
South Dakota
Infested Treated
n.a. n.a.
1,000 0
n.a. n.a.
n.a. n.a.
a. a. n, a.
n.a. n.a.
n.a. n.a.
100 0.25
3*000 0
800 0
5,000 0
950 0
100 0
A,000 3,500*
n.a. n.a.
5 0.25
n. a. n.a.
750 100*
1980
1985
n. a.
n.a.
200
n.a.
h.a.
0
Infested Treated
n.a.
n.a.
o. a.
n. a.
200
n.a.
n.a.
n.a. n.a.
n.a. n.a.
n.a. n.a.
n.a. n.a.
n.a. n,a.
500 1
3,000 25
750 50
6,000 50
950 0
500 1
5,000 4,500*
n.a. n.a.
5 1
n.a. n.a.
500 40o'
n. a.
n. a.
1
Infested
n.a,
n, a.
n, a.
n. a.
n. a.
n. a.
n. a.
500
3,500
700
6,000
850
500
5,500
n. a.
5
n. a.
450
n. a.
n.a.
500
Treated
n.a.
n.a.
n. a.
n.a.
o. a.
n.a.
n. a.
1
50
100
75
0
1
5,000*
n. a.
1
a.a.
400J
n.a. * not available.
*
Number of respondents * 8.
^Data for only the states for which completed questionnaires vere obtained-
*Seed treatment.
5
Resistant varieties planted.
Estimated Control Cost
(dollars per treated acre)
1974 1980 1985
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. xv. a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a,
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. $10.00 $12.00
n.a. n.a. n.a.
n.a. n.a. n.a.
$0.25 $0.50 $0.75
n.a. n.a. n.a.
n.a. n.a. n.a,
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
-------
Table B-5
to
VO
O
WHEAT INSECT PESTS IDENTIFIED BY SURVEY RESPONDENTS
(1974, 1980, 1985)
Acres
(thousands)
Estimated
Control
Cost
(dollars per
1974
1980
1985
Pest
State^
Infested
Treated
Infested
Treated
Infested
Treated
1974
1980
1985
Aphids
Minnesota
60
40
50
25
50
25
$3. 00
$4.00
$5.00
Army cutworm
Montana
150
65
150
65
150
65
3.50
4. 50
5.00
Minnesota
Trace
0
10
6
10
5
n. a.
3.00
5.00
Cutworms
South Dakota
10
5
10
5
10
5
3.00
4.00
4.00
Minnesota
10
4
10
6
10
5
3.00
3.00
5.00
English grain
Washington
10
5
n. a.
n.a.
n.a.
n. a.
3.00
n.a.
n. a.
aphid
Grasshoppers
Montana
1,000
300
1,000
300
1,000
300
3.00
4.00
5.00
Michigan
100
25
100
n. a.
100
n. a.
4.00
n. a.
n. a.
Minnesota
100
60
40
20
40
25
2.00
3.0*0
5.00
Greenbug
Washington
75
50
50
25
50
25
2.50
3.50
4.50
South Dakota
10
5
10
5
10
5
3.00
4.00
4.00
Hessian fly
Montana
50
0
n. a.
n.a.
n. a.
n. a.
0
n. a.
n. a.
Inherent curl mite
South Dakota
25
0
25
25
25
30
0
4.00
4.00
Pale western
Montana
30
25
30
25
30
25
3,50
4.50
5.00
cutworm
Nebraska
21
2
n. a.
n. a.
n. a.
n. a.
3.85
n. a.
n. a.
Wheat stem sawfly
Montana
50
0
n. a.
n. a.
n. a.
n. a.
n, a.
n. a.
n. a.
Minnesota
500
*
400
*
200
*
n. a.
n.a.
n. a.
Wireworms
Washington
2,000
1,000§
2,000
1,000§
2,000
1,000^
1.00
1. 25
1.50
Montana
500
500
500
500
500
500
.50
.50
. 50
Minnesota
20
8
20
10
20
15
<1.00
1.00
2.00
n.a. = not available.
*
Number of respondents = 8.
