(An International Journal for the Americas)
Volume 75, No. 4 December, 1992
TABLE OF CONTENTS
WORKSHOP ON IMPORTANT ARTHROPOD PESTS OF THE
CARIBBEAN BASIN AMENABLE TO BIOLOGICAL CONTROL:
HOMOPTERA, COLEOPTERA, LEPIDOPTERA
McCoY, C. W .- Introduction .................................................................... 399
CRUZ, C., AND A. SEGARRA-Potential for Biological Control of Crop Pests in
the C aribbean ................................................................................ 400
ROSEN, D., AND P. DEBACH-Foreign Exploration: The Key to Classical Biolog-
ical C control ................................................................................... 409
BUCKINGHAM, G. R.-Role of Quarantine Facilities in Biological Control ...... 414
FURGESON, J. S.-Biological Control: An Industrial Perspective ................ 421
ANDREWS, K. L., J. W. BENTLEY, AND R. D. CAVE-Enhancing Biological
Control's Contributions to Integrated Pest Management Through Appropri-
ate Levels of Farmer Participation ................................................. 429
BROWNING, H. W.-Overview of Biological Control of Homopterous Pests in the
C aribbean ...................................................................................... 440
GERLING, D.-Approaches to the Biological Control of Whiteflies .................. 446
OSBORNE, L. S., AND Z. LANDA-Biological Control of Whiteflies with En-
tom opathogenic Fungi ..................................................................... 456
OBRYCKI, J. J.-Techniques for Evaluation of Predators of Homoptera ........... 472
BAKER, P. S., A. KHAN, A. I. MOHYUDDIN, AND J. K. WAAGE-Overview of
Biological Control of Lepidoptera in the Caribbean ........................... 477
GELENTER, W. D.-Application of Biotechnology for Improvement of Bacillus
thuringiensis Based Products and Their Use for Control of Lepidopteran
Pests in the Caribbean ............................................ ...................... 484
ALAM, M. M.-Biological Control of Insect Pests of Crucifers in Selected West
Indian Islands .................................... ........................................ 493
BELLOTTI, A. C., B. ARIAS V., AND 0. L. GUZMAN-Biological Control of the
Cassava Hornworm Erinnyis ello (Lepidoptera: Sphingidae) ................ 506
IGNOFFO, C. M.-Environmental Factors Affecting Persistence of Entomo-
pathogens ..................................................................................... 516
CAPINERA, J. L., AND N. D. EPSKEY-Potential for Biological Control of Soil
Insects in the Caribbean Basin Using Entomopathogenic Nematodes ...... 525
STOREY, G. K., AND C. W. McCoy-Potential for Biological Control of Soil
Insects Using Microbial Pesticides in the Caribbean ........................... 533
BARBERCHEK, M. E.-Effect of Soil Physical Factors on Biological Control
Agents of Soil Insect Pests ............................................................. 539
SIRJUSINGH, C., A. KERMARREC, H. MAULEON, C. LAVIS AND J. ETIENNE-
Biological Control of Weevils and Whitegrubs on Bananas and Sugarcane
in the Caribbean ......................................................................... 548
Continued on Back Cover
Published by The Florida Entomological Society
FLORIDA ENTOMOLOGICAL SOCIETY
OFFICERS FOR 1992-93
President ............................. .............. .. ................... D. F. W illiams
President-Elect .................... .. ... .. ...................... J. E. Pefia
Vice-President ................... ............................ ............. ....... E. M Thoms
Secretary ........................ ... ................. ... D. G. Hall
Treasurer .......................................................... ..................... A. C. Knapp
Other Members of the Executive Committee
J. L. Knapp D. P. Wojcik L. A. Wood J. Hogsette
J. R. McLaughlin O. Liburd D. R. Suiter
J. R. McLaughlin, USDA/ARS, Gainesville, FL ...................................... Editor
Agricultural, Extension, & Regulatory Entomology
James R. Brown-Disease Vector Ecology & Control Center, NAS, Jacksonville, FL
Richard K. Jansson-Tropical Research & Education Center, Homestead, FL
Michael G. Waldvogel-North Carolina State University, Raleigh, NC
Stephen B. Bambara-North Carolina State University, Raleigh, NC
Biological Control & Pathology
Ronald M. Weseloh-Connecticut Agricultural Experiment Sta., New Haven, CT
J. Howard Frank-University of Florida, Gainesville
Chemical Ecology, Physiology, Biochemistry
Louis B. Bjostad-Colorado State University, Fort Collins, CO
Ecology & Behavior
Theodore E. Burk-Creighton University, Omaha, NE
John H. Brower-Stored Product Insects Research Laboratory, Savannah, GA
Forum & Symposia
Carl S. Barfield-University of Florida, Gainesville
Genetics & Molecular Biology
Sudhir K. Narang-Bioscience Research Laboratory, Fargo, ND
Medical & Veterinary Entomology
Arshad Ali-Central Florida Research & Education Center, Sanford, FL
J. E. Pefa-Tropical Research & Education Center, Homestead, FL
Systematics, Morphology, and Evolution
Michael D. Hubbard-Florida A&M University, Tallahassee
Gary J. Steck-Florida State Collection of Arthropods, Gainesville
Willis W. Wirth-Florida State Collection of Arthropods
Business M manager ........................... ......... .... ................. A. C. Knapp
FLORIDA ENTOMOLOGIST is issued quarterly-March, June, September, and De-
cember. Subscription price to non-members is $30 per year in advance, $7.50 per copy;
institutional rate is $30 per year. Membership in the Florida Entomological Society,
including subscription to Florida Entomologist, is $25 per year for regular membership
and $10 per year for students.
Inquiries regarding membership and subscriptions should be addressed to the Busi-
ness Manager, P. O. Box 7326, Winter Haven, FL 33883-7326.
Florida Entomologist is entered as second class matter at the Post Office in DeLeon
Springs and in Winter Haven, FL.
Manuscripts from all areas of the discipline of entomology are accepted for consider-
ation. At least one author must be a member of the Florida Entomological Society.
Please consult "Instructions to Authors" on the inside back cover.
This issue mailed January 15, 1993
Biological Control Workshop-'91: Introduction
WORKSHOP ON IMPORTANT ARTHROPOD PESTS OF THE
CARIBBEAN BASIN AMENABLE TO BIOLOGICAL CONTROL:
HOMOPTERA, COLEOPTERA, LEPIDOPTERA
CLAYTON W. McCoY, CONVENER
This compilation of papers is the result of a workshop held April 8-10, 1991 in
Orlando, Florida. The workshop was organized at the request of, and funded by, the
Caribbean Basin Administrative Group (CBAG). The CBAG encourages cooperative
research among scientists in the Caribbean region and supports research in a wide
range of agricultural disciplines and commodities at the University of Florida, Puerto
Rico, and the Virgin Islands.
Agriculture in most Caribbean areas is characterized by small multiple crop systems
plagued by a diverse complex of pests. Many of these pests are exotic with few natural
enemies. When placed in a tropical environment with a year-round growing season,
many pests have flourished in all agricultural regions. Pesticide use has been widespread
for a number of years now; however, the development of resistance in some major pests
and an increased public concern over the hazards posed to human health and the envi-
ronment has spawned an urgent cry for effective and sound alternative strategies utiliz-
ing biological controls.
The Organizing Committee for the workshop consisted of Clayton W. McCoy and
Fred D. Bennett, co-chairmen, University of Florida; Carlos Cruz, University of Puerto
Rico; Josef Keularts, University of the Virgin Islands; Harold W. Browning and Jorge
E. Pefa, University of Florida; and Dean F. Davis, CBAG program manager. They
recognized a need for action and organized this workshop to provide a scientific forum
for the presentation of current research on the biological control of important homopter-
an, coleopteran, and lepidopteran pests of the Caribbean Region. The committee chose
these three pest groups on the basis of their overall importance and amenability to
biological control. The purpose of the workshop was an follows: (1) assess opportunity
for the application of natural enemies to important agricultural pests within these insect
groups, (2) identify scientists working on these key pests and foster opportunity for
cooperation among scientists, (3) identify opportunity for immediate application of
biological control technologies, and (4) identify knowledge gaps that will require more
The members of the Organizing Committee feel the workshop achieved the purpose
of identifying research needs and fostered information exchange among scientists active
in the Caribbean.
Copies of the symposium may be obtained from:
CBAG Program Manager
University of Florida, IFAS
1022 McCarty Hall
Gainesville, FL 32611
400 Florida Entomologist 75(4) December, 1992
POTENTIAL FOR BIOLOGICAL CONTROL OF CROP PESTS
IN THE CARIBBEAN
CARLOS CRUZ AND ALEJANDRO SEGARRA
Crop Protection Department
College of Agricultural Sciences
University of Puerto Rico
Mayagfiez, Puerto Rico 00708
The potential for biological control of important insect pests of crops of economic
importance in the Caribbean is evaluated. Crops include sugarcane, coffee, bananas,
plantains, sweet potatoes, yams, cassava, tomato, peppers, cabbage, curcubits, some
fruits, pigeon pea and beans. Suggestions are made for biological control of major insect
pests such as Diatraea saccharalis, Diaprepes abbreviatus, Cosmopolites sordidus,
Cylas formicarius, Hypothenemus hampei, Leucoptera coffeella, Plutella xylostella,
Heliothis zea and Diaphania spp. Recommendations are made for the proper evaluation
and selection of candidate pests.
Se informa aqui sobre la evaluacion del potential de control biologico de insects
plagas de los principles cultivos del Caribe. Los cultivos considerados importantes
incluyen cafla, cafe, platanos, guineos, batata, fame, yuca, tomate, pimiento, repollo,
cucurbitaceas, algunas frutas, gandul y habichuelas. Se hacen sugerencias sobre el con-
trol biologico de algunos insects importantes tales como Diatraea saccharalis, Dia-
prepes abbreviatus, Cosmopolites sordidus, Cylasformicarius, Hypothenemus hampei,
Leucoptera coffeella, Heliothis zea y Diaphania spp. Se hacen recomendaciones para
evaluar las posibles plagas y establecer prioridades para el control biologico en el Caribe.
Biocontrol of plant pests in the Caribbean was almost completely neglected with the
advent of pesticides which appeared to be the solution to all pest problems. However,
due to the development of resistance in some major pests, as well as to public awareness
of hazards posed to human health and the environment searches are being made for
seemingly old alternatives.
The tropical environment in the Caribbean facilitates a year round growing season
and supports continuous generations of both pests and beneficial organisms. Agriculture
is characterized by small subsistence farms, multiple crops and pests, pesticide problems
and other problems common to production and marketing.
Biocontrol is now being acknowledged as a major area for further research and is a
component of a USDA initiative in sustainable agriculture. The ARS, the Entomology
Department Administrators (1988) in U.S. Universities and the Experiment Stations
have proposed working groups to prepare national biocontrol research programs. This
has evolved as an effective alternative to the pesticide dilemma.
Biocontrol is one of the most thoroughly studied methods against insects, nematodes,
diseases and weeds. Scientists are now increasing efforts and emphasis on pest control
methods not aimed at wholesale eradication, but rather at maintaining pest populations
below economic thresholds.
This has already started in the Caribbean, where efforts of national and international
institutions have begun to emphasize and implement biocontrol programs. These efforts
Biological Control Workshop-'91: Cruz & Segarra
are directed at identification of research priorities and possible areas for cooperation
within the region. At present, there are several institutions conducting individual re-
search with little or no regional coordination. Biocontrol efforts and findings remain
little known even between neighboring countries. This is due in large measure to in-
adequate communication between research and development organizations in the re-
gion. There are limited means to disseminate and exchange findings (scientific congres-
ses, journals, meetings, seminars, etc.). Lack of funds for research and for attending
local or international meetings is by far the greatest constraint to the development of
biocontrol in the region. A workshop should serve to identify research needs and to
promote interchange of knowledge within the Caribbean community.
There is evidence of the effective biocontrol successes worldwide and in the Carib-
bean (Cock 1985). Bennet (1990) and Alam et al. (1990) summarized the most important
cases of successful biocontrol in the Caribbean. The recent introduction of the citrus
blackfly Aleurocanthus woglumi in Puerto Rico and its outstanding effective control
with the subsequent introduction of the parasitoids Encarsia opulenta and Amitus
hesperidum are a pertinent example (Medina et al. 1991). Chemical control of this pest
has been impossible in most places; only biocontrol has been effective.
In Puerto Rico, as in many other Caribbean countries, biocontrol has been used for
decades. Table 1 presents information regarding the introduction of parasites and pre-
dators into Puerto Rico in the 1900's. Most introductions were made during the 1930's,
up to 1940, just before the pesticide era. As often occurs, data on many of these intro-
ductions are not available since relevant studies were never conducted.
TABLE 1. PARASITES AND PREDATORS OF INTRODUCED PESTS IN THE 1900'S.
Species Natural Enemy Level
and A. pustulans
402 Florida Entomologist 75(4)
TABLE 1. (CONTINUED)
Species Natural Enemy Level
Tetrastichus girff ,.ii,li,,n .. .
Some exotic natural enemies have been recently introduced for the control of certain
insect pests, such as the citrus black fly, the lima bean pod-borer, and the sugarcane
borer and others such as nematodes, diseases and weeds (Cruz & Segarra 1990). Also,
several species of entomophilic nematodes for the control of root weevils were intro-
duced and tested.
Some recent biocontrol attempts have been successful in Caribbean countries, i.e.
the sugarcane borer, the citrus blackfly, the diamondback moth (to a certain level).
On-going projects are devoted to control of the coffee berry borer, Hipothenemus ham-
pei (Baker 1990). Two potential parasitoid candidates have been identified: Prorops
Biological Control Workshop-'91: Cruz & Segarra
nasuta, and Cephalonomia stephanoderes. These parasitoids will be reared and in-
creased for distribution in the areas where the pest has been identified: Colombia,
Jamaica, Mexico, etc. Very positive results are expected from this initiative.
IMPORTANT CROPS AND INSECT PESTS IN THE CARIBBEAN
Several lists of important pests in the Caribbean have been compiled elsewhere.
However, a partial lists of crops and pests which are considered important in the Carib-
bean are presented in Table 2 and potential biocontrol agents are presented in Table 3.
These are based mostly on experience and the literature. Among the most important
crops grown in the Caribbean are coffee, sugarcane, bananas, plantains, root crops
(sweet potato, yams, taniers, cassava), tomato, peppers, cabbage, cucurbits, citrus,
pineapple, avocado, mango, pigeon pea and beans. There are important insect pests
attacking these crops, some of them with a high potential for biocontrol (Table 3).
The Sugarcane Borer
The sugarcane borer, Diatraea saccharalis, is under effective biocontrol in some
countries, but it remains an important limiting factor in several others. Different natural
agents have been successful on some countries, probably indicating the influence of
varying habitats. This pest offers a good opportunity for classical biocontrol. Several
natural enemies have been reported by Bennet (1990) and others. Currently, Cotesia
flavipes, Lixophaga diatraeae, Metagonistylum minense and Trichogramma spp. are
TABLE 2. SOME IMPORTANT CROPS AND THEIR MAIN INSECT PESTS IN THE CARIB-
Crops Main Insect Pests
Sugarcane Diatraea, Diaprepes, Phyllophaga
Coffee Hypothenemus hampei,
Banana and plantains Cosmopolites sordidus
Sweet potato Cylasformicarius, Euscepes
Cassava Silva spp. (Stem shoot fly), mites
Citrus Aleurocanthus woglumi,
Pineapple Batrachedra comosae (gomosis)
Beans Leafhoppers, whiteflies, pod borers
Pigeonpea Pod borers, leafhoppers
Tomato Heliothis, Liriomyza, Keiferia
Pepper Anthonomus eugenii, Myzus
persicae, Thrips palmi, mites, etc.
Cabbage Plutella xylostella
Cucurbits Diaphania spp.
Florida Entomologist 75(4)
TABLE 3. SOME IMPORTANT INSECT PESTS AND THEIR POTENTIAL FOR BIOLOGI-
CAL CONTROL IN THE CARIBBEAN.
Root weevils: Diaprepes abbreviatus,
The coffee berry borer Hypothenemus
Whiteflies: Aleurocanthus woglumi,
Lepidopterous pests: Diatraea spp.
Others: Liriomyza spp.
Ants, Entomopathogenic fungi,
Prorops nasuta, Sephanoderes
Encarsia spp., Amitus sp.,
spp., Microbial control
Cotesia plutellae, Diadegma
insularis, Diadegma sp.
Mirax insularis augmentation
Others Conservation, foreign
Several: Conservation and
Many: Conservation and
Many: Conservation and
Some: Foreign exploration,
conservation and augmentation
the most successful biocontrol agents. Research is needed to continue the improvement
of the existing biocontrol agents.
Historically, the sugarcane rootstalk weevil, Diaprepes abbreviatus, has been among
the most difficult pest to control in the Caribbean. One contributing factor is the high
reproductive potential (5,000 eggs/female) in a situation where very low egg suvivorship
is enough for it to be a pest (Armstrong 1987). Several biological control attempts have
failed except for entomopathogenic nematodes. However, the search for other biocontrol
agents continues. Castro (1986) studied the ants species feeding on neonate larvae. She
found seven feeding on the larvae. Of these, Pheidole subarmata borinquensis and
Pheidolefallax demonstrated the greatest efficiency. In Cuba, Castineiras et al. (1990a,
1990b), reported P. megacephala controlling Cosmopolites sordidus and Cylas for-
micarius. Richman et al. (1983) reported the ants Monomorium floricola and Cre-
matogaster ashmeadi preying on the egg masses of D. abbreviatus in Puerto Rico and
Florida, respectively. Col6n (1986) found 17 species of fungi associated with the larvae
of D. abbreviatus. Pathogenicity tests demonstrated that half of them were able to kill
the larvae. Gliocadium sp. and Fusarium sp. caused the highest mortality. Armstrong
(1981, 1987) studied the efficiency of the egg parasitoid Tetrastichus haitiensis on D.
Biological Control Workshop-'91: Cruz & Segarra
abbreviatus eggs oviposited on sugarcane. He found a low efficiency of the parasitoid
and concluded it was probably due to the difficulty of oviposition through the sugarcane
leaves as compared to other hosts.
Several species of entomopathogenic nematodes have been tested for the control of the
most important root weevils in Puerto Rico. Roman & Figueroa (1985) reported effec-
tive control of the larvae of D. abbreviatus with Steinernema feltiae (=Neoaplectana
carpocapsae). Figueroa & Roman (1990) reported tests with S. glaseri, S. bibioni and
Heterorhabditis heliothidis against D. abbreviatus. Gonzalez (1986) observed some con-
trol using H. bacteriophora and H. heliothidis. Figueroa (1990) obtained excellent re-
sults with S. feltiae and S. bibioni against C. sordidus. A graduate student is currently
evaluating S. feltiae and H. heliothidis against C. formicarius. Several local strains or
species of entomopathogenic nematodes recently found in Puerto Rico are being evalu-
ated. Natural enemies against other curculionid pests such as Exophthalmus spp. and
Pachnaeus spp. are being sought in some Caribbean countries.