^Data for only the states for which completed questionnaires were obtained.
*
Resistant varieties planted.
Seed treatment.
-------
4. Soybeans
Four of the twenty experts included in the soybean survey
responded, and one of these elected not to participate. This level of
response did not provide sufficient data to develop significant forward
pest control trends. Generally pertinent statements included:
(1) Emphasis on other crops and pressure on available
land may cause decrease in soybean acreage beyond
1985 in South Carolina.
(2) A shift is anticipated from fumigant to contact
nematicides if the cost of these chemicals proves
competitive.
(3) Rice-soybean rotation systems could result in
reduced soybean acreage in Arkansas.
5. Tobacco
Three of the eleven tobacco experts responded to the survey.
The response level is not significant. Specific remarks concerning
future pest control follow:
(1) Domestic acreage is a function of worldwide demand.
(2) "Artificial" tobacco was not expected to be compet-
itive in the tobacco market.
A
(3) Class B products such as methomyl, fensulfothion,
carbofuran, monocrotophos, acephate, and disulfoton
will account for 76 and 57 percent of the respec-
tive 1980 and 1985 tobacco insect control markets.
(4) Class C, D, E, F, G, and H products* will account
for 24 and 43 percent of the respective 1980 and
1985 tobacco insect control markets,+ specifically
for budworm and hornworm control. Specificity and
safety factors will contribute to their use.
Selected data on anticipated disease and insect pest problems
in two states appear in Tables B—6 and B—7.
6. Peanuts
Ten of the twelve experts included in the survey replied, two
of whom elected not to participate in the survey. Harvested acreage
was expected to increase; one expert predicted that release of allotment
*See definitions in survey questionnaire in Figure B-l.
^See footnote b. to Table B-l.
297
-------
Table B-6
TOBACCO DISEASES IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Estimated
Control
Acres Cost
(thousands) (dollars per
Disease/ 1974 1980 1985 treated acre)
Organism
State"*"
Infested
Treated
Infested
Treated
Infested
Treated
1974
1980
198 j
Black shank
North Carolina
350
300
350
250
400
225
$28
$35
$35
Granular wilt
North Carolina
350
300
350
250
400
225
28
35
35
Root knot
North Carolina
350
300
350
250
400
225
28
35
35
nematodes
•k
Number of respondents = 3.
^Data for only those states for which completed questionnaires were obtained.
-------
Table B-7
TOBACCO INSECT PESTS IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Acres
(thousands)
1974 1980 1985
Pest State"'" Infested Treated Infested Treated Infested Treated 1974 1980 1985
Aphids
South
Carolina
10
2
11
2
12
2
$ 5
$ 6
$ 9
Budvorm
North
Carolina
400
150
410
100
450
100
7
9
9
South
Carolina
80
78
88
85
92
88
18
20
21
Cutworms
South
Carolina
40
20
44
22
46
23
7
8
n. a
Flea beetles
North
Carolina
300
10
400
5
400
5
5
6
12
South
Carolina
78
60
85
66
89
69
10
11
9
Green peach
North
Carolina
5
2
5
2
5
2
7
9
7
aphid
Hornworms
North
Carolina
150
10
200
5
200
5
7
9
9
South
Carolina
80
78
88
85
92
88
10
11
12
Wireworm
North
Carolina
150
100
150
100
200
100
10
10
10
South
Carolina
60
78
66
85
69
88
11
12
13
n.a. ¦ not applicable.
*
Number of respondents = 3.
t
Data for only those states for which completed questionnaires were obtained.
Estimated
Control
Cost
(dollars per
treated acre)
-------
controls by 1976 or 1977 would cause an acreage increase that might be
limited by equipment availability.
Leaf spot, pod rot, and nematodes required treatment on most
acreage infested in 1974 and were expected to.continue as principal
disease problems. Control of these diseases in 1974 involved 100 per-
cent use of Class A or Class B products. The level of control by 1985
was estimated at 10 to 20 percent with Class C products, 50 percent of
leaf spot control with Class D products, and 20 percent nematode control,
with Class D products. Class H products, specifically resistant peanut
species, were estimated at a level of 40 to 50 percent for pod rot, 25
to 40 percent for leaf spot, and 10 percent for nematode control. All
respondents indicated a 40 to 60 percent increase in the cost of control
by 1985.