The Coffee Berry Borer
The coffee berry borer, H. hampei, is threatening coffee production in the Carib-
bean. Biocontrol research has produced encouraging finding. Besides the two
parasitoids mentioned in the introduction, other potentially effective parasitoids are
under study. An example is Heterospilus coffeicola, which has been classified as effec-
tive but difficult to rear in the laboratory. Other potential candidates include Phymas-
ticus coffee, Aphanognus dictynnu (apparently a hyperparasite of Prorops nasuta),
and some microbial agents such as Beauveria bassiana, Metarhizium anisopliae and
Paecilomyces spp. These are under study at various research centers worldwide
The Coffee Leafminer
Another pest of coffee with a high potential for biological control is the coffee leaf-
miner, Leucoptera coffeella. Gallardo (1988) and Wolcott (1947) have reported 15
parasitoids from Puerto Rico. Gallardo has proposed a biocontrol program utilizing the
augmentation method using the parasitoid Mirax insularis. This idea has extraordinary
merit and should be pursued vigorously. There are many other natural enemies reported
that could be considered when their efficacy and mass rearing techniques are deter-
mined. Foreign exploration is another potential alternative which has not been attemp-
ted for this pest, which supposedly originated in Reunion (Green 1984).
The Diamondback Moth
The diamondback moth (DBM) Plutella xylostella is a lepidopteran pest of impor-
tance to the Caribbean with a significant number of natural enemies. It has outstanding
potential for a successful biocontrol program. In Puerto Rico several parasitoids are
particularly effective. Diadegma insularis may have rates of parasitism up to 90% when
protected from pesticides. One limitation is the very low economic injury level of the
pest. However, the selection of planting dates, use of selective pesticides and the intro-
duction (where not available) of the most effective parasitoids will undoubtedly improve
the level of biocontrol.
Alam (1986) reported complete biocontrol with Trichogramma spp. and Cotesia spp.
Bennett & Yasseen (1972) reported effectiveness of C. plutellae against DBM in Bar-
bados, Montserrat, St. Vincent and Trinidad.
Florida Entomologist 75(4)
The Corn Earworm
The corn earworm (CEW) Heliothis zea is the limiting factor for sweet corn and
other vegetable crops. Chemical control is very difficult, even with an effective insec-
ticide. However, many natural enemies have been observed attacking the eggs and
larval stages of this insect (King & Coleman 1989). The wasp Trichogramma sp. and
the predator Orius pumilio are considered by Figueroa (1983) to be the most important
natural enemies in Puerto Rico. In the US, CEW infestations are considered more
severe in dry than in wet summers because moist conditions favor the development of
several fungal and bacterial diseases. Also several species of predators and parasitic
wasps attack it. Populations of these natural enemies can significantly reduce or almost
eliminate CEW infestations early in the season. Therefore, early season sprays should
be delayed in order to allow populations of natural enemies to develop. Frequent sam-
pling and control measures are necessary in vegetables because even minimal damage
by this and other pests can render produce unmarketable.
Usually the melonworm Diaphania hyalinata (L.) is the most important insect pest
of cucurbits in the Caribbean. Medina et al. (1989) observed several parasitoids and
predators in Puerto Rico. No natural enemies have been studied in detail. In Barbados,
Alam (1986) reported melonworm eggs parasitized by Trichogramma pretiosum, the
larvae by Eiphosoma dentator and by Cotesia sp., and the pupae by Brachymeria sp.
There is a great potential for the biocontrol of this pest, particularly with microbial
insecticides, which are already being used in some countries like Guatemala (Agricola
El Sol, 30 Calle 11-42, Zona 12, Guatemala).
Scale Insects, Whiteflies, Aphids and Thrips
The potential for biocontrol of small insects is well known. Since the introduction of
Rodolia cardinalis in California in 1892 for control of the cottony cushion scale or the
biocontrol of the citrus blackfly in the Caribbean, many outstanding examples have
occurred. In Florida, a biocontrol project is seeking natural enemies for whiteflies. As
a result, several natural enemies have been recorded from Florida, California, the
Caribbean, and Central and South America.
SOME CRITERIA FOR BIOCONTROL TARGET SELECTION
Because there are many biocontrol success stories and the many introduced pests
in the Caribbean, primary emphasis (or first step) should be given to classical biocontrol.
The main task is to identify biocontrol agents (Fry 1989) and introduce them where the
target species are present. There are many known effective biocontrol agents that still
have not been established in countries where target pests are causing crop losses. It is
only recently that many species of important pests have been introduced into Puerto
Rico and several other islands; e.g., the citrus blackfly, A. woglumi; the pepper weevil,
Anthonomus eugenii; the beet armyworm, Spodoptera exigua; the tomato pinworm,
Keiferia lycopersicella; the imported fire ant, Solenopsis invicta; Thrips palmi; the
sugarcane thrips, Fulmekiola serrata; and the avocado lace wing bug, Pseudacysta
persea. Most recently an apparently new race of the whitefly, Bemicia tabaci has in-
vaded the Caribbean, being more destructive and with a wider range of plant hosts than
the common whitefly.
Biological Control Workshop-'91: Cruz & Segarra 407
Designating priority to pests should be an important goal of local and regional efforts
to establish biocontrol programs. Pests, needs, and goals will vary among Caribbean
countries. Therefore, a system to evaluate candidate pests should be developed based
primarily on three components: 1) an assessment of current or potential impact of the
pest to the regional economy; 2) a thorough appraisal of institutional policy and logistic
capabilities; and 3) an evaluation of prior biocontrol achievements against the pest.
When applied within the context of local conditions, these criteria should insure that
the initial phases of a biocontrol initiative in the Caribbean will be as risk free as
Pests should be targeted according to crop importance, impact on the crop, and
finally a cost and benefit analysis. Those pests which attack a major commodity with
highly predictable temporary consistency, where current management methods are no
longer effective, and which are not under quarantine regulations are excellent candi-
dates. Similarly, pests with high economic injury levels, that are non-vectors, and which
are sole 'key' pests ought to be given preference as targets. Finally, if funds are scarce,
priority should be given to pests with natural control agents which have proven effective
elsewhere in contrast to projects requiring more expensive operations like mass rearing
or foreign exploration.
The depth of biocontrol achievements against a pest should be the final component
of its assessment as a target. Pests where natural enemy taxonomy is well known as
well as those where the study of natural mortality is available should be considered
first. Targets where successful programs exist abroad should be favored initially.
ALAM, M. M. 1986. Vegetable pests and their natural enemies in Barbados, West
Indies, Proc. 22nd Caribbean Food Crops Society. 25-29 August, 1986. St. Lucia.
ALAM, M. M., J. C. REID, AND G. MULLER. 1990. The present status and future
needs of biological control in the Caribbean community. Caribbean meetings on
Biological Control, 5-7 November 1990, Guadeloupe, F.W.I.
ARMSTRONG, A. 1981. Distribuci6n de Diaprepes abbreviatus L. (Coleoptera:Cur-
culionidae) y Tetrastichus haitiensis Gahan (Hymenoptera:Eulophidae) en las
areas cameras del norte y noroeste de Puerto Rico. M.S. Thesis, RUM-UPR. 83
ARMSTRONG, A. 1987. Parasitism of Tetrastichus haitiensis Gahan on egg masses of
Diaprepes abbreviatus in Puerto Rico. J. Agric. Univ. P.R. 71(4): 407-409.
BAKER, P. 1990. Biological control of the coffee berry borer. CARAPHIN News, No.
2, IICA, Trinidad and Tobago.
BENNETT, F. D. 1990. An overview of Classical Biological Control in the Caribbean
and some examples of the utilization of entomophagous insects. Caribbean Meet-
ings on Biological Control, 5-7 November 1990, Guadeloupe.
BENNETT, F. D. AND, M. YASSEEN. 1972. Parasite introduction for the biological
control of three pests in the Lesser Antilles and British Honduras. PANS 18:
CASTINEIRAS, A., A. BORGES, AND 0. OBREGON. 1990a. Biological control of Cylas
formicarius elegantulus (Summ.). Caribbean Meetings on Biological Control, 5-7
November 1990, Guadeloupe.
CASTINEIRAS, A., T. CABRERA, A. CALDERON, M. LOPEZ, AND M. LUJAN. 1990b.
Lucha biol6gica contra Cosmopolites sordidus, Caribbean Meetings on Biological
Control, 5-7 November 1990, Guadeloupe, F.W.I.
CASTRO, S. 1986. Hormigas depredadoras de larvas neonatas de Diaprepes ab-
breviatus (Coleoptera:Curculionidae). M.S. Thesis, RUM-U.P.R. 72 pp.
COCK, M. J. W. 1985. Review of Biological Control of Pest in the Commonwealth
Caribbean and Bermuda up to 1982. Commonwealth Inst. Biol. Cont. Tech.
Comm. #9. 244 pp.
Florida Entomologist 75(4)
COLON, I. 1986. Estudio preliminary de generous de hongos aislados de larvas de Diap-
repes abbreviatus en el area oeste de Puerto Rico y su evaluaci6n como control
biol6gico. M.S. Thesis, R.U.M.-U.P.R. 30 pp.
COUNCIL OF ENTOMOLOGY DEPARTMENT ADMINISTRATORS. 1988. Research Initia-
tives, A Research Agenda for Entomology, Prepared by the Subcommittee on
Research Initiatives, T. Don Canerday, Chairman, Univ. of Georgia, Ga.
CRUZ, C., AND A. SEGARRA. 1990. Recent biological control experiences in Puerto
Rico. Caribbean Meetings on Biological Control, 5-7 November 1990,
FIGUEROA, W. 1990. Biocontrol of the banana root borer weevil, Cosmopolites sor-
didus (Germar), with steinernematid nematodes. J. Agric. Univ. P.R. 74(1): 15-
FIGUEROA, W., AND J. ROMAN. 1990. Parasitism of entomophilic nematodes on the
sugarcane rootstalk borer, Diaprepes abbreviatus L. (Coleoptera:Curculionidae),
larvae. J. Agric. Univ. P.R. 74(2): 197-202.
FIGUEROA, E. 1983. Ciclo vital y enemigos naturales de Heliothis zea en maiz. M.S.
Thesis, R.U.M.-U.P.R. 110 pp.
FRY, J. M. 1987. Natural enemy data bank. CAB International Institute of Biological
Control, United Kingdom, 185 pp.
GALLARDO, F. 1988. Faunal survey of the coffee leafminer (Leucoptera coffeella)
parasitoids in Puerto Rico. J. Agric. Univ. P.R. 72(2): 255-62.
GONZALEZ, N. 1986. Evaluaci6n de dos nematodos entom6fagos Heterorhabditis
heliothidis y H. bacteriophora (Rhabditoidea:Heterorhabditidae) en el control de
la larva de Diaprepes abbreviatus (L.) (Coleoptera:Curculionidae). M.S. Thesis,
R.U.M.-U.P.R. 36 pp.
GREEN, S. D. 1984. A proposed origin of the coffee leafminer Leucoptera coffeella
(Gu6rin-M6neville) (Lepidoptera:Lyonetidae). Bull. Entomol. Soc. Am. 30: 30-
KING, E. G., AND R. J. COLEMAN. 1989. Potential for biological control of Heliothis
species. Annu. Rev. Entomol. 34: 53-75.
MEDINA, S., E. ABREU, F. GALLARDO, AND R. FRANQUI. 1989. Natural enemies of
the melonworm, Diaphania hyalinata L. (Lepidoptera:Pyralidae), in Puerto
Rico. J. Agric. Univ. P.R. 73(4): 313-320.
MEDINA, S., A. SEGARRA, AND R. FRANQUI. (in press). La mosca negra ed los cit-
ricos, Aleurocanthus woglumi Ashby (Homoptera:Aleyrodidae) en Puerto Rico.
J. Agric. Univ. P.R.
RICHMAN, D. B., W. F. BUREN, AND W. H. WHITCOMB. 1983. Predatory arthropods
attacking the eggs of Diaprepes abbreviatus (L.) (Coleoptera:Curculionidae) in
Puerto Rico and Florida. J. Georgia Entomol. Soc. 18: 335-342.
ROMAN, J., AND W. FIGUEROA. 1985. Control of the larva of the sugarcane rootstalk
borer, Diaprepes abbreviatus (L.) with the entomogenous nematode Neoaplec-
tana carpocapsae Weiser. J. Agric. Univ. P.R. 69(2): 153-158.
WOLCOTT, G. 1947. A quintessence of sensitivity: The coffee leaf miner. J. Agric.
Univ. P.R. 31: 215-219.
Biological Control Workshop-'91: Rosen & DeBach
THE KEY TO CLASSICAL BIOLOGICAL CONTROL
DAVID ROSEN' AND PAUL DEBACH2
'Entomology and Nematology Department, University of Florida,
Gainesville, FL 32611-0740; permanent address: The Hebrew University,
Faculty of Agriculture, Rehovot 76100, Israel
2Department of Entomology, University of California, Riverside, CA 92521
Importation and establishment of exotic natural enemies, known as "classical" biolog-
ical control, is the best alternative to the wholesale use of toxic chemical pesticides. The
statistics of worldwide importation projects are discussed, and a plea is made for en-
hanced foreign exploration for natural enemies.
La importacion y el establecimiento de enemigos naturales exoticos, conocido como
control biologico clasico, es la mejor alternative a el uso unilateral de pesticides quimicos
toxicos. Se discuten las estadisticas de los projects de importacion de enemigos
naturales a nivel mundial y se hace una peticion para el fomento de exploraciones en el
With rising concern about environmental quality, modern agriculture has been ex-
periencing a severe pesticide crisis. While effective pest control is an absolute necessity
in most cropping systems, it has become increasingly evident that unilateral reliance
on toxic chemicals may involve serious hazards of chemical pollution affecting man,
domestic animals, plants and wildlife. Resurgences and upsets of target and non-target
pests, triggered by rapid development of resistance coupled with destruction of benefi-
cial natural enemy populations, have become common occurrences, and the ever-mount-
ing cost of chemical pesticides has often become prohibitive. Safer, environmentally
compatible and less expensive alternatives to wholesale chemical control are therefore
Of the various alternatives available to us-selective chemicals, cultural, mechanical,
physical and autocidal controls, etc.-"classical" biological control, or the importation
of exotic natural enemies and their establishment in new habitats, has been by far the
most successful to date. Other selective control methods may serve as important compo-
nents of integrated pest management whenever they are applicable, but classical biolog-
ical control remains the most promising approach for the foreseeable future. When it
works, as it often does, biological control offers effective, permanent, inexpensive and
virtually hazard-free solutions to serious pest problems. Worldwide, attempts at class-
ical biological control have been made against 416 species of insect pests belonging to
eight orders, and a significant degree of success has been achieved against two-fifths
of them: 75 species have been brought under complete control by imported natural
enemies, 74 under substantial control, and 15 under partial control in at least one
country. Many of these successes have been repeated in other countries, so that al-
together there have been at least 384 successful projects worldwide, including 156 that
have been rated as complete successes, 164 as substantial, and 64 as partial successes
(see DeBach & Rosen 1991). Achievements with biological control of weeds have been
410 Florida Entomologist 75(4) December, 1992
as impressive, but this falls beyond the scope of this conference. Outstanding examples
in the Caribbean Basin have included the citrus blackfly, Aleurocanthus woglumi Ashby
(Homoptera: Aleyrodidae), and the sugar-cane borer, Diatraea saccharalis (Fabricius)
(Lepidoptera: Pyralidae), and-in Florida-the Florida red scale, Chrysomphalus
aonidum (L.), the purple scale, Lepidosaphes beckii (Newman), and the citrus snow
scale, Unaspis citri (Comstock) (Homoptera: Diaspididae). Several projects are cur-
rently addressing other pests.
The benefit-to-cost ratio of classical biological control has been calculated as 30:1
(DeBach & Rosen 1991). Compared with a 3:1 ratio for chemical control (Pimentel et
al. 1981), this is indeed remarkable. The cost of researching, developing and marketing
a new pesticide is now at least $20 million, and the process may take at least ten years.
The cost of discovering a new natural enemy, on the other hand, rarely exceeds several
thousand dollars. Repeat projects, where a successful natural enemy is simply transfer-
red to another country, may cost no more than the small amount required for a few
airmail shipments; but even new projects, involving foreign exploration "from scratch",
may be relatively inexpensive. Unfortunately, unlike augmentative biological control,
where natural enemies can be mass-reared and sold to consumers for periodic releases,
there is no direct profit incentive for private enterprise to import and establish exotic
natural enemies. Such incentives can, however, be provided by government or agro-in-
Some have cited a low rate of establishment of imported natural enemies as justifi-
cation for the claim that classical biological control hardly ever works. The rate of
establishment is actually very good: Out of a total of 4226 natural enemy importations
that have been recorded to date in 153 countries, about 30 percent have resulted in
establishment, and the fate of an additional 22 percent is still unknown (DeBach &
Rosen 1991). This is an impressive record, considering that, by comparison, only one
in 15,000 or 20,000 candidate chemicals ever makes it to the marketing stage as a
pesticide (DeBach & Rosen 1991). However, the rate of establishment in itself is, of
course, irrelevant as a measure of success of biological control. If nine natural enemies
fail to become established and the tenth provides complete, permanent control of a
serious pest, it is the benefit-to-cost ratio of the entire project that should be taken as
the criterion for its success, and we have already seen that this ratio is excellent.
In view of this outstanding record, it is almost unbelievable that out of about 10,000
insect pest species recorded worldwide, only some 400 species have ever served as
targets for classical biological control efforts. In other words, over 95 percent of the
world's insect pests have had no biological control importation project directed against
them. This is indeed a remarkable indictment against our profession! (DeBach & Rosen
Exciting breakthroughs in molecular biology should not overshadow the fact that
importation of exotic natural enemies is still the best method for controlling a pest. The
only alternative, that of creating new superior natural enemies through genetic en-
gineering, is still far from offering practical solutions (Beckendorf & Hoy 1985), and
until it does, foreign exploration for exotic natural enemies should always receive the
first priority. Thorough ecological study of agro-ecosystems can be most enlightening
and improve our understanding of pest-natural enemy interactions. However, with all
our simulation, modeling and sophisticated analysis capabilities, classical biological con-
trol remains essentially an empirical endeavor: We simply cannot predict the perform-
ance of a natural enemy in a new habitat. On the other hand, competition among natural
enemy species does not present a threat to the success of a biological control project.