The principal insects requiring control included the lesser
cornstalk borer and thrips. These insects, along with the southern
corn rootworm and potato leafhopper, were anticipated as future insect
pests requiring control. The 90 to 100 percent level of control attrib-
uted to Class A and B products in 1974 was expected to decrease by 1985
to a level of 20 to 50 percent for lesser cornstalk borer and 80 percent
for thrips. Class C products, especially TH 6040, were estimated as
representing 40 percent of the 1985 insect control market. The use of
pheromones for control of the lesser cornstalk borer was estimated at
10 percent for this market by 1985.
Selected data on anticipated disease and insect pest problems
in four states appear in Tables B-8 and B-9.
7. Alfalfa
A total of six questionnaires were mailed in the alfalfa sur-
vey with replies received from two experts. The low response level
only warrants mention that grasshopper and alfalfa weevil are problems,
with natural control adequate for alfalfa weevil in Minnesota.
8.. Sorghum
Six questionnaires were sent in the sorghum survey; three
replies were received, two of which from respondents electing not.to
participate. The comments from the one respondent indicated that resis-
tant sorghum varieties will be employed for greenbug control.
9. Sugarbeets
Four of the eleven experts surveyed responded. Three of the
responses were related to insect control. The one expert in disease
control indicated that growers would be interested in more effective
control of rhizoctonia crown rot.
300
-------
Table B-8
PEANUT DISEASES IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Acres
(thousands)
Estimated
Control
Cost
(dollars per
reated acre)
Disease/
1974
1980
1985
t
Organism
Stated
Infested
Treated
Infested
Treated
Infested
Treated
1974
1980
1985
Damping-off
Alabama
75
10
100
75
120
8
$ 4.00
$ 5.00 $
i 5.00
Leaf spot
Alabama
Texas
North Carolina
201
285
168
190
225
168
250
285
n. a.
245
250
n.a.
300
290
n. a.
300
250
n.a.
30.00
3.50
10.00
40.00
4.00
n. a.
50.00
5.00
15.00
Nematodes
Alabama
Texas
North Carolina
10
200
168
5
40
65
25
225
n.a.
20
50
n.a.
40
225
n. a.
35
50
n. a.
10.00
12.00
10.00
10.00
20.00
n. a.
15.00
25.00
12.00
Peanut rust
Texas
45
20
45
30
45
30
3. 50
4.00
5.00
Pod rot
Alabama
Texas
North Carolina
25
140
168
10
45
25
30
190
n.a.
28
75
n.a.
40
195
n.a.
38
75
n. a.
10.00
12.00
5.00
10.00
20.00
n. a.
12.00
25.00
3.00
Root-knot
nematode
Alabama
Texas
60
60
40
20
100
80
90
40
120
85
110
40
10.00
12.00
10.00
20.00
15.00
25.00
Root rot
Alabama
1
0
20
18
30
20
n. a.
25.00
30.00
Sclerotium
rolfsii
Alabama
150
10
200
180
220
200
20.00
30.00
30.00
Seedling
disease
Alabama
201
0
250
100
300
150
n.a.
15.00
15.00
Seed rot
Alabama
201
200
250
250
300
150
2.00
2.00
2.00
Southern
blight
North Carolina
168
40
n.a.
n.a.
n. a.
n.a.
5.00
n.a.
6.00
Web blotch
Alabama
Texas
0
260
0
160
80
260
60
200
200
260
180
200
30.00
3.50
40.00
4.00
50.00
5.00
n.a. * not available.
Number of respondents = 8.
Data for only the states for vhich completed questionnaires were obtained.
-------
Table B-9
PEANUT INSECT PESTS IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Estimated
Control
Acres Cost
(thousands) (dollars per
1974 1980 1985 treated acre)
Pest
State*
Infested
Treated
Infested
Treated
Infested
Treated
1974
1980
1985
Burrowing bug
Texas
30
28
35
35
60
50
$ 6.25
$ 6.00
$ 7.50
Granulate
Alabama
100
50
125
70
150
80
4.00
4.00
4.00
cutworm
Heliothis spp.
Texas
35
60
150
100
170
140
2.25
3.50
5.00
Lesser corn-
Alabama
80
20
110
30
130
50
4.00
4.00
4.00
stalk borer
Texas
170
80
170
120
200
180
5.50
6.50
8.00
Potato leaf
Virginia
80
80
80
80
80
80
10.00
11.00
12.00
hopper
North Carolina
168
168
n. a.
n. a.
n. a.
n. a.
1.50
n.a.
n. a.
Southern corn
Virginia
75
75
75
75
75
75
10.00
11.00
12.00
rootworm
North Carolina
40
40
n. a.
n. a.
n.a.
n.a.
6.00
n. a.
n. a.
Thrips
Alabama
190
100
240
150
285
200
4.00
4.00
4.00
North Carolina
168
168
n. a.
n. a.
n.a.
n. a.
1.50
n. a.
1.50
Virginia
100
100
100
100
100
100
10.00
11.00
12.00
Two-spotted
North Carolina
75
75
n. a.
n. a.
n.a.
n.a.
8.00
n. a.
n. a.
spider mite
n.a. = not available.
•k
Number of respondents = 8.
^Data for only the states for which completed questionnaires were obtained.
-------
All three insect experts Indicated that sugarbeet root maggot
required treatment on a total of 186 thousand of 221 thousand infested
acres covered by responses in 1974. Two experts anticipated treatment
on 85 thousand of 125 thousand acres, and 95 thousand of 130 thousand
acres, by 1980 and 1985 respectively.
Cutworms were also indicated as a control problem. Two of
the three respondents felt that Class A and B products would provide
future control of insect problems. In addition, one expert specifically
mentioned the use of growth inhibitors (Class C) for aphid control and
resistant sugarbeet varieties (Class H) for toot maggot control, each
at an anticipated level of 5 percent of total control by 1985. Plant
resistant varieties were anticipated as providing 90 percent control of
leafhoppers by 1980.
10. Tomatoes
Three replies from six of the experts included in the survey
were received; however, the data provided were incomplete. The respon-
dents anticipated that future control of the tomato disease complex
including early blight, septoria leaf blight, gray leafspot, fruit
anthracnose, and leaf mold would be provided by Class A and B products,
particularly systemics on irrigated acreage. They also felt that, if
shown to be viable, microbial control could gain a large percentage of
the market by 1985.
11. Cole Crops
All four of the experts included in the survey replied, with
one disease expert electing not to participate. Only 1974 data were
provided by the other disease expert who indicated that mildew and botry-
tis were the most important disease organisms.
Data on insect control were also incomplete; however, one
expert indicated that control of cabbage looper, diamonback moth, and
cabbage root aphid was required on most infested acres, and that improved
B.t. and polyhedral virus strains might provide control up to 40 percent
by 1980.
12. Potatoes
A total of six replies, two from disease experts and four from
insect experts, were received from the fourteen experts included in the
survey. The disease control data were incomplete; one expert antici-
pated future control would be provided by Class A and Class B products,
specifically systemics.
303
-------
All four Insect respondents indicated that 100 percent of
insect control in 1974 was provided by Class A products. Respondents
anticipated the following methods to control all of the identified
insect pests by 1980 and 1985:
(1) Class A (registered "second generation" pesticides):
50 to 100 percent by 1980 and 20 to 100 percent by
1985.
(2) Class B (new chemical products of the same general
type as in Class A): 10 to 30 percent by 1980 and
10 to 80 percent by 1985, with use for control of
potato leaf hopper and the potato flea beetle reach-
ing 80 percent by 1985.
(3) Class C (novel new chemical products): 10 to 20
percent by 1980 and 40 to 50 percent by 1985, with
juvenile hormone analog control of aphids.