Multiple importations may result either in complementary action of several natural
enemies coexisting in the same habitat, or in competitive displacement of inferior species
by one that is better adapted to that particular habitat, but in each case the outcome
Biological Control Workshop-'91: Rosen & DeBach
is more effective biological control. There is, therefore, no substitute for foreign explo-
ration. Importation of as many exotic natural enemies as can be discovered is the key
to success in pest management.
Inasmuch as foreign exploration and importation of natural enemies are basically
empirical, there is no need for extensive preliminary research. Knowledge of systema-
tics-both of the target pest and of its natural enemies-and basic biology is an essential
prerequisite. In order to be utilized, a natural enemy obviously has to be recognized as
distinct, and inadvertent importation of hyperparasites or other noxious organisms
should of course be avoided. However, basic ecological research, important as it may
be for our understanding of ecosystems, should not delay the actual importation of
natural enemies. Time spent in studying existing host-natural enemy complexes, etc.,
before new natural enemies are imported, can result in real loss to society. The loss an
investigator can be credited with may be calculated as the time spent from the start of
a project until the actual importation is made, times the annual savings accrued by
ultimate success of biological control (DeBach 1966). One could cite quite a few good
examples for such unnecessary delays.
The Caribbean Basin has seen one of the very first attempts to use air transportation
for the transfer of natural enemies. It was Dr. J. G. Myers who, as early as 1932, used
fledgling PanAm flights to ship sugar-cane borer parasites from Cuba to Antigua. With
modern air transport, there is really no excuse for not bringing the natural enemies in.
In the past, the transfer of live organisms from one continent to another required
prolonged odysseys on land and sea, and explorers had to maintain suitable host plants
with live cultures of host insects, or make judicial use of whatever refrigerating facilities
were available on board ships at the time, for their colonies of natural enemies to
survive the long journey. Many importation failures resulted from these formidable
difficulties, as well as from poor handling techniques (see van den Bosch 1968, Hoy
1985). Modern-day explorers may still encounter tremendous difficulties and hazards,
especially in some so-called developing countries-Dr. Myers' exploits in the Amazon
Basin stand out as an excellent example, and conditions have not improved much over
there-but the actual transfer of live natural enemies from one corner of the globe to
another can usually be accomplished within 24 to 48 h, and improved techniques such
as the use of humiditrons (DeBach & Rose 1985) make the task even easier. The rate
of successful establishment of imported natural enemies is therefore constantly improv-
ing, and explorers can devote most of their energy to the actual search.
Classical biological control is still far from reaching the point of diminishing returns.
Not only have the vast majority of pest species never served as targets for importation
projects, but the vast majority of natural enemy species still remain to be discovered.
The fauna of large world areas is very little known, and it has been estimated that
between 70 and 90 percent of all parasitic Hymenoptera-by far the most important
group of biological control agents-are still undescribed, and that we have no biological
information whatsoever about 97 percent of all existing species (DeBach & Rosen 1991).
Most of the task is still ahead of us, and at the current rate of foreign exploration it is
anybody's guess how many potentially powerful weapons will become extinct before we
even have a chance to discover them!
The title of this conference-"Important Arthropod Pests of the Caribbean Basin
Amenable to Biological Control: Homoptera, Coleoptera, Lepidoptera"-is somewhat
misleading. It may be misunderstood to imply that other groups of pests are either not
important, or not amenable to biological control. If there is one lesson that may be
learned from an analysis of past projects, it is that classical biological control is broadly
applicable to essentially all pest situations. Successes have indeed been achieved against
Homoptera, Coleoptera and Lepidoptera, but also against various other arthropod
pests, as well as against diverse organisms such as snails and rabbits, weeds and plant
Florida Entomologist 75(4)
pathogens. In spite of the opinions of some, that biological control may work only under
certain favorable conditions, it has been very successful under very diverse cir-
cumstances: in continental areas as well as on islands, in temperate as well as tropical
climates, on annual as well as perennial crops and in forests, and against direct as well
as indirect pests, including species of medical and veterinary importance. In fact, al-
though the prospects may seem better in certain situations than in others, as a general
rule the amount of success in biological control has been directly correlated with the
amount of research and importation work carried out. Further efforts are likely to yield
many further successes (Rosen 1985, DeBach & Rosen 1991).
Both exotic and native pests have proved amenable to classical biological control.
The most common scenario is, of course, that of an exotic pest invading a new area,
leaving its natural enemies behind. Foreign exploration in the land of origin is, there-
fore, a logical starting point for any classical biological control project. However, pests
may acquire additional natural enemies as they spread into new areas, and natural
enemies may also become accidentally established in faraway lands, or may have differ-
ent strains in different areas. Some climatic matching can be very useful: If the target
pest is a problem on a tropical island in the Caribbean, it stands to reason that explora-
tion for its natural enemies should focus on similar areas. However, it should be remem-
bered that natural enemies have been known to surprise us in the past in this respect,
and there is no way of predicting their performance in a new habitat. Whenever possi-
ble, exploration throughout the range of distribution of the target pest should be carried
out, and a broad genetic spectrum of each natural enemy species should be imported
from numerous localities.
Some outstanding successes have been achieved by importing the natural enemies
of related species, so when all else fails this option should not be discounted. However,
the idea should not be carried to extremes: The so-called "new associations theory"
(Hokkanen & Pimentel 1984), which claims that natural enemies that have not evolved
with the target pest are somehow more likely to effect biological control than do "old
associations", is simply not true and is not borne out by the statistics of worldwide
importation projects. Thus, although the parasites and predators of related species
should certainly not be ignored, those that have evolved with the target pest in its land
of origin should continue to be the first choice (DeBach & Rosen 1991).
The main problem with foreign exploration is that, by definition, it has to be per-
formed in foreign lands, and administrators are often reluctant to provide funds for trips
abroad. Although only modest funds are sometimes required, some funds are neverthe-
less always needed. As early as in the cottony-cushion scale project in California, at-
tempts at foreign exploration were at first thwarted by an act of Congress prohibiting
employees of the USDA from traveling abroad, and it took some adroit political maneuv-
ering to send Albert Koebele to attend an exposition in Melbourne and bring about
what, to paraphrase a contemporary expression, has become "the mother of all biological
control projects". Unfortunately, official attitudes have not changed much during the
century that has elapsed since then. This, coupled with the rise of rather petty environ-
mental concerns about possible detrimental effects of biological control and the tendency
of bureaucrats to over-regulate importation activities, has resulted in a regrettable
decline of classical biological control. These adverse trends ought to be reversed.
In many cases, natural enemies can be obtained from knowledgeable colleagues in
other countries, without resorting to foreign travel. International organizations may
also be of great help-the IIBC, or International Institute of Biological Control, for-
merly known as the CIBC, is currently the best one, maintaining stations on several
continents with trained personnel collecting and shipping natural enemies upon request.
However, in the final analysis, there is no substitute for dedicated foreign explorers
from the importing country, who are willing to go out there by themselves and search
Biological Control Workshop-'91: Rosen & DeBach 413
for natural enemies to solve a pest problem back home. There is a great shortage of
such trained explorers, and the attitude of many of our universities and research institu-
tions, which prefer so-called basic, paper-oriented scientific research to foreign explora-
tion for biological control agents, is not very helpful. If these trends toward over-scien-
tification and bureaucratization continue unhindered, they may eventually bring about
the demise of classical biological control. It is in the better interests of all mankind that
foreign exploration for exotic natural enemies be encouraged and adequately funded,
and that dedicated foreign explorers be rewarded for their contributions to problem-sol-
BECKENDORF, S. K., AND M. A. HOY. 1985. Genetic improvement of arthropod nat-
ural enemies through selection, hybridization or genetic engineering techniques,
pp. 167-187 in M. A. Hoy and D. C. Herzog [eds.], Biological control in agricul-
tural IPM systems. Academic Press, Orlando.
DEBACH, P. 1966. Theory, theorems and practice of biological control. Paper pre-
sented at the Annual Meeting of the Entomological Society of America, Portland,
Oregon, 6 pp.
DEBACH, P., AND M. ROSE. 1985. Humidity control during shipment and rearing of
parasitic Hymenoptera. Chalcid Forum 4: 11-13.
DEBACH, P., AND D. ROSEN. 1991. Biological control by natural enemies, 2nd ed.
Cambridge University Press, Cambridge, xv + 440 pp.
HOKKANEN, H., AND D. PIMENTEL. 1984. New approach for selecting biological con-
trol agents. Canadian Entomol. 116: 1109-1121.
HoY, M. A. 1985. Improving establishment of arthropod natural enemies, pp. 151-166,
in M. A. Hoy and D. C. Herzog [eds.], Biological control in agricultural IPM
systems. Academic Press, Orlando.
PIMENTEL, D., J. KRUMMEL, D. GALLAHAN, J. HOUGH, A. MERRILL, I.
SCHREINER, P. VITTUM, F. KOZIOL, E. BACK, D. YEN, AND S. FIANCE.
1981. A cost-benefit analysis of pesticide use in U.S. food production, pp. 27-54
in D. Pimentel [ed.], CRC handbook of pest management in agriculture. Vol. 2.
CRC Press, Boca Raton, FL.
ROSEN, D. 1985. Biological control, Chapter 13, pp. 413-464 in G. A. Kerkut and L.
I. Gilbert [eds.], Comprehensive insect physiology, biochemistry and pharmacol-
ogy. Vol. 12, Insect Control. Pergamon Press, Oxford.
VAN DEN BOSCH, R. 1968. Comments on population dynamics of exotic insects. Bull.
Entomol. Soc. America 14: 112-115.
Florida Entomologist 75(4)
ROLE OF QUARANTINE FACILITIES IN
GARY R. BUCKINGHAM
Agricultural Research Service
U.S. Department of Agriculture
c/o Florida Biological Control Laboratory
P.O. Box 147100, Gainesville, FL 32614-7100
Quarantine facilities play an important role in classical biological control programs.
Imported natural enemies can be identified, cleaned of diseases and hyperparasites, and
studied in a secure quarantine facility without risk to the environment. Personnel at
the facilities advise other researchers about regulations and techniques for importation,
document results of the importations through shipment record forms, annual reports,
and voucher specimens, and conduct research on biologies and host ranges of natural
enemies. Not only does the quarantine facility help allay the fears of the public about
the dangers of importing foreign organisms, it also protects the researcher from blame
for accidental introductions not associated with the researcher's project.
Las unidades cuarentenarias juegan un papel decisive en los programs de control
biologico clasico. Los enemigos naturales importados pueden ser identificados, asi como
tambien sus enfermedades o hiperparasitos pueden ser estudiadas en las unidades de
cuarentena sin ningun riezgo a el medio ambiente. El personal engargado de estas
unidades, aconseja a los investigadores acerca de los reglamentos y tecnicas de importa-
cion, y document los resultados de las importaciones, a traves de registros de envio,
reports anuales, especimenes tipo, e investiga la biologia y el rango de hospederos de
los enemigos naturales. La unidad cuarentenaria ayuda no solamente a aliviar el temor
public en lo referente a las consecuencias de la importacion de organismos foraneos,
sino que tambien proteje al investigator de culpabilidad por la introduction accidental
de organismos no asociados con el project de investigation.
Classical biological control programs involve shipments of natural enemies between
countries usually with insect and plant hosts included. Often the hosts are present in
the receiving country, but not always. Sometimes natural enemies are collected from
insects or plants closely related to the pest rather than from the pest itself. In these
cases the host of the natural enemy might become a new pest if it were inadvertently
released. In addition, the host material often includes natural enemies of the desired
To house these natural enemies and their hosts upon arrival, a network of quarantine
facilities has been organized. These facilities are not part of the traditional quarantine
inspection system that prohibits entry of unwanted guests into the country or into a
state. They are operated by or for researchers not by or for inspectors. This distinction
has occasionally been overlooked by foreign agricultural officials surprised and disap-
pointed that they were not visiting an inspection facility when they visited our quaran-
tine facility at the Florida Biological Control Laboratory (FBCL), Division of Plant
Industry, Florida Department of Agriculture and Consumer Services, in Gainesville.
Approximately 15 facilities that process insect shipments are present throughout the
Biological Control Workshop-'91: Buckingham 415
United States. Some of the larger or longest established ones in addition to the FBCL
are in Riverside and Albany, California; Newark, Delaware; Stoneville, Mississippi;
Columbia, Missouri; Bozeman, Montana; and College Station and Temple, Texas.
ROLE OF THE QUARANTINE FACILITY
The quarantine facility plays an important role in the "biological control drama".
Like an effective actor, it dominates the second act of a three-act improvisational play.
During the first act, it whispers advice to other actors from the wings. Bursting upon
stage in the second act as a sword-slashing swash-buckler, it protects the stage from
all intrusions. With the trained memory of a seasoned actor, it documents the impro-
vised movements and dialogue from the second act, saving them for future playwrights
and actors. Finally, during every performance it researches new movements, gestures,
and dialogue to improve itself and to ensure success of the final act.
Of course, biological control is not an improvisational play that can be manipulated
as the actors and director wish, but it does have many similarities. For example, it
plays to both empty and full houses; it has limited engagements and record-breaking
long runs; it has road tours; and its existence is often dependent on the financial support
of a "producer" and the interest of the ticket-buying public. Like a classic Shakespeare
play, classical biological control provides generations of followers with a performance
that changes subtly with each new theater group and new director.
Departing from the drama analogy, I believe the principal functions of a quarantine
facility and its personnel can be characterized as follows:
1. Advise foreign explorers and other researchers about regulations governing im-
portation of natural enemies and about shipping techniques.
2. Protect the environment from unwanted introductions of pests and of natural
enemies of biocontrol agents.
3. Document introductions of biocontrol agents into and out of quarantine for the
benefit of researchers, administrators, and the historical record.
4. Conduct basic research, for example, studies of biologies of natural enemies, of
host ranges of natural enemies especially weed control agents, and of exotic insects for
projects not associated with biological control.
In this paper I discuss these functions to aid those readers who wish to construct a
quarantine facility, to modify a laboratory room for quarantine use, or to import natural
enemies through an existing facility. If classical biological control is to develop fully in
the Caribbean area, quarantine facilities will undoubtedly be needed. Further informa-
tion about quarantine procedures can be found in the paper by Rose & Harrison (1990).
As interest in classical biological control grows, more and more researchers who are
inexperienced in foreign exploration or in importation of agents need advice before
beginning a project. The quarantine facility provides a highly visible and ready source
of information about regulations, permits, shipping labels and techniques, etc. This will
be especially true in a country or state with a new quarantine facility and new emphasis
on natural enemy importations.
In the United States, introduction of insects and pathogens for control of agricultural
pests is regulated by the Animal and Plant Health Inspection Service (APHIS) of the
United States Department of Agriculture (USDA) and by the Departments of Agricul-
ture of the individual states. The researcher completes a form (Form PPQ 526) request-
ing permission to introduce the natural enemy. This form must be submitted to state
officials for approval before it is forwarded to APHIS. It includes the names of the pest
Florida Entomologist 75(4)
and its natural enemies, the source of the material to be introduced, the names of the
researchers, and information about the project and future disposition of the natural
enemies. If the request is approved, APHIS sends distinctive shipping labels to the
researcher along with a signed copy of the form. These labels are placed on the outside
of the package by the foreign explorer or collaborator to alert customs and agricultural
inspectors at the port of entry. I usually recommend that researchers also place a copy
of the permit in an envelope addressed to the plant quarantine inspector and attached
to the outside of the package. National regulations concerning introductions of biocon-
trol agents are currently being examined for possible changes. It behooves all countries
to have a stream-lined permitting process that protects against unsafe introductions yet
facilitates movement of packages through ports of entry.
In addition to information about regulations, quarantine personnel are often asked
about shipping techniques and clearance procedures at ports of entry. Whenever possi-
ble international shipments that cannot be hand carried should be sent by air freight or
by courier service rather than by international air-mail, unless the postal service's
Express Mail International service is available and has been previously tested. A ship-
ment of alligatorweed flea beetles (Agasicles hygrophila Selman & Vogt) that I sent
from Florida to Beijing, People's Republic of China, was received by the scientist within
3 days using Express Mail International service. This was faster than two preceding
attempts to have the flea beetles hand carried to Beijing. Within the United States,
various express delivery services and the postal service's Express Mail system offer
convenient, dependable overnight service to most locations.
Shipping containers come in all shapes and sizes. The container with the natural
enemies should be enclosed in a cloth bag and packaged inside another container to
prevent escape in case the outside package is damaged. Information about shipping
containers can be obtained from Bartlett & van den Bosch (1964), Boldt & Drea (1980)
and Hendricksen et al. (1987).
Researchers wishing to import specific natural enemies should make sure that
quarantine personnel have authoritatively identified specimens ahead of time for com-
parison in quarantine with the living material. This greatly accelerates the release from
quarantine and can increase the chance of successful colonization if the biology of the
identified natural enemy has been previously reported and can be consulted immediately
by quarantine personnel. This provision of specimens to quarantine personnel ahead of
time can not be over-emphasized. Natural enemy identification is extremely difficult
without specimens for comparison.
PROTECT THE ENVIRONMENT
Quarantine facilities are foremost a filter system used to prevent plant pests and
natural enemies of biocontrol agents from entering the country along with the flow of
biocontrol agents. They protect not only the environment, but also the researcher whose
work is viewed as a threat by many people. It is not unusual to discover new immigrant
insects during a project because of the increased sampling in a crop or habitat. If the
project were not conducted in quarantine, the researcher might be showered with blame
for the unwanted "introduction", thus jeopardizing the project and possibly future pro-
jects. This protection is overlooked by some researchers who utilize the quarantine
facility merely because regulations require it.
Facilities differ greatly in size and complexity, however they generally share several
safety features. Doors remain locked and admittance is restricted to authorized person-
nel. Some facilities have recently installed push button locks with codes that can be
changed monthly or when projects finish. There is at least one anteroom and often two.
Biological Control Workshop-'91: Buckingham
In the FBCL facility, the first anteroom is illuminated continuously and the second
which opens into quarantine has no light. White laboratory coats are donned in the
darkened anteroom. Light is provided by a small window between the rooms which is
also a baffled insect trap to catch insects that might have been carried on the laboratory
coats. Negative air pressure inside the facility helps prevent insects on or near the door
from being sucked outside when the door is opened. Floors, walls, ceilings, and windows
in quarantine facilities must be carefully sealed to prevent insect escape. Air intakes
and exhausts are either covered with fine mesh metal screens or the air-handling system
is recycled through large, very efficient filters as it is at the FBCL. All waste material
leaving quarantine is autoclaved or sterilized in another manner. Many facilities have
tanks for treatment of waste-water. Greenhouses must have wire-reinforced glass or
very strong tempered glass. Because the greenhouses are tightly sealed a cooling sys-
tem is essential. When the air-conditioning system disfunctions at the FBCL, we run
a lawn sprinkler on top of the greenhouses to cool them below lethal temperatures.
Some organizations may need and could support large quarantine facilities like those
at Stoneville, Mississippi (313 m2, two 25 m2 greenhouses) and at the FBCL (173 m2,
four 18 m2 greenhouses). However, countries or organizations with lesser needs and
smaller budgets can have safe, efficient quarantine rooms like those at the University
of California, Albany, California (60 m2) and at the Center for Biological Control in
Central America (CBCCA), El Zamorano, Honduras (32 m2) (R. Cave, personal com-
munication). With care, an existing laboratory room could be renovated as a quarantine
room. Detailed descriptions of the preceding quarantine facilities are reported in Leppla
& Ashley (1978). Organizations wishing to build large facilities are cautioned to consider
carefully the large maintenance and energy costs that these facilities require and to
provide for these costs. Insect colonies can not be moved to another building in case of
equipment failure and if they die out they can not be replaced readily as can native or
adventive insects used in most research programs.
Equally as important as the facility for protecting the environment and perhaps
more important are the personnel who work in the facility. Each facility has a quarantine
officer who is responsible for enforcing rules and ensuring maintenance. At smaller
facilities the quarantine officer often handles most of the material, but at large research
facilities this is usually not possible. The quarantine officer must ensure that the identity
of a known biocontrol agent is confirmed either by a comparison with specimens or by
a specialist. This is true even with pure colonies from other research laboratories. A
shipment of laboratory parasitized fly puparia sent to the FBCL from an overseas
research program yielded a pure colony of parasites, but the wrong species. A detailed
examination of the shipping containers showed that the permitted species had emerged
(and died) earlier in transit. The "pure" colony had contained two species. Even the
most careful rearing programs occasionally have contaminated colonies especially when
rearing difficult to distinguish microhymenoptera.
The quarantine officer or other personnel must have sufficient taxonomic training
to immediately separate individuals by species for rearing when field collected host
material is received. The parent generation is often dead before an identification is
obtained from specialists outside quarantine. Schlinger & Doutt (1964) provided a key
to the families of entomophagous insects which is a handy reference for quarantine
personnel. The general rule is to rear natural enemies for at least one generation to
prevent release of obligate hyperparasites and to discover if there is a problem with
pathogens. Weed control agents, especially Lepidoptera, occasionally have disease prob-
lems when reared in quarantine. Goodwin (1984) discussed diseases commonly encoun-
tered during insect rearing. At the FBCL we are fortunate to have strong taxonomic
support locally from Division of Plant Industry and University of Florida taxonomists
and pathologists and from ARS/USDA insect pathologists.
Florida Entomologist 75(4)
Diseased specimens of biocontrol interest may be sent to quarantine facilities by
insect pathologists conducting foreign exploration or by other foreign explorers. Dead
specimens are released to pathologists. Until recently diseased material was usually
sent directly to pathologists, but that appears to be changing. Regulations concerning
handling of insect pathogens are currently being formulated.
Often while the foreign explorer is packaging a shipment to send to quarantine and
while quarantine personnel are processing an incoming shipment, documentation of the
shipment seems unimportant compared to the need to keep the insects alive. Surpris-
ingly, documentation may be equally or even more important. Many shipments arrive
in quarantine with all material dead; many attempts to colonize species in or out of
quarantine fail and many species are not established in the field. Without adequate
quarantine documentation there would be no record of those projects. The information
on the importation forms and the voucher specimens are the only tangible results of the
project. Researchers often have difficulty recognizing the importance of documentation
and are wary of providing information they think might detract from their future pub-
lications. However, rarely are there conflicts with publications and every quarantine
facility should document projects as carefully as possible. Fisher (1964) illustrated a
card that was used to document shipments at University of California facilities. Coulson
(1987) illustrated and discussed record forms (941 & 942) used by federal and some state
quarantine facilities, including the FBCL, to document shipments into and out of
quarantine, respectively. Both the California and federal forms document the collecting
locations, collectors, dates, species shipped and emerged, packaging, transport method,
numbers shipped, condition received, disposition of material and other information.
Many quarantine facilities prepare an annual report of importations and exportations
that is distributed to other facilities and interested persons. It can usually be obtained
by writing the quarantine officer. These reports are valuable for communication but are
especially valuable as an historical record. They are usually produced from computer
data bases similar to that described by Dysart (1981). Coulson et al. (1988) published a
national summary of insects released in the United States during 1981 based heavily
upon the quarantine shipment forms. Publication of updates is planned.
Documentation of projects through a voucher insect collection is an important re-
sponsibility of quarantine facility personnel. The voucher collection, including specimens
of both the hosts and the natural enemies, can be maintained at the facility or in a
museum, although collections at both locations would be best.
Many introductions into quarantine are relatively easy to process and to move out
of quarantine. They may be pure laboratory cultures from well known institutions; they
may be a natural enemy group, for example Aphytis (parasites of armored scales) with
a proven safety record; they may be diseased insects that can be transferred to the
pathologist when they die. Other introductions are not so easy to process. Research of
both short and long duration may be needed before they can be cleared for release. New
species of parasites or species with unknown biologies in genera or families that include
hyperparasites must be reared and their behavior carefully observed. Fisher (1964)
suggested techniques to determine if a species is a hyperparasite. Predators and occa-
sionally parasites may require studies of their host ranges if they are not well known.
This is especially true today with the concern that society has for non-target native
organisms. Carabid beetles that eat mole cricket (Scapteriscus) eggs have been studied
Biological Control Workshop-'91: Buckingham
in the FBCL for five years in an attempt to evaluate their threat to native Orthoptera
or other groups.
Biology studies may be conducted while the researcher is waiting for a long-lived
natural enemy to die to obtain a specimen for identification or while awaiting an identifi-
cation of specimens that have been sent from one taxonomist to another. During a
project at the FBCL on fall armyworm (Spodoptera frugiperda (Abott & Smith)) para-
sites from South America, a researcher conducting laboratory cross-mating studies with
armyworms obtained permits to receive the pest moths from quarantine before we
obtained an identification and permits to release the parasites. Some projects involve
research that is considered safer if conducted in quarantine. During a project on fire
ant (Solenopsis invicta Buren) natural enemies, a suspected parasitic ant associated
with fire ant queens was studied in quarantine.
Weed control research requires long term studies in quarantine especially if there
has been little foreign research. Coulson & Soper (1989) discussed protocol for these
types of studies. The researcher must demonstrate through host range tests that the
control agent will not endanger crops or native plant populations. A federal inter-agency
Technical Advisory Group advises APHIS officials on the safety of the release. A similar
state committee advises Florida officials who must agree to the release before forward-
ing the application to APHIS. Agents tested at the FBCL during a project to control
the submersed aquatic weed Hydrilla verticillata (L.F.) Royle, were cleared for release
in as little as one year and as long as five years.
Various types of research in addition to biocontrol research may be conducted at
quarantine facilities. Extensive cross-mating studies to produce sterile hybrids of
Heliothis moths were conducted at the Stoneville quarantine facility. A multi-year pro-
ject was conducted at the FBCL to develop rearing techniques for species of mosquitoes
from California. The FBCL was chosen for the project because of local expertise in
mosquito rearing. Exotic cockroaches, orchid bees, and phengodid beetle larvae have
been confined to the FBCL quarantine in support of local research projects. Quarantine
is also used for rearing larvae of newly discovered immigrant insects to obtain adults
Countries conducting classical biocontrol projects should have a quarantine facility
to process the imported material or should have use of a regional facility. Formerly, a
quarantine room at the Commonwealth Institute of Biological Control in Trinidad
functioned like a regional facility for portions of the Caribbean basin because of the
many projects it supported. Perhaps a similar facility could be developed again. Because
a newly invading pest usually threatens many neighboring countries, a concerted control
effort with a system of several regional quarantines should be more efficient than
quarantines in every country. The quarantine facility at the CCBCA, El Zamorano,
Honduras, is planned as a regional quarantine for Central America.
As discussed previously, quarantine facilities need not be large. A renovated labora-
tory room could be sufficient for many small projects. I anticipate an increase in quaran-
tine facilities of all sizes in the United States as weed control projects increase and as
requirements for host range testing of parasites and predators increase. Quarantine
facilities are safe as currently operated and help allay public fears about importation of
exotic insects. I know of no reports of field establishments of pests that have escaped
from a quarantine facility in the United States.
If new quarantine facilities are developed throughout the Caribbean, it is imperative
that a documentation system be instituted and that voucher specimens be kept and
carefully maintained. It would be helpful if emphasis were placed on energy efficient
420 Florida Entomologist 75(4) December, 1992
design of the structure compatible with the quarantine function. This might help avoid
the large maintenance costs associated with quarantine facilities, especially as they
As pest control moves from the chemical age to the biological age, personnel at
quarantine facilities will help lead the way.
BARTLETT, B. R., AND R. VAN DEN BOSCH. 1964. Foreign exploration for beneficial
organisms, Chapter 9, pp. 283-304 in P. Debach and E. I. Schlinger [eds.],
Biological Control of Insect Pests and Weeds. Reinhold. New York, New York.
BOLDT, P. E., AND J. J. DREA. 1980. Packaging and shipping beneficial insects for
biological control. FAO Plant Prot. Bull. 28: 64-71.
COULSON, J. R. 1987. Release[s] of beneficial organisms, pp. 378-385 in L. Knutson,
F. C. Thompson, and R. W. Carlson [eds.], Biosystematic and biological control
information systems in entomology. Agr. Zool. Rev. 2.
COULSON, J. R., A. CARRELL, AND D. L. VINCENT. 1988. Releases of beneficial
organisms in the United States and territories 1981. USDA Misc. Publ. 1464.
COULSON, J. R., AND R. S. SOPER. 1989. Protocols for the introduction of biological
control agents in the United States, pp. 2-35 in R. P. Kahn [ed.], Plant Protection
and Quarantine. Vol. III Special Topics. CRC Press. Boca Raton, Florida.
DYSART, R. J. 1981. A new computer data bank for introduction and release of bene-
ficial organisms. Chapter 8, pp. 121-128 in G. C. Papavizas [ed.], Biological con-
trol in crop production. Beltsville Symposia in Agricultural Research 5. Al-
lanheld, Osmun. Totowa, New Jersey. 461 pp.
FISHER, T. W. 1964. Quarantine handling of entomophagous insects, Chapter 10, pp.
305-327 in P. Debach and E. I. Schlinger [eds.], Biological control of insect pests
and weeds. Reinhold. New York, New York. 844 pp.
GOODWIN, R. H. 1984. Recognition and diagnosis of diseases in insectaries and the
effects of disease agents on insect biology, pp. 96-129 in E. G. King and N. C.
Leppla [eds.], Advances and challenges in insect rearing. ARS/USDA. New Or-
leans, Louisiana. 306 pp.
HENDRICKSON, JR., R. M., S. E. BARTH, AND L. R. ERTLE. 1987. Control of relative
humidity during shipment of parasitic insects. J. Econ. Entomol. 80: 537-539.
LEPPLA, N. C., AND T. R. ASHLEY. 1978. Facilities for insect research and produc-
tion. USDA Tech. Bull. 1576. 86 pp.
ROSE, M., AND W. HARRISON. 1990. Quarantine laboratory procedures, Chapter 2,
pp.7-9 in D. H. Habeck, F. D. Bennett, and J. H. Frank [eds.], Classical biolog-
ical control in the southern United States. Southern Coop. Series Bull. 355. 196
SCHLINGER, E. I., AND R. L. DOUTT. 1964. Systematics in relation to biological con-
trol, Chapter 8, pp. 247-280 in P. Debach and E. I. Schlinger [eds.], Biological
control of insect pests and weeds. Reinhold. New York, New York. 844 pp.
Biological Control Workshop-'91: Ferguson 421
BIOLOGICAL CONTROL: AN INDUSTRIAL PERSPECTIVE
J. SCOTT FERGUSON
Vero Beach, Florida
The CIBA-GEIGY Corporation is the U.S. subsidiary of CIBA-GEIGY Limited,
Basel, Switzerland, one of the largest agricultural chemical companies in the world. The
focus of the Insect Control Sector is toward newer, safer technologies that fit into IPM
programs and are socially and environmentally acceptable. As part of this strategy
CIBA-GEIGY has major efforts in biological control, with priorities in Bacillus thurin-
giensis, Beauveria bassiana and Trichogramma spp. research. CIBA-GEIGY's first
Bacillus thuringiensis product, coded CGA-237218, is a transconjugate from Bacillus
thuringiensis subsp. aizawai and Bacillus thuringiensis subsp. kurstaki parents. This
transconjugate strain shows increased lepidopteran activity over that of the parental
strains, especially against pests in the genera Heliothis, Spodoptera, Pieris, and
Mamestra. EPA approval is expected in 1992. A cooperative effort between CIBA-
GEIGY and the University of Florida is in progress to develop a commercial Beauveria
bassiana product for control of the citrus root weevil complex in citrus. CIBA-GEIGY
also has cooperative projects underway with Canadian institutions and the USDA to
develop the parasitic wasp, Trichogramma minutum, for control of the eastern spruce
budworm, Choristoneura fumiferana, an economically important pest of Canadian
La corporation CIBA-GEIGY es una subsidiaria de CIBA-GEIGY Ltd., de Basel,
Suiza, la cual es una de las companies agroquimicas mas grandes del mundo. La filosofia
de la compania con respect al control de insects, se basa en el desarrollo de tecnologias
nuevas que ante todo sean inofensivas, que se puedan implementar en programs de
control integrado de plagas, y que sean aceptadas tanto social como ambientalmente.
Como parte de esta estrategia, la CIBA-GEIGY ha colocado sus mayores esfuerzos en
el area del control biologico con prioridad en Bacillus thuringiensis, Beauveria bassiana
y Trichogramma spp. el primer product de la CIBA-GEIGY con B. thuringiensis,
codigo CGA-237218, es una bacteria que ha sido transformada geneticamente y que
tiene como padres B. thuringiensis subsp. aizawai y B. thuringiensis subsp. kurstaki.
Esta bacteria es mas active que sus padres en el control de lepidopteros, especialmente
plagas del genero Heliothis, Spodoptera, Pieris, y Mamestra. Se espera que el EPA la
apruebe en 1992. La CIBA-GEIGY con la cooperation de la Universidad de Florida
estan desarrollando un product commercial de Beauveria bassiana para el control de los
gorgojos de las raices de citricos. La CIBA-GEIGY tambien esta cooperando con in-
stituciones canadienses y el USDA para desarrollar una avispa, Trichogramma
minutum, que parasite el gusano oriental de los apices del pino, Choristoneura
fumiferana, el cual es una plaga muy important en los bosques canadienses.
CIBA-GEIGY Corporation is the U.S. subsidiary of one of the largest chemical
companies in the world-CIBA-GEIGY Limited, headquartered in Basel, Switzerland.
CIBA (an acronym for Chemical Industry of BAsel) Limited began in 1884 as a manufac-
turer of synthetic dyes for the silk industry and soon ventured into the field of
medicines. The company later extended its research and production activities to other
branches of chemistry. Geigy was formed in 1758 and was also primarily concerned with
422 Florida Entomologist 75(4) December, 1992
dyestuffs but branched out in the 20th century into agricultural chemicals and phar-
maceuticals. In 1970 CIBA Limited and Geigy merged to form CIBA-GEIGY, which
today has affiliated companies in 50 countries. The Agricultural Division of the U.S.
company is headquartered in Greensboro, North Carolina, and has been an important
member of the agricultural industry for more than 40 years, manufacturing and dis-
tributing insecticides, herbicides, fungicides, plant growth regulators and, since 1974,
seeds (with the acquisition of Funk Seeds International). In 1983 CIBA-GEIGY opened
its Agricultural Biotechnology Research Unit which is dedicated to creating, in the
laboratory, new plants that cannot be developed through traditional breeding, including
the incorporation of resistance to insects.
Over the years many successful and useful insecticides have been developed by
CIBA-GEIGY: DZNR (diazinon), CuracronR (profenfos) and SupracideR (methidathion),
for example. In the 1970's new research resulted in the development of the insect
growth inhibitor cyromazine (TrigardR), that is highly selective for dipterous pests
(especially agromyzid leafminers) and effective at the very low rate (for the 70's) of 140
grams active ingredient per hectare (g ai/ha). Since then, newer insect growth inhibitors
have been found by CIBA-GEIGY that are active at rates as low as 1.0-2.0 g ai/ha.
As an industry leader in innovative technology, CIBA-GEIGY has focused its re-
search efforts in insect control toward newer and safer technologies. These technologies
include new chemical leads, behavior modifying chemicals, chemosterilants and growth
inhibitors, expertise in integrated pest management and the development of biological
control agents. To date the development of biological controls has focused on 1) Bacillus
thuringiensis, 2) the entomopathogenic fungi Beauveria brongniartii and B. bassiana,
and 3) parasites belonging to the genera Trichogramma and Encarsia. The following
is a discussion of CIBA-GEIGY's efforts in biological control.
CIBA-GEIGY's efforts with Bacillus thuringiensis (Bt) were initiated about 10
years ago, with major efforts in our Basel, Switzerland, laboratories. However, CIBA-
GEIGY Research and Development personnel worldwide are involved in collecting sam-
ples for isolation of new, potentially more active strains and in field testing of strains
that are in development.
The first strain of commercial development by CIBA-GEIGY is coded CGA-237218
(U.S. Patent 4,935,353; NCTC accession number 11821), with the suggested trademark
in the USA of AgreeTM. CGA-237218 is a transconjugate from Bt subsp. aizawai and
Bt subsp. kurstaki parents, with the HD-135 mutant serving as the receptor of genetic
material from the HD-191 mutant. The lepidopteran activity of this transconjugate is
increased over that of the parents, especially against pests in the genera Heliothis,
Spodoptera, Pieris and Mamestra (Table 1).
In the USA, field tests with CGA-237218 have been conducted primarily on vegeta-
ble crops, especially cole crops, tomatoes and leafy vegetables by CIBA-GEIGY, univer-
sity and private contract researchers. The effective use rate is 1120-2240 g/ha. Figure
1 is a summary of lepidopterous pest control in cole crops from 10 field trials across the
USA, demonstrating the activity of CGA-237218 on 3 major pests-Pieris rapae,
Trichoplusia ni, and Plutella xylostella. P. rapae, which is quite sensitive to Bt's, was
easily controlled by all rates of CGA-237218. T. ni required higher rates for effective
control while P. xylostella was intermediate in this respect.