(4) Class D (microbials): 5 to 20 percent by 1980 and
40 to 50 percent by 1985, with polyhedrosis virus
control of lepidopterous larvae.
(5) Class E (pheromones): 5 to 10 percent by 1980 and 1985
with baited traps possible for use against wireworms
and two-spotted mites.
Specific comments concerning the use of particular types of
control products included:
(1) Pyrethroid derivatives could represent up to 50 per-
cent of the total potato insect control market by 1985.
(2) Green peach aphid resistance will stimulate the devel-
opment of juvenile hormone analogs for control.
(3) Polyhedrosis virus has been proven effective for
control of lepidopterous larvae and will be used,
provided the possibility of mutation to toxic strains
is eliminated.
Selected data on anticipated disease and insect pest problems
in three states appear in Tables B-10 and B-ll.
13. Apples
Five of the nine experts replied, with two respondents electing
not to participate in the survey. Responding disease control experts
indicated that some antagonistic fungi or other microbials may provide
control for apple scab and powdery mildew by 1985, and that the ideal
control would be a systemic fungicide requiring one application for
season-long control.
304
-------
Table B-10
Disease/
Organism
Black-leg
bacteria
Deep-pitted
scab
Early blight
Early blight
foliage lesion
Early blight
tuber lesion
State1
Washington
Washington
Wisconsin
Washington
Washington
Powdery mildew Washington
Washington
Ring-rot
bacteria
Seed piece
decay
VerticiIlium
wilt
Wisconsin
POTATO DISEASES IDENTIFIED BY SURVEY RESPONDENTS
(1974, 1980, 1985)
1974
Ac res
(thousands)
1980
1985
Infested Treated Infested Treated Infested Treated
n. a.
50
70
30
10
n. a.
25
Washington 55
n. a.
50
50
5
n. a.
25
n. a.
n. a.
55
80
40
n. a.
n. a.
30
n. a.
n. a.
n. a.
55
40
10
n. a.
n.a.
30
n. a.
n. a.
n.a.
58
100
n. a.
n.a.
n.a.
35
n.a.
n. a.
n. a.
58
40
n.a.
n.a.
35
n.a.
Estimated
Control
Cost
(dollars per
treated acre)
1974
n. a.
tu a.
1980 1985
n. a.
n, a.
n.a. n.a.
$ 50.00 $70.00 $80.00
30.00 40.00 40.00
tl. a.
25.00
n. a.
1.50
100.00
n.a. n.a.
n.a.
n.a.
n. a.
n. a.
n.a. n.a.
n. a.
n.a. = not available.
•k
Number of respondents = 6.
^Data for only the states for which completed questionnaires were obtained.
-------
Table B-ll
U1
O
tr.
POTATO INSECT PESTS IDENTIFIED BY SURVEY RESPONDENTS
(1974, 1980, 1985)
AcTes
(thousands)
Estimated
Control
Cost
(dollars per
1974
1980
1985
Pest
Stated
Infested
Treated
Infested
Treated
Infested
Treated
1974
1980
1985
Aphids
Minnesota
5°
25
25
15
25
trace
$ 3.00
$ 5.00
n.a.
Colorado potato
Washington
45
20
55
30
75
50
15.00
17.00
$20.00
beetle
Minnesota
5
5
5
5
10
10
3.00
5.00
5.00
Wisconsin
1
1
2
2
2
2
14.00
14.00
15.00
Cutworms
Minnesota
25
15
20
20
20
20
3.00
4.00
5.00
Wisconsin
5
5
6
6
7
7
7.50
8.50
10.00
European corn
Wisconsin
3
3
5
5
6
6
7.00
8.00
10.00
borer
Flea beetles,
Wisconsin
50
50
55
55
58
58
15.00
18.00
20.00
leaf hopper,
potato aphid
Green peach aphid
Washington
85
75
100
90
120
100
50.00
60.00
70.00
(leaf roll virus
Wisconsin
50
7
55
10
58
10
12.00
14.00
15.00
vector)
Leaf hoppers
Minnesota
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n. a.
n. a.
n.a.