CGA-237218 is in full development within CIBA-GEIGY, with full submission for
registration expected by mid-1991 and anticipated sales under the AgreeTM trademark
in late 1992. Initial marketing of a "50%" wettable powder product containing 2.5-5.0%
delta endotoxin (as determined by chromatography) is planned. Additional strains are
Biological Control Workshop-'91: Ferguson
0 20 40 60 80
% control (live larvae)
M 560 g/ha
EJZ3 1120 g/ha
B3 2240 g/ha
Fig. 1. Summary of lepidopterous pest control in cole crops with CGA-237218.
TABLE 1. BIOLOGICAL ACTIVITY OF CGA-237218 VERSUS ITS PARENTAL STRAINS.
LC5o (ppm) in Diet Bioassay
Insect Species HD 191 HD 135 CGA-237218
Heliothis armigera 48.0 228.0 44.0
Heliothis virescens 5.8 205.0 4.8
Spodoptera littoralis 10,000 445.0 330.0
Pieris brassicae 0.98 1.2 0.72
Mamestra brassicae >10,000 185.0 162.0
Florida Entomologist 75(4)
Insects are attacked in nature by fungi representing all classes. The 2 most impor-
tant groups of entomopathogenic fungi are the Deuteromycetes and Phycomycetes (Fer-
ron 1978). The genus Beauveria (Deuteromycetes: Moniliales) is the best known and
contains 2 species, B. bassiana and B. brongniartii (MacLeod 1954).
Entomopathogenic fungi differ from bacteria (e.g., Bacillus thuringiensis) and vir-
uses in that the most common route of infection is through the surface of the integument,
although infection can occur through the spiracles and possibly via ingestion (Gardner
& Noblet 1978). This advantage allows for the infection of insects independently of their
feeding activity. With Beauvaria, the generalized infection cycle proceeds as follows:
conidia contact the integument, germinate, penetrate (by mechanical pressure and/or
enzymatic activity) via production of a germ tube, enter the haemolymph, where hyphal
growth continues, and spread throughout the body resulting in death due to organ
destruction and/or toxin production (Madelin 1966, Roberts 1981). In the haemolymph
the fungus produces hyphal bodies, also called blastospores, which spread the infection
throughout the body by giving rise to more hyphae. In nature the blastospore stage is
found only inside the insect body. After host death, hyphae emerge from the cadaver
to produce conidia for further disease spread.
Although Beauvaria has the advantage of not having to be ingested, there are also
some disadvantages, the most notable of which are 1) the inactivation of infective units
by ultraviolet light (UV) and heat and 2) the problem of mass production and formulation
of infective units required for commercialization. To date there has been no successful
mass production and formulation of a Beauveria preparation on a commercial scale.
Beauvaria conidia and blastospores cannot be spray dried because temperatures of
50-70C kill them (Roberts & Campbell 1977). As a comparison, B. thuringiensis is
spray dried at over 100'C. However, Beauvaria conidia can be frozen for months with-
out detrimental effects (Muller-Kogler 1967). UV light is also detrimental to conidia and
may be the key reason for the disappointing results from foliar applications to control
the Colorado potato beetle, Leptinotarsa decemlineata (Coleoptera:Chrysomelidae)
(Gaugler et al. 1989). Roberts and Campbell (1977) suggest that insect control with
entomopathogenic fungi in soil habitats may be more rewarding (than foliar habitats)
due to higher moisture content, moderate temperatures and protection from UV light.
One of the first pests successfully controlled by the application of an en-
tomopathogenic fungus was the May beetle or European cockchafer, Melolontha
melolontha (Coleoptera:Scarabaeidae), by the fungus Beauveria brongniartii (Keller &
Zimmerman 1989). Adult cockchafers swarm between the end of April and the beginning
of June, with most swarming concentrated along borders of forests (Keller et al. 1989).
A feeding, mating and egg maturation period of about 7-10 days is followed by a return
of the female to breeding sites, where eggs are deposited at a depth of approximately
10-20 cm. Larvae eclose after 4-6 weeks, passing through 3 larval instars during their
2 year development (Keller et al. 1989). Larvae feed on roots of a variety of orchard,
vineyard and pasture crops, where sufficient numbers can cause heavy damage. In
addition, such high value crops as strawberries are also attacked.
B. brongniartii has been observed as a pathogen of cockchafers for over 100 years.
A major problem in using B. brongniartii to control cockchafers in the field has been
introducing the inoculum into the soil where it can come into contact with larvae. There
are 2 possible methods of contaminating the breeding sites: 1) introduction of inocula
by mechanical methods and 2) introduction using the egg-depositing females as vectors
(Keller et al. 1989). In the cockchafer control program in Switzerland and Italy, swarm-
ing adults are sprayed (by helicopter) with a blastospore suspension, with the treated
adult female bringing the infective unit into the soil when she returns to oviposit (Keller
1986). For several seasons B. brongniartii blastospore suspension was produced by
Biological Control Workshop-'91: Ferguson
CIBA-GEIGY Pharma Division, Basel, Switzerland, in a 4000 liter fermenter. The goal
of the cockchafer control program is to reduce populations over time, as the fungus acts
slowly and long-term observations are necessary to evaluate efficacy (Keller & Zimmer-
Although blastospores are not infective units in nature, they are as infective as
conidia in bioassays (S. Keller, personal communication). However, because conidia are
the infective units in nature and are more resistant to adverse environmental conditions,
therefore probably more tolerant to formulating, a commercial Beauveria product would
be most feasible if conidia could be produced in culture and formulated to enhance
stability and ease of application. For these reasons, CIBA-GEIGY has conducted re-
search in liquid production of Beauveria conidia. In the USA a cooperative effort be-
tween CIBA-GEIGY and the University of Florida is in progress to develop B. bassiana
for control of the citrus root weevil complex. This complex is made up of 5 weevil
species: Artipus floridanus, Diaprepes abbreviatus, Pachnaeus litus, P. opalus and
Pantomorus cervinus (Coleoptera:Curculionidae). Citrus root weevils appear to be an
excellent candidate for control by B. bassiana for several reasons: 1) a highly virulent
citrus root weevil strain of B. bassiana exists; 2) eggs are laid in the citrus tree canopy
and, upon eclosion, neonate larvae drop to the soil to burrow to the roots, thus a
susceptible stage is present in the soil habitat where conidial survival is more likely (as
discussed by Roberts and Campbell, 1977); and 3) no effective larval controls are cur-
To date, the infectivity of liquid-produced B. bassiana conidia versus aerial conidia
(produced on Sabourad Dextrose Agar [SDA]) has been compared in laboratory bioas-
says on larvae of 3 insect species-Colorado potato beetle, Leptinotarsa decemlineata
(Coleoptera:Chrysomelidae), Diabrotica balteata (Coleoptera:Chrysomelidae), and the
little leaf notcher, Artipusfloridanus. In addition, the infectivity of liquid-produced B.
brongniartii conidia versus the CIBA-GEIGY blastospore suspension was compared in
a laboratory bioassay on third instar M. melolontha (Coleoptera:Scarabaeidae) larvae
by Dr. S. Keller, Swiss Federal Research Institute, Reckenholz, Switzerland. In all 4
cases liquid-produced conidia were as infective as SDA-produced conidia and the blastos-
pore suspension (Figs. 2 & 3).
Research is in progress to optimize conidial yields in liquid culture and develop a
commercially acceptable formulation of Beauveria bassiana.
Another area of biological control that CIBA-GEIGY is pursuing is the use of the
parasitic wasp species, Trichogramma minutum and T. pretiosum (Hymenoptera:
Trichogrammatidae), for insect control. Trichogramma spp. wasps parasitize the eggs
of over 200 lepidopteran species, including such economically important pests as the
tobacco budworm, Heliothis virescens (Noctuidae); the bollworm/corn earworm/tomato
fruitworm, Helicoverpa zea (Noctuidae); the eastern spruce budworm, Choristoneura
fumiferana (Tortricidae) and the European corn borer, Ostrinia nubilalis (Pyralidae).
Effective pest control with Trichogramma spp. has been demonstrated in Europe where
the European corn borer is controlled by T. evanescens on approximately 20,000 ha of
There are 2 major obstacles to overcome in the commercial development of Tricho-
gramma spp. as insect control agents: 1) production costs and scale up and 2) unreliable
quality. Lesser problems include method of application and effective rates of release.
However, there are opportunities as well: one production process could be utilized to
control over 100 lepidopteran pests, cost effective control with Trichogramma would
increase the volume of use, and the social and environmental acceptability of biological
Florida Entomologist 75(4)
A. floridanus ;
0 20 40 60 80 100
Source of Conidia
liquid = aerial (SDA)
Fig. 2. Comparison of Beauveria bassiana infectivity with liquid-produced versus
aerial-produced (on SDA) conidia.
CIBA-GEIGY has 2 cooperative Trichogramma projects in progress in North
America. The largest, most visible and furthest advanced is the Spruce Budworm Con-
trol Project in cooperation with the University of Toronto, University of Guelph and
the Ontario Ministry of Natural Resources. The eastern spruce budworm, Choris-
toneurafumiferana, is a major forestry pest in Canada. From 1977 to 1981 the average
annual loss (defoliation) due to this pest was 44 x 106m', equivalent to 66% of the volume
harvested in the same period (Gross 1985). Five Canadian provinces (Ontario, British
Columbia, Manitoba, Quebec and Nova Scotia) have adopted a non-chemical approach
to forest insect pest control (Smith et al. 1990). In 1979, 1% of the forest area treated
with insecticides was treated with Bacillus thuringiensis subsp. kurstaki (B.t.k.) and
99% with chemical insecticides while in 1988, of the 750,000 ha treated, 64% was treated
with B.t.k. and 36% with chemical insecticides (Smith et al. 1990). The long-term goals
of the CIBA-GEIGY Canada project are the eventual replacement of current chemical
controls with Trichogramma minutum and enhancing the growth and development of
Biological Control Workshop-'91: Ferguson
100 r- -.
/ /A ___
Days post inoculation
Fig. 3. Comparison of Beauveria brongniartii infectivity with liquid-produced
conidia versus liquid-produced blastospore suspension on L3 Melolontha melolontha.
our knowledge in the area of biological control. The 3 step research project will investi-
gate 1) rearing in vivo, 2) species selection and 3) Trichogramma minutum releases
into areas where eastern spruce budworm is a problem. To this end, CIBA-GEIGY is
in the process of building a Trichogramma rearing facility near Toronto. In tests con-
ducted by Smith et al. (1990) rates of 8-23 x 106 females/ha resulted in 42-83% reduction
in larval densities the following year while a release of 23 x 106 females/ha, followed by
one application of B.t.k. the following spring, reduced spruce budworm populations by
The second cooperative project with Trichogramma is with the USDA to develop
an in vitro production method for T. minutum and T. pretiosum.
/ 'A "
428 Florida Entomologist 75(4) December, nrnz
Additionally, another CIBA-GEIGY biological control project is in progress in
Europe to develop another parasitic wasp species, Encarsia formosa, for biological
control of whiteflies in protected crops (i.e., glasshouse vegetable production).
NEMATODES AND VIRUSES
CIBA-GEIGY has exclusive marketing and development rights for professional turf
and ornamental uses of entomogenous nematodes produced by the Biosys Corporation.
The first labeled product will be Exhibit (Steinernema carpocapsae strain 27) for control
of various turf and ornamental pests, including larvae of: black vine weevil, Otiorhyncus
sulcatus (Coleoptera:Curculionidae); strawberry root weevil, 0. ovatus (Coleoptera:
Curculionidae); the Japanese beetle, Popillia japonica (Coleoptera:Scarabaeidae); fun-
gus gnats, Bradysia spp. (Diptera:Mycetophilidae); and various armyworms, cutworms,
sod webworms and billbugs in turf.
Additionally, CIBA-GEIGY has tests in progress to evaluate insect pathogenic vir-
uses as insect control agents.
In summary, CIBA-GEIGY has a large effort in biological insect controls, with
major priorities in Bacillus thuringiensis, Beauveria bassiana and Trichogramma spp.
research and development. These efforts are in harmony with CIBA-GEIGY's strategy
to focus on newer, safer technologies that fit into IPM programs and are socially and
FERRON, P. 1978. Biological control of insect pests by entomogenous fungi. Ann. Rev.
Entomol. 23: 409-442.
GARDNER, W. A., AND R. NOBLET. 1978. Effect of host age, route of infection, and
quantity of inoculum on the susceptibility of Heliothis virescens, Spodoptera
eridania, and S. frugiperda to Beauveria bassiana. J. Georgia Ent. Soc. 13: 215-
GAUGLER, R., S. D. COSTA, AND J. LASHOMB. 1989. Stability and efficacy of
Beauveria bassiana soil inoculations. Environ. Entomol. 18: 412-417.
GROSS, H. L. 1985. The impact of insect and diseases on the forests of Ontario. Cana-
dian For. Serv. Info. O-X-366. 96 pp.
KELLER, S. 1986. Control of May beetle grubs (M. melolontha L.) with the fungus
Beauveria brongniartii (Sacc.) Petch, pp. 525-528 in R. A. Samson, J. M. Vlak
& D. Peters [eds.]. Fundamental and applied aspects of invertebrate pathology.
Foundation of the Fourth Internatl. Colloquium of Invertebr. Pathol.
Wageningen, The Netherlands.
KELLER, S., E. KELLER, C. SCHWEIZER, J. A. L. AUDEN, AND A. SMITH. 1989.
Two large field trials to control the cockchafer M. melolontha L. with the fungus
Beauveria brongniartii (Sacc.) Petch. BCPC Mono. No. 43 Progress and Pros-
pects in Insect Control.
KELLER, S., AND G. ZIMMERMAN. 1989. Mycopathogens of soil insects. Chapter 10,
pp. 240-270 in N. Wilding, N. M. Collins, P. M. Hammond & S. F. Webber
[eds.], Insect-fungus interactions. Academic Press, London.
MACLEOD, D. M. 1954. Investigations on the genera Beauveria Vuill. and
Tritirachium Limber. Canadian J. Botany 32: 818-890.
MADELIN, M. F. 1966. Fungal parasites of insects. Ann. Rev. Ent. 11: 423-448.
MULLER-KOGLER, E. 1967. On mass cultivation, determination of effectiveness, and
standardization of insect pathogenic fungi, pp. 339-353 in Van der Loan [ed.],
T-v i ^ f\r\c^
. -- a- m W -
Biological Control Workshop-'91: Andrews et al. 429
Proc. International Colloq. Insect Pathol. Microbial Control, 1966. North Holland.
ROBERTS, D. W. 1981. Toxins of entomopathogenic fungi, Chapter 23, pp. 441-464 in
D. Burges [ed.], Microbial control of pests and plant diseases, 1970-1980.
Academic Press, London.
ROBERTS, D. W., AND A. S. CAMPBELL. 1977. Stability of entomopathogenic fungi.
Misc. Publ. Ent. Soc. Amer. 10: 19-76.
SMITH, S. M., J. R. CARROW, AND J. E. LAING. 1990. Inundative release of the egg
parasitoid, Trichogramma minutum (Hymenoptera:Trichogrammatidae),
against forest insect pests such as the spruce budworm, Choristoneura
fumiferana (Lepidoptera:Tortricidae). The Ontario Project 1982-1986. Memoirs
Ent. Soc. Canada No. 153.
ENHANCING BIOLOGICAL CONTROL'S CONTRIBUTIONS
TO INTEGRATED PEST MANAGEMENT THROUGH
APPROPRIATE LEVELS OF FARMER PARTICIPATION
KEITH L. ANDREWS,12 JEFFERY W. BENTLEY,' RONALD D. CAVE'
'Department of Crop Protection
Escuela Agricola Panamericana
P.O. Box 93
Tegucigalpa, Honduras C.A.
2Department of Entomology
University of Florida
Most recent development literature calls for greater farmer participation in agricul-
tural research and technology transfer. Interestingly, biological control specialists do
not seem to be involved in the trend. Four methodological models for developing and
implementing biological control are proposed and then analyzed for their applicability
to the Caribbean and Central America. Profit-generating biological inputs can be de-
veloped without farmer involvement, but grower involvement is required in the im-
plementation phase. Inoculative releases and classical biological control do not require
farmer involvement in the implementation phase, but may benefit from farmers' support
and participation in the research and development phase. Alternatively, conservation
and manipulation techniques require extensive farmer involvement in both the research
and implementation phases. Unfortunately, biological control researchers generally ig-
nore farmers as collaborators, even when their participation is key for implementation
in heterogeneous agroecological and socioeconomic environments. Biological control in
developing neotropical countries is seriously limited by financial and personnel con-
straints; a series of difficult strategic and operational decisions must be made if biological
control is to contribute significantly to IPM in the area.
La literature de desarrollo reciente hace un fuerte Ilamado para una mayor participa-
cion de agricultores en la investigation agricola y transferencia de tecnologias. Sorpren-
430 Florida Entomologist 75(4) December, 1992
dentemente, los especialistas en control biologico parecen no estar involucrados en la
discussion. Se proponen y analizan cuatro models metodologicos para el desarrollo e
implementation del control biologico por su aplicabilidad en la zona del Caribe y Centro
America. Se puede desarrollar products biologicos que generen ganancias sin involu-
crar al productor en la fase de implementation. Las liberaciones inoculativas y el control
biologico clasico no requieren de la participation del agricultor en la fase de implemen-
tacion, pero puede beneficarse del apoyo y participation de los agricultores en las fases
de investigation y desarrollo. Alternativamente, las tecnicas de conservation e im-
plementacion requieren de la participation extensive del agricultor en las fases de inves-
tigacion e implementation. Desafortunadamente, los investigadores en control biologico
generalmente ignoran a los agricultores como colaboradores, aun cuando su participation
es clave para la implementation en ambientes agroecologicos y socioeconomicos
heterogeneos. El control biologico en los paises neotropicales en desarrollo esta
seriamente limitado por la falta de recursos financieros y personales; se precisa tomar
una series de decisions estrategicas y operacionales para que el control biologico pueda
contribuir significativamente al MIP en el area.
Regional IPM has a mixed record of accomplishments. There have been several
major success stories in Central America, most notably in capital-intensive export crops
such as sugar cane, coffee, and bananas (Hansen 1987, Andrews & Quezada 1989).
However, there has been virtually no impact on small producers. This appears to be
true of the Caribbean as well. This may be due to a conspiracy of inimical socioeconomic
factors such as inadequate and discontinuous research and outreach efforts, poor infras-
tructure and inappropriate macroeconomic and social policies (Andrews & Bentley
1990). However, we consider that the fundamental problem has been inappropriate
strategy and modus operandi of past and existing IPM research and extension pro-
In this paper, we make several suggestions as to how biological control may contrib-
ute more to IPM, focussing on methodological issues, especially farmer participation.
Programmatic statements about farmer participation in agricultural innovation were
provided by Biggs & Clay (1981), Chambers & Jiggins (1987), Rhoades (1987), Richards
(1989); see Amanor (1989) for a list of hundreds of other references. Symposia edited
by Youngberg & Sauer (1990) and Bentley (in press c) focused on farmer participation
in agricultural development; the latter includes several references to plant protection.
FARMER PARTICIPATION IN PEST MANAGEMENT IMPLEMENTATION
Pest management techniques can be categorized as a function of the degree to which
farmers participate in their implementation.
1. The Farmer Bypass Model attempts to use standardized technologies which are
applicable over wide areas. This model supposes minimal farmer involvement. It is best
represented by the sterile insect method which requires only imposed compliance on
the part of farmers. It is also exemplified by the approach taken by plant breeders.
Farmers' involvement is limited to merely choosing the seed variety to sow. Perhaps
the most extreme form of the farmer bypass model is classical biocontrol, requiring
absolutely no farmer involvement. The introduction and establishment of exotic natural
enemies is something done for and to farmers' production systems, and it does not
require their active participation or sanction. Similarly, inoculative releases of natural
enemies carried out by government or industry-for example, a sugar mill-also benefit
farmers without involving them.
There are several major successes which can be attributed to these techniques; all
of them have involved minimal extension. In fact, it can be argued that the success of
Biological Control Workshop-'91: Andrews et al.
the farmer bypass model is due to the straightforward research uncomplicated by out-
reach. This technique looks at bugs and not very much at people; it can be highly cost
effective when successful.
2. Farmer as Protagonist Model. An alternative approach recognizes the ecological
and socioeconomic heterogeneity of agroecosystems. It is impossible to extend inflexi-
ble, standardized technologies over large areas. Efforts to centralize delivery and use
of IPM procedures and services like scouting and spraying are incompatible with the
need to make and implement rapid, fine-tuned site specific decisions at the farm level
(Goodell 1984). Those who know the Caribbean and Central America understand the
problems confronting IPM specialists in the region because farms located only a couple
of kilometers from one another are often more different than those located at the far
ends of the US cornbelt (Andrews & Bentley 1990).
This approach makes use of flexible technologies which are applied locally, and farm-
ers are viewed as the protagonists of change, that is the implementors of the
technologies. Most IPM programs focus on field-level control and management and
require some form of scouting as the basis for making decisions regarding the use of
chemical, mechanical or cultural practices. Inundative releases and the use of microbial
pesticides also view the farmer as protagonist. Conservation and manipulation efforts
depend on local initiative to be successful.
The literature reports some successes using the farmer-as-protagonist model, espe-
cially among large scale, capital intensive farmers. However, many of the successes are
based on the existence of pest management advisors who serve as key links, and in fact
are more the implementors than are the farmers themselves. Among small holders we
find few non-chemical successes despite extensive (although sometimes flawed) exten-
FARMER PARTICIPATION IN PEST MANAGEMENT RESEARCH
Biological control research and development (R&D) commonly involves three princi-
pal steps: Step 1. Inventory determines which natural enemies exist in the agroecosys-
tem. Step 2. Evaluation determines, generally quantitatively, what these natural
enemies are doing to regulate pest populations. Step 3. Sometimes referred to as manip-
ulation, but probably better called action, this step involves research to develop
technologies to improve the impact of the natural enemy complex. It can take three
a. Conservation/manipulation involves the protection and enhancement of natur-
ally occurring biological control through modification of the habitat;
b. Augmentation takes place through the liberation of mass reared natural
c. Classical biological control involves the introduction and establishment of exo-
tic natural enemies
There are at least six general justifications for farmer participation in R&D efforts. All
of them seem generally applicable to biocontrol.
Knowledge. First, farmers are viewed as a source of indigenous knowledge
(Thurston 1990, Farrington & Martin 1987). Farmers are knowledgeable, especially
about cultural practices, germplasm and botanical insecticides (see Ileia Newsletters
No. 6 and 4(3), Glass & Thurston 1978, Altieri 1986, Litsinger et al. 1978, Risch 1981,
Matteson et al. 1984). Interestingly, these authors make little mention of farmer knowl-
edge of biological control. Could it be that farmers have little to contribute to biological
control research and development efforts? Huang & Pei (1987) reported that Chinese
farmers consciously and effectively manipulate ants for control of pests in citrus trees.
Florida Entomologist 75(4)
Asian farmers use ducks to control paddy pests (NAS 1977). Altieri et al. (1977) con-
tended that farmers often selectively weed fields in a way that leaves plants that im-
prove naturally occurring biological control. Vertebrate predators are consciously pro-
tected and birds are creatively manipulated by rice farmers in Sri Lanka (Upawansa
1989). Not withstanding the previous three Asian and one Latin American examples,
we hypothesize that biological control is rarely, if ever, an important element of tradi-
tional farmers' paradigms. Unfortunately, few conventional modern farmers appreciate
it either. Virtually all references to small farmer biocontrol (e.g. Hellpap 1986, Hansen
1987) focus on their use of research-derived techniques, not indigenous practices. The
idea that proliferation of some insects is desirable is counterintuitive (Goodell 1984).
Empowerment. The second justification for involving farmers in research is that
disenfranchised farmers may be empowered through this activity. It is important to
show the positive aspects of traditional systems, to realize and fully document what
farmers are doing right, and recognize that it may be very difficult to farm better under
the marginal conditions that they confront. Innovative, participative R&D activities can
help rural people to develop confidence and pride in their own knowledge systems and
technological capabilities. With an enhanced sense of worth and power, farmers are
better able to select and adapt from among exogenous technologies offered them, and
consciously preserve desirable elements of their endogenous technologies (Richards
1985, Altieri 1989, Thrupp 1989).
We need to make it clear when we talk about legitimizing traditional systems, we
are not supporting the simplistic notion that modern day poor farmers should continue
to use the same technology that they used generations earlier; rather, tradition must
co-exist with innovation. Farmers constantly reinterpret traditional knowledge in an
ever-changing world and innovate. Farmers are conservative only insofar as they main-
tain technologies that work (Rhoades 1987, Richards 1989, Bentley 1992). Johnson
(1972) goes so far as to say that farmers are too willing to change, and adopted green
revolution technologies too quickly.
Empowerment through effective participation has strong political dimensions since
it can help marginalized people develop a sense of solidarity and collective political
bargaining power (Thrupp 1989). This attitude appears to us to be a prerequisite for
genuine development in the polarized, inegalitarian societies of the Caribbean and Cen-
Focus. Third, farmers' participation in planning and evaluation of R&D helps to
focus that research. Farmers can contribute to discussions that lead to strategic deci-
sions regarding what investigation is to be done, what this experimentation should
accomplish, what evaluation criteria are most important, and what is the time frame to
be used for judging the productivity of the program. Farmers also can help to identify
opportunities for change as well as constraints to that change. Their participation may
help to evaluate progress. The underlying assumption is that researchers trying to
develop workable, cost-effective technologies are more effective when they are account-
able to producers. This is the opposite attitude to that expressed to the senior author
by an unnamed University of Florida Institute of Food and Agricultural Sciences ad-
ministrator who reported that his job was "to isolate his researchers from grower
Advisory and commodity boards may provide intellectual support for short and long
term research efforts. Wealthy producers may use their political power to allocate
public funds and orient researchers' work. The autonomous or semi-autonomous Com-
modity Research Institutions which work on behalf of coffee and sugarcane producers
in the region have been highly effective in developing IPM programs. This may be due
to the political and financial strength of the beneficiary farmers which assures that their
needs and priorities are responded to by researchers and technical people. Finally, in
Biological Control Workshop-'91: Andrews et al.
the transnational fruit companies which have operated in Central America and the
Caribbean for nearly a century, researchers do not act in isolation from the people in
charge of production; both answer to the same boss. Therefore, researchers are account-
able to the production managers who participate in strategic planning and orientation
of research and development efforts; just as the researchers have to keep their research
practical, the production people are compelled to implement the research results (An-
drews in press).
Unfortunately, smallholders are rarely involved in strategic planning of investiga-
tion. The failure to involve them may be the most important factor limiting IPM R&D
in developing neotropical countries. Unless farmers are involved in strategic planning
it is unlikely that technologies useful to them will be developed. Certain alternative
groups (e.g. non-governmental private voluntary organizations) may stimulate strategic
participation routinely; unfortunately these groups generally lack scientific expertise to
make full use of the farmers' suggestions and respond fully to their needs.
Support. Fourth, farmer participation in R&D efforts can be justified in terms of
the financial, logistical and intellectual support they provide to the operational aspects
of experimentation. Farmer collaborators often provide inputs, labor and land to sup-
port experimentation. Commercial farmers may provide partial support for research,
as in the case of commodity boards. Farmers may also provide intellectual input in the
field. This may include plot selection, development of evaluation criteria, or interpreta-
tion of experimental results. Who knows better than the farmer if a new variety has
desirable agronomic characteristics or if a proposed cultural practice fits into the produc-
tion system (Bentley & Melara 1991). Farmers rarely adopt technologies without mod-
ification; they adapt technologies by creatively combining elements of existing and pro-
posed technologies (Horton 1984, Rhoades 1987). The participation of farmers in on-farm
trials is very common in Central America; it is in many ways the essence of the farming
systems research and extension approach (Hildebrand 1986). It may be highly cost
effective and desirable. However, with small-holders it may be a serious imposition. It
costs farmers considerable time, effort and resources to participate in this way. We feel
that it is difficult to justify this category of involvement if at least some of the cooperat-
ing farmers or their representatives have not previously participated extensively in the
strategic orientation of the research program. Ashby (1986) discussed different
mechanisms for including small farmers in on-farm experimentation involving seed and
Farmers may have considerable impact on experiments to evaluate cultural prac-
tices, weed management and host plant resistance. However, it is not clear what role
they may have in biological control. Ard6n et al. (in press) reported that Honduran
farmers who collaborated in our cabbage IPM program contributed more than half of
the ideas to on-farm experiments when cultural practices were studied. They contri-
buted less than 1/3 of the ideas on biological control and less than 1/5 of the ideas in
studies of pest biology. It may be that farmers are masters of cultural practices, weed
science (Chac6n & Gliessman 1982) and development of new varieties under their par-
ticular farm conditions. However, it seems they lack either the paradigm or the research
tools to participate extensively in development of useful biocontrol technologies.
Validation. Also, farmer participation in the R&D effort allows researchers to com-
pare technologies' performance under realistic farm conditions including management
skill levels that differ from those of experiment stations (Lockeretz 1987). Farmers can
play an important role in the local adaptation and popularization of technologies (Horton
1986, Bunch 1982). Their involvement in polishing techniques and applying them effec-
tively to local conditions is a step that only they can carry out. Unfortunately, in the
development literature this activity is oftentimes referred to as participatory research.
This top-down approach is so common that a recent anthology allegedly on farmer
Florida Entomologist 75(4)
participation in technology generation described little more than case after case of
adaptation of fertilizers and new seed varieties in on-farm trials (Matlon et al. 1984).
The implication is made that this level of participation is adequate (e.g. Reid et al. 1988).
It is not. It is only a final phase in the continuum of activities from strategic to applied
to adaptive research. Moreover, if farmer participation is limited to this important
stage, they will not significantly influence which research results are made available to
Modern research organizations, especially those doing strategic and applied re-
search, often fail at the stage of local adaptation. By creating alliances with private
voluntary organizations (PVOs) and certain non-governmental organizations (NGOs,
e.g. grower groups) research organizations can maximize their effectiveness. PVO and
NGO collaborators are excellent community organizers and are capable of stimulating
validation and adaptation efforts (Altieri 1989, Farrington & Biggs 1990).
Extension. Finally, participation jump starts extension efforts; it gives at least a
few farmers experience with the new technologies, thereby allowing them to adapt and
by word of mouth to begin to extend these technologies to their neighbors. Follow up
outreach by farmer-experiments is an integral part of the village level adaptive research
described by most advocates of farmer participation.
WHEN IS IT WORTHWHILE TO INVOLVE FARMERS IN BIOCONTROL?
In most cases, farmers are the ones who have to do what is necessary to make a
given technology work, and in most cases they must repeat it every single year. In
order to implement a technology, they need to know of it, understand it, believe in it,
want to apply it, and have the resources to use it. (Rogers 1983 provides a longer and
more technical list of the requirements for technology adoption and use.) With biological
control techniques like conservation and manipulation, farmers have to be reached by
linkage farmers, paratechnical people or extensionists. Each farmer has to try out the
technologies, then satisfy him or herself on all counts and conclude that the technologies
are feasible, efficacious and profitable. However, with the bypass method, farmers are
involved in neither validation nor implementation. Because it is so simple it is no wonder
researchers and policy makers alike prefer the bypass model! However, how widely
applicable is it? In the following section, we judge the potential of each biological control
approach as a function of farmer participation required to make it work.
There are two approaches that are subcategories of the farmer bypass model.
1. Nonparticipatory research-Nonparticipatory implementation. This common
bypass model is represented by classical biocontrol. Farmers benefit but participate
neither in the R&D nor implementation phase. Only researcher-initiated classical
biocontrol fits in this category.
2. Participatory research-Nonparticipatory implementation. This is an uncommon
model in which grower participation is required prior to a bypass implementation phase.
Farmers may mandate classical biocontrol programs (Van Driesche 1989) and perhaps
modify temporarily their agricultural practices to help establish beneficial organisms.
This appears to have been the case with the Vedalia beetle in California over 100 years
ago. They may cooperate by withholding some or all spray applications while releases
are made. After establishment, however, farmer involvement is limited to the role of
beneficiary. This approach may also be represented by inoculative releases of natural
enemies in which either governmental organizations or grower organizations such as
sugar mills make periodic liberations of natural enemies in the farmers' fields and farm-
ers participate only as beneficiaries.
Is this model not the best of all worlds? Farmers provide strategic input and direc-
tion to the development of technologies, then sit back and enjoy the fruits of the
Biological Control Workshop-'91: Andrews et al.
technologies. However, beyond classical biocontrol and inoculative releases we feel that
no other options exist for using this model.
There exist two variants of the farmer-as-protagonist approach.
3. Nonparticipatory research-Participatory implementation. This common, conven-
tional model is top-down and assumes that basic and strategic researchers pass ideas
and information to applied researchers who then pass their results to extensionists, and
finally the extensionist moves a well-polished product to the user. It is undeniable that
this model has been effective when focused on development of inputs such as chemical
insecticides. It is probably effective when microbial insecticides or inundative releases
of parasitoids or predators are involved. (It is the prevailing model represented in this
symposium.) Because chemical and microbial inputs can be packaged and farmers taught
to use them in relatively simple ways, this model has been successful. Private companies
work closely with public sector researchers to develop the technologies, then undertake
massive, profit-motivated outreach programs to make them work.
However, this approach is not effective when information rather than inputs is to
be extended. It cannot be useful for the development and popularization of conservation
and manipulation techniques. Nor is it an effective approach to use where agroecosys-
tems are heterogeneous because generally, it is impossible for researchers to under-
stand the complex realities confronting farmers and to adjust technologies for all these
complexities, repeating the process in each ever-changing agroecological and
socioeconomic domain (Biggs & Clay 1981). This model will fail if researchers do not
take into consideration all key factors as they carry out their research; unfortunately,
many of these factors (especially socioeconomic ones) fall outside the biological control
researchers' area of expertise and interest.
Except for research and development of inputs, this approach has been and continues
to be a major waste of scarce biological control resources. If we really want biocontrol-
based IPM to work in the Third World, biocontrol specialists working on information-
laden approaches to be used in heterogeneous environments should avoid this model,
using the following one instead.
4. Participatory research-Participatory implementation. This is a costly and chal-
lenging approach that may be the only way to develop and implement conservation and
manipulation techniques. It assumes local control of the research process, a local focus
and local benefit. It is presently extremely uncommon. Is it a fruitful, cost-effective
Our enthoentomological studies have shown that farmer knowledge can be extensive
but partial. While farmers have extensive understanding of wasp and bee natural his-
tory and a complex folk taxonomy, they lack the key concept that some insects eat or
develop on others, i.e. they lack the basic paradigm for biological control. We have
observed, however, that some farmers, when exposed to biocontrol concepts, begin to
experiment with large predators (as Polybia spp.) and seek ways to incorporate easily
observable natural enemies into their systems.
It is our hypothesis now that it is not necessary to provide farmers with finished
biocontrol technologies; it may be sufficient to provide them with key missing informa-
tion (e.g. wasps eat worms that eat crops) and let them apply the new information
locally, developing their own local technologies. The key questions to be answered are:
will significant numbers of farmers use the information we provide to develop means to
manipulate vertebrates, wasps and other large predators? Can similar techniques be
used to stimulate farmers' innovations with smaller, less easily observed natural
enemies? Can researchers effectively backstop farmers who have been trained in biocon-
trol precepts? Can the informal research of farmers and the formal research of scientists
be cost-effectively linked to develop efficacious, locally useful new biocontrol proce-
dures? In 1992, we will provide biological control training to over 500 farmers and
extensionists in the hope that answers to these questions will be positive.
436 Florida Entomologist 75(4) December, 1992
CONCLUSIONS AND RECOMMENDATIONS
Based on the preceding analysis we arrive at six hypotheses that border on being
1. Education of technology users is key. Farmers and extensionists who do not
understand the basic precepts of biocontrol and its potential neither adopt it nor partici-
pate in its development or local adaptation.
2. We must face a hard question. Where do we have the highest probability of
obtaining the greatest benefit for the least research and implementation costs? We must
apply our scarce resources there. How many scientists with postgraduate specialization
in biocontrol are working (i.e. not administrating) in the entire region which is made
up of more than 20 countries and over 50 million people. Where do we start? What
should our priorities be? Among the high priority options we identify are:
- Natural enemy complex enhancement with known winners. This approach promises
a few large scale success. Quezada (1989) listed 17 of the best candidates for classical
biological control efforts in Central America. This approach requires biological con-
trol specialists and does not necessarily need farmer participation. It is highly cost-
effective when successful. This is a major focus of the work carried out by
Zamorano's Center for Biological Control in Central America.
- Conservation and manipulation efforts almost certainly will permit many small, local
successes. A few biological control specialists may effectively multiply their impact
by working with the many IPM generalists who work with farmers in different
agroecological and socioeconomic domains. We need to allocate a substantial propor-
tion of our scarce resources to extension and education to make this approach work.