Two-spotted mites
Washington
30
25
40
30
50
50
10.00
12.00
15.00
Wire worms
Washington
40
30
20
20
20
20
3.00
4.00
5.00
Minnesota
Uirevorms,
Wisconsin
12
12
13
13
14
14
7.00
23.00
25.00
white grubs
n. a. = not available.
*
Number of respondents ¦ 6.
Data only for states for which completed questionnaires were obtained.
-------
Data on insect control provided by one expert indicated an
expected decrease in the use of Class A products to 10 percent of the
total average annual cost of control per treated acre by 1985.' Class B
products would account for 40 to 60 percent control, especially against
the codling moth, and Class C products (specifically juvenile hormone
analogs) would provide 20 to 50 percent control (mostly against apple
aphids). Class F products (controlled release of natural predators and
parasites) were estimated as accounting for 60 percent of the average
cost of McDaniel mite control per acre by 1985. The controlled release
of sterile insects was considered a definite future control possibility,
currently limited by its cost.
Selected data on anticipated disease and insect pest problems
in two states appear in Tables B-12 and B-13.
14. Citrus
Two of the four experts included in the citrus survey responded,
and provided information specific to California. Disease control of
three of the seven principal disease pests involved the use of conven-
tional (Class A) products in 1974 and respondents anticipated future
control to rely on their use. They expected that either no treatment
or the removal of infested trees would continue as control methods for
the other four disease pests.
Respondents indicated that thrips, mites, red scale, and leaf
hoppers required treatment with Class A products on 690 thousand of the
810 thousand acres treated in 1974; 1980 and 1985 estimates of treat-
ment were 893 out of 1,051 and 280 out of 410 thousand treated acres,
respectively. The reduction in acreage requiring either Class A or
Class B product treatment by 1985 was seen resulting from the use of
parasites and predators for up to 50 percent of scale and mite control.
Selected data on anticipated disease and insect pest problems
in California appear in Table B-14 and B-15.
15. Grapes
Neither of the two experts contacted responded.
16. Survey Questionnaire
A copy of the survey questionnaire used to collect the data
in this Appendix is presented in Figure B-l following p. 311.
307
-------
Table B-12
APPIE DISEASES IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Estimated
Control
Acres Cost
(thousands) (dollars per
1974 1980 1985 treated acre)
Disease/Organism State^ Infested Treated Infested Treated Infested Treated 1974 1980 1985
Apple scab Oregon 5 2 5 2 5 2 $9.00 $9.90 $11.00
Phytophora Washington 20 20 n. a. n. a. n.a. n.a. 20.00 n.a. n.a.
Powdery mildew Oregon 5 2 5 2 5 2 9.00 9.90 11.00
Washington 60 40 n.a. n.a. n.a. n.a. 15.00 n.a. n.a.
n.a. = not available.
*
Number of respondents = 3.
^Data for only the states for which completed questionnaires were obtained.
-------
Table B-13
Pest
APPLE INSECT PESTS IDENTIFIED BY SURVEY RESPONDENTS*
(1974, 1980, 1985)
Stated
1974
Acres
(thousands)
1980
1985
Estimated
Control
Cost
(dollars per
treated acre)
Infested Treated Infested Treated Infested Treated 1974
1980
1985
Apple aphids
Washington
93.5
50
60
50
60
50
$20.00
$30.00
$ 50.00
Cat facing bugs
Washington
20
20
n. a.
n. a.
n. a.
n. a.
n.a.
n. a.
n. a.
Codling moth
Washington
93.5
93.5
100
100
110
110
40.00
60.00
100.00
European red mite
Washington
60
60
60
60
60
60
23.30
40.00
60.00
McDaniel mite
Washington
93.5
45
50
25
60
20
22.27
30.00
50.00
San Jose scale
Washington
75
60
80
70
90
80
25.00
30.00
45.00
n. a. « not available.
*
Number of respondents = 3.