- Since the private sector is likely to invest increasingly large sums in microbial and
other augmentative approaches, resource-limited public sector institutions and de-
velopment programs should not focus their scarce human and financial resources on
what is rapidly becoming a viable alternative due to the efforts of scientists employed
in developed countries and by transnational companies. Regional biocontrol
specialists should let IPM generalists and pesticide application technologists (who
need new ways to contribute) do the screening and fit the new inputs into IPM
systems. Unfortunately, the grant money and publication opportunities presented
by this area will increasingly tempt biocontrol specialists away from the alternative
3. Certain social equity issues need to be addressed as we decide how best to use
our scarce biocontrol research and development resources. Classical biocontrol is prob-
ably one of the most scale-neutral technologies available; it is democratic and is as likely
to benefit small scale producers as it is to benefit large ones. Conservation and manip-
ulation techniques, on the other hand, appear to always favor one socioeconomic group
or another; what works with large scale, capital intensive producers will not be useful
among small holders, and vice versa. We can, therefore, choose to produce conservation
and manipulation technology that favors either the privileged or the disenfranchised.
Through development of conservation and manipulation technologies we can address in
a small way class differences. Augmentation techniques, on the other hand, are probably
useful only-or certainly mainly-to the relatively well-to-do. Patented, packaged prod-
ucts which are available only in relatively large villages and in cities are unlikely to help
isolated, poor farmers.
4. Farmers should be involved in the strategic planning of all biocontrol research
and development efforts. While they may contribute less to biological control than to
some other aspects of IPM, their participation in setting research goals and milestones
is important. Their participation in integrating various approaches, including biological
control, into existing production systems is key. Participation provides significant op-
portunities for grower education, consciousness-raising and feedback.
Biological Control Workshop-'91: Andrews et al.
5. Even though it is the most common model, the nonparticipatory research-par-
ticipatory implementation approach is not a viable method in the developing world
except for production of inputs to be marketed for profit. It has not worked in the past,
and there is no reason to believe that it will work in the future. Each of the other three
approaches has a legitimate role and is appropriate to certain expected outcomes.
6. Given the numbers of agrosocioeconomic situations and the limited resources of
biocontrol research and development people, if we choose to do conservation and manip-
ulation we had better look for novel research approaches. We must look for allies to
facilitate the farmer involvement which is central to the success of conservation and
manipulation efforts. Independent farmers, grassroots PVOs and community based
NGOs, especially grower organizations, are effective and willing allies. We can provide
them with missing information and play a supportive role for their local research efforts;
it is time that we give up the alternative idea that they support us in our research
Farmer participation is not a panacea (Bentley & Andrews 1991, Bentley in press
d). Farmer participation in biological control may not be as fruitful as it is in certain
other IPM tactics. It may in fact, not be necessary in efforts to carry out classical
biological control or certain kinds of inoculative releases. However, effective IPM pro-
grams based on conservation and manipulation of existing beneficial enemies are out of
the question unless client farmers are involved in the strategic planning, experimenta-
tion and implementation phases. To be effective collaborators, farmers need to be edu-
cated in the ideas that underlie biological control and believe in them. We are likely to
continue to self limit ourselves and make minimal impact among target groups until
biological control researchers and policy-makers: 1) spend significant efforts to educate
farmers in biocontrol and 2) understand and act on the fundamental difference between
techniques that do and do not require participation.
We thank Dres. Grace Goodell and Ana M. Andrews for ideas and references. We
appreciate anonymous reviewers' comments which improved the ms. One reviewer con-
firmed our thesis that this paper is necessary; no, it should not "be published in a
sociology journal"-it's for you. This work was supported in part by grants from
USAID/Honduras and ROCAP. DPV/EAP publication No. 340 and Florida Journal
ALTIERI, M. 1986. Bases ecol6gicas para el desarrollo de sistemas agricolas alter-
nativos para campesinos de Latinoambrica. CIRPON. Revista de Investigaci6n
ALTIERI, M. 1989. Agroecology: a new research and development paradigm for world
agriculture ecosystems and environment. 27: 37-46.
AMANOR, K. 1989. 340 Abstracts on Farmer Participatory Research. London: Over-
seas Development Institute. Agricultural Administration (Research and Exten-
sion) Network. Network Paper 5.
ANDREWS, K. L. 1988. Latin American research on Spodopterafrugiperda (Lepidopt-
era: Noctuidae). Florida Entomol. 71: 630-653.
ANDREWS, K. L. (in press). La investigaci6n agricola aplicada es participativa except
en el caso de los pequefios agricultores. CEIBA 31(2).
ANDREWS, K. L., AND J. W. BENTLEY. 1990. IPM and resource-poor Central Amer-
ican farmers. Global Pesticide Monitor 1: 1, 7-9.
ANDREWS, K. L., AND J. R. QUEZADA [eds.]. 1989. Manejo Integrado de Plagas
Insectiles en la Agricultura Escuela Agricola Panamericana. El Zamorano, Hon-
duras. 623 pp.
Florida Entomologist 75(4)
ARDON, M., R. SANCHEZ, C. SANCHEZ, AND M. MORA. in press. Participaci6n de
agricultores y tecnicos en un program de investigaci6n en Manejo Integrado de
Plagas en Repollo. CEIBA 31(2).
ASHBY, J. A. 1986. Methodology for the participation of small farmers in the design
of on-farm trials. Agricultural Administration 22: 1-19.
BENTLEY, J. W. 1989. What farmers don't know can't help them: the strengths and
weaknesses of indigenous technical knowledge in Honduras. Agriculture and
Human Values 6: 25-31.
BENTLEY, J. W. 1991. Conocimiento y experiments espontaneos de campesinos hon-
durefos sobre el maiz muerto. Revista Manejo Integrado de Plagas. 8: 16-26.
BENTLEY, J. W. 1992. Today there is no misery: The ethnography of farming in
Northwest Portugal. Tucson: University of Arizona Press. 169 pp.
BENTLEY, J. W. in press a. Facts, fantasies and failures of farmer participation: intro-
duction to the symposium. CEIBA 31(2).
BENTLEY, J. W. in press b. Etno-ecologia de abejas y avispas sociales en Honduras.
Proceedings Tercer Congreso MIP, Managua, Nicaragua.
BENTLEY, J. W. [ed.]. in press c. Proceedings of the Symposium "Participaci6n del
agricultor en investigaci6n y extension agricola". CEIBA 31(2).
BENTLEY, J. W., AND K. L. ANDREWS. 1991. Pests, peasants and publications: An-
thropological and entomological views of an integrated pest management pro-
gram for small-scale Honduran farmers. Human Organization 50: 113-124.
BENTLEY, J. W., AND W. MELARA. 1991. "Experimenting with Honduran Farmer-
Experimenters". ODI Agricultural Administration (Research and Extension)
Network. Newsletter 24: 31-48.
BIGGS, S. D., AND E. J. CLAY. 1981. "Sources of innovation in agricultural technol-
ogy." World Development 9(4): 321-336.
BROKENSHA, D. W., D. M. WARREN, AND Y. O. WERNER [eds.]. 1980. Indigenous
knowledge systems and development. University Press of America. Lanham,
MD. 466 pp.
BUNCH, R. 1982. Two ears of corn. World Neighbors. Oklahoma City, Oklahoma. 251
CHACON, J. C., AND S. R. GLIESSMAN. 1982. "Use of the 'non-weed' concept in tra-
ditional tropical agroecosystems of south-eastern Mexico." Agro-Ecosystems 8:
CHAMBERS, R., AND J. JIGGINS. 1987. Agricultural research for resource-poor farm-
ers Part II: A parsimonious paradigm. Ag. Admin. and Exten. 27: 102-128.
FARRINGTON, J., AND S. BIGGS. 1990. NGOs, agricultural technology and the rural
poor. Food Policy 15: 479-491.
FARRINGTON, J., AND A. MARTIN. 1987. Farmer participatory research: a review of
concepts and practices. London: Overseas Development Institute. Agricultural
Administration (Research and Extension) Network. Discussion Paper 19. Lon-
FRANCIS, C., J. KING, J. DEWITT, J. BUSHNELL, AND L. LUCAS. 1990. Participat-
ory strategies for information exchange. American J. of Alternative Agriculture
GLASS, E. H., AND H. D. THURSTON. 1978. Traditional and modern crop protection
in perspective. BioScience 28: 109-115.
GOODELL, G. 1984. Challenges to international pest management research and exten-
sion in the Third World: Do we really want it to work? Bulletin of the ESA
GOODELL, G., K. L. ANDREWS, AND J. LOPEZ. 1990. The contributions of agr6nomo-
anthropologists to on-farm research and extension in integrated pest manage-
ment. Agricultural Systems 32: 321-340.
HANSEN, M. 1987. Escape from the pesticide treadmill: Alternatives to pesticides in
developing countries. Institute for Consumer Policy Research Consumers Union.
Mount Vernon, N. Y. 185 pp.
HELLPAP, C. 1986. Nicaragua's revolution in IPM. ILEIA Newsletter 6: 14-16.
Biological Control Workshop-'91: Andrews et al. 439
HILDEBRAND, P. E. [ed.]. 1986. Perspectives on farming systems research and ex-
tension. Lynne Rienner Publishers. Boulder, Co. 167 pp.
HORTON, D. 1984. Social scientists in agricultural research: Lessons from the Man-
taro Valley Project, Perl. International Development Research Centre. Ottawa.
HORTON, D. 1986. Farming systems research: Twelve lessons from the Montaro Val-
ley Project. Agricultural Administration 23: 93-107.
HUANG, H. T., AND PEI YANG. 1987. The ancient cultured Citrus ant: A tropical ant
is used to control insect pests in southern China. BioScience 37(9): 665-671.
ILEIA NEWSLETTER. Information centre for Low External Input and Sustainable Ag-
riculture, Leusden Netherlands.
JOHNSON, A. W. 1972. "Individuality and experimentation in traditional agriculture".
Human Ecology 1(2): 149-159.
LITSINGER, J. A., E. C. PRICE, AND R. T. HERRERA. 1978. Small farmers' pest
control practices in rainfed rice, corn and grain legumes in three Philippine pro-
vinces. Philippine Entomology 4: 65-86.
LOCKERETZ, W. 1987. Establishing the proper role for on-farm research. American
J. of Alternative Agric. 2: 132-136.
MATLON, P., R. CANTRELL, D. KING, AND M. BENOIT-CATTIN [eds.]. 1988. Coming
full circle: Farmers' Participation in the Development of Technology. Interna-
tional Development Research Centre. Ottawa. 176 pp.
MATTESON, P. C., M. A. ALTIERI, AND W. G. GAGNE. 1984. Modification of small
farmer practices for better pest management. Ann. Rev. Entomol. 29: 383-402.
NATIONAL ACADEMY OF SCIENCES. 1977. Insect control in the People's Republic of
China. Report No. 2 Committee on Scholarly Communication with the People's
Republic of China. Washington, D. C. 218 pp.
QUEZADA, J. R. 1989. Utilizaci6n del control biol6gico clasico, pp. 195-210 in K. L.
Andrews and J. R. Quezada [eds.] Manejo Integrado de Plagas Insectiles en la
Agriculture. Escuela Agricola Panamericana. El Zamorano, Honduras.
REID, WALTER V., J. N. BARNES, AND B. BLACKWELDER. 1988. Bankrolling suc-
cesses: A portfolio of sustainable development projects. Environmental Policy
Institute and National Wildlife Federation. Washington, D.C.
RHOADES, R. E. 1987. Farmers and experimentation. Discussion Paper 21. Agricul-
tural Administration Unit. Overseas Development Institute. London.
RICHARDS, P. 1985. Indigeneous agricultural revolution. Westview Press. Boulder,
Co. 189 pp.
RICHARDS, P. 1989. Farmers also experiment: A neglected intellectual resource in
African science. Discovery and Innovation 1: 19-25.
RISCH, S. J. 1987. Indigenous knowledge systems and strategies of pest control. pp.
00-00 in S. Gliessman [ed.] Methods in agroecology. Springer Verlag.
ROGERS, E. M. 1983. Diffusion of innovations. The Free Press. New York, N.Y. 453
THRUPP, L. A. 1989. Legitimizing local knowledge: from displacement to empower-
ment for Third World people. Agriculture and Human Values 6: 13-24.
THURSTON, H. DAVID. 1990. Plant disease management practices of traditional farm-
ers. Plant Disease 74(2): 96-102.
UPAWANSA, G. K. 1989. Ancient methods for modern dilemmas. Ileia Newsletter 6: 9-
VAN DRIESCHE, R. G. 1989. How organic producers can make classical biological con-
trol work for them. American J. of Alternative Agriculture 4: 169-172.
YOUNGBERG, G., AND R. J. SAUER [eds.]. 1990. Proceedings of the symposium
"Learning from each other: new models for sustainable agriculture research and
information." American J. of Alternative Agriculture 5: 147-186.
Florida Entomologist 75(4)
OVERVIEW OF BIOLOGICAL CONTROL
OF HOMOPTEROUS PESTS IN THE CARIBBEAN
H. W. BROWNING
University of Florida, IFAS
Citrus and Education Center
700 Experiment Station Road
Lake Alfred, FL 33850
The potential for biological control of homopterous pests of agricultural crops in the
Caribbean region is supported by a rich history of successes and some biological and
ecological attributes which favor this approach to pest management. A review of the
discovery and importation of parasites and predators of citrus blackfly, Aleurocanthus
woglumi Ashby into Caribbean Basin countries and subsequent repeated utilization of
these natural enemies in newly invaded areas provides evidence that: 1.) Success with
a pest in one area is generally transferrable, particularly if cooperation can be estab-
lished; 2.) More than one natural enemy species can be utilized, and several species can
contribute to the ultimate success of the project; 3.) Natural enemies from different
regions are sometimes effective under varying conditions in the target areas of introduc-
tion; 4.) Successful suppression of a serious pest over a broad area can be obtained with
little initial investment via classical biological control; and 5.) No deleterious effects of
the successful biological control effort on citrus blackfly were identified, compared to
those associated with repeated use of chemical pesticides over a similar area of infesta-
tion. Other examples cited provide evidence for the continued emphasis on development
and implementation of biological control of the Homoptera.
El potential para el control biologico de homopteros plaga en la cuenca del Caribe
esta basado en exitos y tambien en la calidad ecologica de esta region, lo cual favorece
el enfoque del manejo integrado de plagas. Al revisar los resultados de los hallazgos y
la importacion de parasites y predadores de la mosca negra Aleurocanthus woglumi
Ashby en los paises del Caribe, y la consiguiente utilizacion de estos parasites y preda-
dores en regions recientemente invadidas por la mosca negra, podemos concluir que:
1. El exito en el control biologico de una plaga en una region puede ser transferido a
otras regions, si se establece una cooperation adecuada. 2. Se puede utilizar mas de
una especie de enemigo natural y varias species pueden a su vez contribuir a el exito
del project. 3. Los enemigos naturales provenientes de otras regions son algunas
veces efectivos bajo las condiciones de las areas de nueva introduction. 4. El exito de
la supresion de una plaga en una extensa region puede ser alcanzado mediante una poca
inversion inicial, empleando el control biologico clasico. 5. Comparado con el uso continue
de pesticides, no se observaron efectos negatives en este esfuerzo de control biologico
de la mosca negra. Se citan otros ejemplos para dar evidencia de que se debe continuar
con el desarrollo e implementation del control biologico del orden Homoptera.
The Caribbean Basin Advisory Group's (CBAG) support of Classical Biological Con-
trol is an indication of the current interest in this critical element of Integrated Pest
Management. Indeed, biological control is becoming an area of focus for agricultural
pest control worldwide. The historical success of biological control in the Caribbean has
been advanced through investment of funds, development of facilities, and support of
scientific expertise from a number of institutions, including CBAG. The Commonwealth
Biological Control Workshop-'91: Browning
Agricultural Bureau International Institute of Biological Control, (CAB IIBC, formerly
CIBC), Caribbean Agricultural Research and Development Institute (CARDI), Institut
National de la Recherche Agronomique (INRA), Escuela Agricola Panamericana El
Zamorano (EAP), University of West Indies, various U.S. State Agricultural Experi-
ment Stations, and other educational and research organizations should be credited with
facilitating these developments. It behooves the research community to continue its
investment in Caribbean projects having a good probability for success. Homoptera are
just such an example of a target group for classical biological control projects with a
high probability of success. Moreover, they have repeatedly demonstrated the potential
to become key pests in diverse agricultural settings worldwide, and have also shown
their vulnerability to biological control agents. Thus, the focus on Homoptera in the
CBAG workshop and the emphasis in future biological control efforts is undeniably
DeBach et al. (1970), pointed out that more than 2/3 of all successful examples of
classical biological control worldwide have involved homopterous pests, including over
80% of all which were deemed a complete success. In this review and many others,
coccids are extremely well represented among biological control attempts, and even
more prominent among those which achieved some level of success. More recent esti-
mates of the success of biological control of homopterous pests relative to other target
groups have been provided by other symposia authors (Rosen 1992).
Whether the reported rates of success with Homoptera are due primarily to the level
of effort expended or to other factors, compelling biological and ecological arguments
exist for the continued attention on their biological control. Some of the attributes of
the homopteran pests supporting this conclusion include:
1.) Homoptera attack a wide range of economic plant species, representing a broad
range of plant families (i.e. coffee, citrus, mango, avocado, papaya, coconut, guava,
pineapple, sugarcane, tomato and other vegetables);
2.) Geographic affinities among Caribbean islands and between these islands and
adjacent larger land masses (e.g. Republic of Trinidad and Tobago and northern
South America) promote common pest and biological control opportunities;
3.) Sessile habits of most Homoptera relative to other insect orders lead to
localized, definable populations which but for the small size of the insects are rela-
tively easy to study;
4.) General adaptation of Homoptera to subtropical and tropical climates allows for
year-around populations with mixed age structures on perennial and annual host
plants. These factors all lead to stable host plant, pest insect, and natural enemy
5.) Homopterous pests are continually spread across geographic areas through in-
festations on vegetative plant material, and the degree to which this material is
moved through commerce and other routes within or into the Caribbean is substan-
tial. This movement both within the Caribbean and worldwide continues the slow
spread of these pests, leading to cosmopolitan distribution for many of them. Thus,
the opportunity for shared research and, particularly, reciprocity in the exchange
of candidate natural enemies for classical biological control is enhanced.