^Data for only the states for which completed questionnaires were received.
-------
Table B-14
CITRUS DISEASES IDENTIFIED BY SURVEY RESPONDENTS
(1974, 1980, 1985)
Acres
Disease/Organism
State*
1974
1980
1985
Infested Treated
Infested Treated
Infested Treated
Foot rot
California
2
1
2
- 1
2
1
Navel end rot
California
3
n. a.
4
n. a.
4
n. a.
Oak root fungus
California
< 1
n. a.
n. a.
n. a.
n. a.
n. a.
Penicillium species
California
300
n. a.
320
n. a.
330
n. a.
Psorosis
California
1
n. a.
1
n. a .
n. a.
n. a.
Quick decline
California
4
n. a.
2
n. a.
2
n. a.
Septoria spot
California
1
n. a.
n. a.
n. a.
n. a.
n. a.
Estimated
Control
Cost
(dollars per
treated acre)
1974 1980 1985
n. a.
n. a.
n. a.
n. a.
n. a.
n. a.
n. a.
n. a.
n. a.
n. a.
n. a .
n. a.
n. a.
n. a.
n. a.
n. a.
n. a,
n, a.
n. a.
n. a.
n. a.
n.a. = not available.
•k
Number of respondents = 2.
Data for only the states for which completed questionnaires were received.
-------
Table B-15
CITRUS INSECT PESTS IDENTIFIED BY SURVEY RESPONDENTS
(1974, 1980, 1985)
1974
Acres
(thousands)
1980
1985
Estimated
Control
Cost
(dollars per
treated acre)
Pest
Stated
Infested
Treated
Infested
Treated
Infested
Treated
1974
1980
198-
Citricolar scale
California
200
10
208
18
100
10
$25
$33
$35
Citrus cutworms
California
90
80
100
95
100
80
23
30
35
Leafhoppers
California
200
150
300
225
100
75
25
30
35
Leaf roller
California
40
30
50
45
50
40
15
20
25
Mites
California
100
90
200
180
20
15
20
33
35
Red scale
California
190
180
190
188
100
90
87
110
130
Thrips
California
300
270
330
300
150
100
50
65
75
*
Number of respondents = 2.
^Data from only the states for which completed questionnaires were obtained.
-------
April 1975
U>
I—'
IO
: TRENDS IN INSECT AND DISEASE CONTROL
Crop
1. According to U.S D.A. data, the acreage planted to this crop(s) in 1974 was as indicated below. Please provide your best estimates of the number of
acres likely to be planted to this crop in 1980 and 1985.
1974 1980 1985
Planted Acreage (thousands of acres): Your state
Total United States
Average Farm Value (dollars per harvested acre): Your state
Total United States
(Remarks on trends in acreage)
FIGURE B-1 SAMPLE QUESTIONNAIRE—PESTICIDE USE PARAMETERS
-------
Crop
2. Please list below the principal insect and disease pests for this crop(s) IN YOUR STATE for which control was necessary in 1974
and for which control will be available and purchased in 1980 and 1985. While admittedly difficult to do so, try to "guesstimate"
to the best of your ability the probable levels of infestation, treatment, and control cost per treated acre for these pests.
Principal Intact Pests
1974
1980 | 1985
Infested
Acres
(thousands)
Treated
Acres
(thousands)
Average Cost of
Control*
(dollars per treated
acre in 1974 dollars)
Infested
Acres
(thousands)
Treated
Acres
(thousands)
Average Cost of I , ,
« . fi Infested
Control St
,, 8 Acres
(dollars per treated 9 . ^ , v
• -in . (thousands)
acre in 19/4 dollars) X
T reated
Acres
(thousands)
Average Cost of
Control*
(dollars per treated
acre in 1974 dollars)
|
|
|
|
I
|
|
J
I
|
8
I
|
I
TOTAL
1
Principal Disease Pert*
1
B
|
|
I
I
I
I
J
|
|
I
|
J
I
|
I
J
I
I
TOTAL
I
*We recognize that specific product applications may control more than one pest, thus making it difficult to identify average per-acre control costs for each pest. Please fill in
this portion of this question in a manner most representative of the existing and anticipated control practices in your state.
FIGURE B-1 SAMPLE QUESTIONNAIRE—PESTICIDE USE PARAMETERS (Continued)
-------
3. What practical problems do you foresee over the next ten years in your state in the control of these pests on
this crop(s), with regard to resistance, availability of control products, costs, environmental problems, and
other aspects of effective and economic pest management?
4, Do you think that the per-acre control costs for these pests in 1980 and 1985 will increase or decrease as a
percentage of total per-acre crop production costs? If so, by how much? Please explain.
FIGURE B-1 SAMPLE QUESTIONNAIRE—PESTICIDE USE PARAMETERS (Continued)
-------
5. The following is a listing of classes of products that are currently being used, or have the potential of being used in integrated or other
management systems, for the control of the listed pests.
Class A. Registered "second generation" pesticides, e.g., chlorinated
hydrocarbons, org* no phosphates, dithiocarba mates, etc.
Class 6. New chemical products of the same general type as in Class A.
Class C. Novel new chemical products, e.g., juvenile hormones and
analogues, chrtin inhibitors, ovicides, ecdysones, etc.
Please indicate on a percentage control cost basis the role that you feel these classes of products play and are likely to play in the
control of the indicated pests IN YOUR STATE ON THIS CROP(Sl.
Class 0. Microbials, e.g., virvses, bacteria.
Class E. Pheromones.
Class F. Controlled release of natural parasites and predators.
Class G. Controlled release of sterile insects.
Class H. Other
Principal Insect Pests
LEVEL OF CONTROL BY CLASS OF PRODUCTS
(express as a % of the average cost of control per treated acre)
1974
1980
1985
A
B
C
D
E
F
G
H
! a
B
C
D
E
F
G
H
A
B
C
D
E
F
G
H
1
TOTAL
1
Principal Disease Pejts
|
J
J
J
1
I
J
I
|
|
|
I
|
TOTAL
1
. 1
U>
I—
FIGURE B-1 SAMPLE QUESTIONNAIRE—PESTICIDE USE PARAMETERS (Continued)
-------
6. If you indicated that some level of future control is likely to be effected on this crop(s) with products in classes B through H, please list the specific
products you think will be used, and on what pests, as outlined below.
Will Be Used to Control:
% Market Share in
Spaeific Q*a B Through H Products You Think Will 8* Used Type of Pest(s) 1980 1985
W
»-»
Ov
FIGURE B-1 SAMPLE QUESTIONNAIRE—PESTICIOE USE PARAMETERS (Continued)
-------
7. The following are among the critical factors influencing the commercial availability and use of pesticide products and pest management techniques.
Efficacy Cost of control Ease of application
Specificity Short-term human safety Grower's attitude
Mode of action Long-term human and Deregistration of
Pest resistance environmental safety existing pesticides
For the specific products in classes B through H you listed opposite as obtaining a certain level of pest control market share in your state by 1980
and/or 1985, please list and discuss the factors which you think will be most responsible for their reaching this position.
Sptific CliM B Through H Product* You Listed in Question 6. Factors, Discussion _
CO
h-»
FIGURE B-1 SAMPLE QUESTIONNAIRE—PESTICIDE USE PARAMETERS (Continued)
-------
8. If you have any other comments or remarks on the trends in insect and disease pest management for this crop{s) — in your state or nationally — over the
next ten years, please use the space below.
9. When you have completed this survey, please return to: Thomas A. Blue
Manager, Agricultural Chemicals
Chemical industries Center
Stanford Research Institute
Menio Park, California 94025
Your assistance is deeply appreciated. When the final survey results are published, we will be glad to send you a complimentary copy. Please indicate
below where it should be sent: *
'The contents of individual surveys will not be divulged but we do need to know who responded in case we have additional questions and in order to distribute survey results.
FIGURE B-1 SAMPLE QUESTIONNAIRE—PESTICIDE USE PARAMETERS (Concluded)
------- |