In the Caribbean region, a large number of species occur for which classical biological
control has been attempted and for which some level of success has resulted. The review
of biological control in Commonwealth Caribbean and Bermuda discussed over 30
species of Homoptera for which some noticeable level of effort has been expended (Cock
1985). Similarly, Clausen's (1978) world review of introduced parasites and predators
highlights many biological control projects in the Caribbean and Bermuda, focusing on
natural enemies of Homoptera. Targets of classical biological control in the Caribbean
Florida Entomologist 75(4)
and Bermuda which are presented in these reviews or have arisen through recent
invasion into the region include:
Whitefly (Family Aleyrodidae)
Aleurocanthus woglumi Ashby, citrus blackfly
Aleurodicus cocois (Curt.), coconut whitefly
Metaleurodicus cardini (Back)
Mealybugs (Family Pseudococcidae)
Antonina graminis (Mask.), Rhodesgrass scale
Dysmicoccus boninsis (Kuway.), gray sugarcane mealybug
Dysmicoccus brevipes (Ckll.), pineapple mealybug
Nipaecoccus nippae (Mask.), coconut mealybug
Phenacoccus gossypii (Tns. & Ckll.), Mexican mealybug
Planococcus citri (Risso), citrus mealybug
Pseudococcus longispinus Targ., longtailed mealybug
Saccharicoccus sacchari (Ckll.), sugarcane mealybug
Armored Scales (Family Diaspididae)
Aonidiella aurantii Mask., California red scale
Aspidiotius destructor Sign., coconut scale
Carulaspis minima (Targ.), Bermuda cedar scale
Comstockiella sabalis (Comst.), palmetto scale
Insulapsis newsteadi (Sulc.), Bermuda cedar scale
Parlatoria ziziphi Lucas, black Parlatoria scale
Pseudaulacaspis pentagon (Targ.), oleander scale
Unarmored scales (superfamily Coccoidea)
Asterolecanium pustulans (Ckll.), oleander pit scale
Coccus viridis (Green), green scale
Icerya purchase Mask., cottony cushion scale
Orthezia insignis Browne, greenhouse orthezia
Orthezia praelonga (Dgl.)
Parasaisettia nigra (Nietn.), nigra scale
Protopulvinaria mangiferae (Green), mango shield scale
Pulvinaria psidii Mask., green shield scale
Vinsonia stellifera (Westw.). stellate scale
Aeneolamia spp., sugarcane froghopper
Saccharosydne saccharivora (Westw.), West Indian canefly
Siphaflava (Forbes), yellow sugarcane aphid
It is not the aim of this paper to review all of the biological control efforts on
Homoptera in this region, but rather it is to highlight some factors leading to successful
biological control and to identify some areas for continued research and implementation
over the next 5-10 years.
By way of an excellent example, it is useful to review the case with citrus blackfly,
Aleurocanthus woglumi, in the Caribbean and neighboring areas. The citrus blackfly
is an Old World species native to Asia which is relatively restricted in its host range,
utilizing primarily plants in the genus Citrus (Family Rutacea). When numbers on
citrus increase, populations spill over onto other hosts such as apple, coffee, mango,
and fig. Dowell & Steinberg (1990) and Dowell et al. (1981) investigated the relative
suitability of various host plants for citrus blackfly development.
The citrus blackfly lays its eggs in a spiral orientation. Immature stages settle and
feed on the underside of leaves. The movement of infested plant material provides the
primary mechanism by which this pest invades new areas.
Biological Control Workshop-'91: Browning
A. woglumi was first recorded from Jamaica and the Caribbean in 1913 (Edwards
1932). In Jamaica, it became widely distributed on island citrus, coffee, and other hosts
within one year of its detection. It was reported three years later in Cuba (Clausen &
Berry 1932). A USDA biological control effort against this pest was begun in Cuba in
1929, with exploration for exotic natural enemies focusing in Malaysia, Java, and
Sumatra. These efforts led to the introduction of Erotmocerus serious Silv., Encarsia
divergens (Silv.), and the coccinellids Catana clauseni Chapin and Scymnus smithianus
Clausen & Berry into Cuba. Encarsia smith (Silv.), and E. merceti Silv. were collected
but did not survive shipment. By 1933, E. serious was widespread and considered a
success. The parasite was then moved to Haiti and Panama in 1931. C. clauseni was
only temporarily established.
Jamaica: During this period, efforts to augment the local ant, Crematogaster brevis-
pinosa Mayr, were undertaken in Jamaica, as was inoculation of field populations with
the entomogenous fungus Aschersonia aleyrodis Webber (Edwards 1932). Some success
was reported. Introduction of the coccinellid Delphastis catalinae Horn was suggested,
but apparently was never attempted. In 1931, Erotmocerus serious was introduced via
Cuba, and by 1936 it was considered a success in Jamaica as well as in Cuba. This
parasite also was shipped to Costa Rica from Jamaica in 1933 to combat the blackfly.
Bahamas: Blackfly was present in the Bahamas before 1916, and its suppression by
E. serious in Cuba led to E. serious' introduction into the Bahamas in 1931. Establishment
followed but control was not achieved. Catana clauseni was introduced in 1936, but no
records of its establishment exist. By 1947, however, the blackfly was under control by
E. serious. A 1972 outbreak on Long Island resulted in the introduction of Encarsia
opulent (Silv.) from Barbados in the same year.
Mexico: In Mexico, occurrence of citrus blackfly date to 1935. E. serious was imported
from the Panama Canal zone in 1938, but failed to establish. Repeated shipments (12)
of parasites from the Canal Zone in 1943 and substantial distribution efforts in western
Mexico also were unsuccessful. Thus foreign exploration was initiated. Malaysia collec-
tions in 1948 led to introductions of Encarsia smith and E. divergens, but these were
unsuccessful. During 1949-50, foreign exploration obtained Encarsia clypealis, E.
opulenta, E. merceti, and Amitus hesperidum Silv. in western India and Pakistan.
These parasites were shipped to Mexico. A massive program of parasite distribution
ensued, since been unequalled, with over 1,600 people employed at the peak of the
program. During the period 1951-53, some 1.2 million E. clypealis, 2.8 million E.
opulenta, and 242 million A. hesperidum were released across large regions of Mexico
infested with blackfly (Smith & Maltby 1964). Because of the success of A. hesperidum
in Mexico, it was introduced into Ecuador in 1955, where it became established (Clausen
Blackfly was first recorded in Barbados in 1964. Erotmocerus serious was introduced
in 1964 from Jamaica, and Encarsia opulenta in 1965 from Mexico (Bennett 1965). Both
species were established, but E. opulenta dominated. Soon thereafter, blackfly popula-
tions were so low that it was difficult to field collect parasites (Pschorn-Walcher &
Blackfly was reported from the Cayman Islands in 1940's-50's, and two introductions
of Erotmocerus serious were attempted: one from Jamaica during this period and a
second through CIBC in 1966, but it was not detected in subsequent evaluations. How-
ever, in 1970, Encarsia opulent was found, indicating that it had been probably intro-
duced with Erotmocerus serious in 1966 (Bennett 1970, 1971).
During the 1950's-60's, blackfly was an occasional problem in Jamaica, and, thus,
Encarsia opulenta was introduced from Mexico in 1964. Once established in Jamaica,
E. opulent displaced Erotmocerus serious (Van Whervin 1968).
Other areas peripheral to the Caribbean have been invaded by the citrus blackfly.
Movement of the pest from Mexico into Texas occurred in the late 1970's, causing
Florida Entomologist 75(4)
serious damage to citrus (Summy et al. 1983). Amitus hesperidum and Encarsia
opulenta were introduced from Mexico and again proved successful in reducing pest
populations below damaging levels.
Florida, likewise, was invaded by the blackfly, with the initial introduction being
reported on the Keys in 1931. Eradication of this limited infestation was successful by
1937 (Newell & Brown 1939), but the insect again became established in Florida in 1976,
this time along the east coast. An enormous Federal-State eradication effort was under-
taken for an estimated $15 million but it failed. The introductions of A. hesperidum and
E. opulenta from Texas in 1976 successfully suppressed the pest (Dowell et al. 1981).
The most recent blackfly introduction in the Caribbean region is Puerto Rico in 1988.
In 1989, small numbers of Amitus hesperidum and Encarsia opulent were introduced
from Florida into the San Juan area (H. Browning, unpublished). By late 1990, even
though blackfly had spread to several locations on the island, the parasites reduced
populations to densities that made field recovery of blackfly or parasites difficult. Ef-
forts are ongoing to spread parasites to new areas of infestation across the Island.
The preceding review of the suppression of citrus blackfly by parasites supports
several guidelines regarding biological control of Homoptera in particular and perhaps
to biological control in general.
1. Success with a pest in one area generally indicates that control is transferrable
to another, particularly if coordination is established. Introduction of a parasite
complex has led to commercial control of this serious pest in many areas because of
2. More than one natural enemy species was utilized and contributed to the ulti-
3. Natural enemies from different regions were effective under varying conditions
in the target areas of introduction.
4. Successful suppression of a serious pest over a broad area can be obtained with
little initial investment via classical biological control. The cost to most of the suc-
cessful recipient areas was limited to the expense of shipping the material from a
5. No deleterious effects of the successful biological control effort against citrus
blackfly were identified, compared to those associated with repeated use of chemical
pesticides over a similar area of infestation.
Clausen (1978) and Cock (1985) provide detailed accounts of the biological control of
other homopterous pests in the Caribbean region, including those listed earlier. Impor-
tant additional points are illustrated by this broader review of cases involving the
biological control of Homoptera, including:
1. The importance of taxonomy to biological control is inestimable. In addition to
natural enemy taxonomy, pest and even host plant taxonomic support can mean the
difference between success and failure. This expertise should be enlisted at the
outset of a biological control project.
2. Many successful cases of biological control in the Caribbean involving Homoptera
have included importation or augmentation with predators, particularly coccinellids.
Obrycki (this publication) discusses evaluation techniques for these natural enemies.
Evaluation of the predator impact will lead to a better understanding of the role
that they play in regulating homopterous populations.
3. The Caribbean region has a rich reserve of natural enemies which is available
for export to cooperating scientists in other parts of the world. Reciprocal arrange-
ments could result, and lead to less expensive and more efficient foreign exploration
for the natural enemies to combat other Caribbean pests.
Biological Control Workshop-'91: Browning
4. New pest species continue to invade the Caribbean region, and established pests
continue to spread within this area. Many additional examples of partial or complete
biological control successes could be achieved through the development and expan-
sion of cooperative interactions.
One of the primary objectives of the CBAG Biological Control Workshop was to
identify additional opportunities for biological control in the Caribbean region. Some
potential areas for such cooperation in the biological control of Homoptera are:
1. Evaluations of existing natural enemies, focusing on both native and introduced
species and the role they play in pest population dynamics. A tremendous need
exists to document the outcome of previous natural enemy introductions, because
there has been a dearth of information in many introduction projects. These en-
deavors should: a) identify target niches for additional natural enemy introductions,
augmentations, or conservation; b) reduce redundant introductions; and c) assist in
the planning of future foreign exploration and importations.
2. Plan for foreign exploration to enhance the biological control of existing pests
and to develop strategies for new invaders or for species expanding their distribution
within the Caribbean region.
3. Coordinate efforts and scientists to increase reciprocity among and between
areas, in order to enhance success rates.
4. Provide taxonomic support to assist in biological control projects. A faunistic
study of the natural enemies of Homoptera with a taxonomic component from the
outset would be highly productive.
Biological control of Homoptera in the Caribbean region must be considered a prior-
ity for future development, and can build on a rich history of success. Native and newly
introduced species of Homoptera will continue to plague agriculture in this area, and
biological control offers a permanent, inexpensive, and environmentally sound approach
for their management.
The support of the Caribbean Basin Advisory Group for the presentation of this
paper at the 1991 Biological Control Workshop is appreciated. Florida Agricultural
Experiment Station Journal Series No. R-02766.
BENNETT, F. D. 1965. Report on a visit to Barbados, June 28-July 1, 1965. CIBC
Unpub. Rep., 3 pp.
BENNETT, F. D. 1970. Report on a visit to Grand Cayman, B.W.I., to assess the
possibilities of biological control June 29-July 4, 1970. CIBC Unpub. Rep., 6 pp.
BENNETT, F. D. 1971. Some recent successes in the field of biological control in the
West Indies. Revista peruana de entomologia agricola. 14: 369-73.
CLAUSEN, C. P. [ed.] 1978. Introduced parasites and predators of arthropod pests and
weeds: A world review. Agric. Handbook 480. U.S. Dept. Agric., Washington,
D.C. 545 pp.
CLAUSEN, C. P., AND P. A. BERRY. 1932. The citrus blackfly in tropical Asia and
the importation of its natural enemies into tropical America. USDA, Tech. Bull.
320, 58 pp.
COCK, M. W. J., [ed.] 1985. A review of biological control of pests in the commonwealth
caribbean and Bermuda up to 1982. Commonwealth Inst. Bio. Contr., Tech.
Comm. No. 9, Commonwealth Agric. Bur., London. 218 pp.
Florida Entomologist 75(4)
DEBACH, P., D. ROSEN, AND C. E. KENNETT. 1970. Biological control of coccids by
introduced natural enemies, Chapter 7 in Biological Control, C. B. Huffaker
[ed.], Plenum Press, New York. 511 pp.
DOWELL, R. V., AND B. STEINBERG. 1990. Influence of host plant characteristics and
nitrogen fertilization on development and survival of immature citrus blackfly,
Aleurocanthus woglumi (ashby) (Homoptera: Aleyrodidae). J. Applic. Entomol.
DOWELL, R. V., R. H. CHERRY, G. E. FITZPATRICK, J. A. REINERT, AND J. L.
KNAPP. 1981. Biology, plant-insect relations, and control of citrus blackfly.
Florida Agric. Exper. Stn. Tech. Bull. 818: 49 pp.
EDWARDS, W. H. 1932. Importation into Jamaica of a parasite (Erotmocerus serious
Silv.) of the citrus blackfly (Aleurocanthus woglumi Ashby). Entomol. Bull. 6,
Dept. Agric. Jamaica. 12 pp.
NEWELL, W., AND A. C. BROWN. 1939. Eradication of the citrus blackfly in Key
West, Florida. J. Econ. Entomol. 32: 680-682.
PSCHORN-WALCHER, H., AND F. D. BENNETT. The successful biological control of
the citrus blackfly (Aleurocanthus woglumi (Ashby)) in Barbados, West Indies.
Pest Articles and News Summaries (PANS) 13: 375-84.
ROSEN, D. 1992. ??.
SMITH, H. D., AND H. L. MALTBY. 1964. Biological control of citrus blackfly in
Mexico.USDA Tech. Bull. 1311. 30 pp.
SUMMY, K. R., F. E. GILSTRAP, W. G. HART, J. M. CABALLERO, AND I. SAENZ.
1983. Biological control of citrus blackfly in Texas. Environ. Entomol. 12: 782-
VAN WHERVIN, L. W. 1968. The introduction of Prospaltella opulenta Silvestri into
Jamaica and its competitive displacement of Erotmocerus serious Silvestri. Pest
Articles and News Summaries (PANS). 14: 456-64.
APPROACHES TO THE BIOLOGICAL CONTROL
Department of Zoology,
The George S. Wise Faculty of Life Sciences,
Tel Aviv University, Ramat Aviv 69978, Israel
Biological control has been successful for some whitefly species but not for others.
Natural enemies available for control of whiteflies include fungi, predators and
parasitoids. Each of these groups has an important place in the ecosystem, but knowl-
edge of their biology and utilization is limited and available mainly for parasitoids.
There is no proven method for predicting the success of natural enemies. However, the
number of host stages that are not vulnerable to enemy attack, as well as the host
refuges, should be reduced in both time and space. This can be accomplished by integ-
rating the use of different natural enemies, plant resistance and selective insecticides.
Comparative studies of successful cases of whitefly biocontrol are suggested as a means
to increase our knowledge of the necessary attributes of successful natural enemies.
Biological Control Workshop-'91: Gerling 447
El control biologico de moscas blancas ha resultado exitoso para algunas species y
no para otras. Los enemigos naturales que se encuentran a disposicion para dicho control
son hongos, depredadores y parasitoideos. Cada uno de estos grupos ocupa un lugar
important en el ecosistema, pero el conocimiento de su biologia y modo de utilizacion
es limitado, y existe mayormente para los parasitoideos. Parte de los problems actuales
de plagas se tipifican por extensas areas de esparcion de las moscas blancas, alrededor
de las cuales se forman los complejos de enemigos locales. No hay un metodo probado
para la estimacion a-priori del exito de un enemigo natural; la eficacia de un enemigo
tal debe ser verificada en el campo. No obstante se sugiere que la selection del enemigo
sea guiada por la necesidad de reducir el numero de estados hospederos que no son
vulnerable al ataque del enemigo y de reducir los refugios del hospedero en el tiempo
y el espacio. Esto puede realizarse integrando el uso de diferentes enemigos naturales,
de la resistencia de las plants y de insecticides selectivos. Se sugieren studios com-
parativos de casos exitosos en control biologico de moscas blancas, como medio para
aprender mas acerca de los atributos necesarios para enemigos naturales exitosos.
Biological control of arthropods, the utilization of organisms (or natural enemies) to
control arthropod pests, has been practiced for over 100 years. During the last decade
notable contributions have been made to the theory and practice of such control. Critical
issues dealing with insect behavior, competition, diversity, and aggregation, as well as
models and their relevance to biological control have been examined (Waage & Great-
head 1986, Mackauer et al. 1990). Likewise, much information about whitefly pest
status, ecology, and biological control agents has been reviewed (Gerling 1990a). Draw-
ing upon this body of knowledge, I shall present some of the approaches to biological
control of whiteflies.
For practical purposes, biological control has been divided into two categories: In-
oculative (or classical), in which the natural enemy is usually introduced only once and
is then expected to continue to multiply and control the pest; and inundative, in which
the natural enemy is seen as a "biological insecticide", may be introduced repeatedly,
and is primarily aimed at producing immediate pest suppression. Recently, van Lente-
ren (1986) employed the term "seasonal inoculative release method" for repeated re-
leases with a long-term effect.
Whiteflies have been the subject of biological control attempts since early this cen-
tury; first using the classical approach, and later also through the inundative or seasonal
inoculative methods. Some attempts have been successful (Clausen 1978, van Lenteren
& Woets 1988), whereas others (Gerling 1986) have failed, with the species involved
causing severe economic damage (Byrne et al. 1990). In addition, formerly successful
biological control is now being challenged due to the need (or apparent need) for insec-
ticidal treatments against other pests (Dowell 1990).
The proper utilization of natural enemies depends upon an adequate understanding
of whitefly and enemy taxonomy, ecology and behavior. While ecological and behavioral
traits will be discussed here, taxonomy is beyond the scope of this review; however, its
importance should not be underestimated. The taxonomy of both whiteflies (Bink-
Moenen & Mound 1990) and their enemies (Viggiani & Mazzone 1979, DeBach & Rose
1982, Hayat 1989) is currently in a state of flux. Consequently, accurate determinations
are not always available.
THE WHITEFLY-ENEMY COMPLEX
Whiteflies are homopterans belonging to the suborder Sternorrhyncha. Their life
cycle includes an egg, four nymphal instars, a so-called pupal instar, and winged adults.
448 Florida Entomologist 75(4) December, 1992
All of the immature stages except for the young first instar "crawler" are sessile.
Whiteflies usually have from 2 to 7 generations per year, most of which occur during
the summer, since they are mainly warm climate insects (Bink-Moenen & Mound 1990).
In addition to direct plant damage caused by sucking its sap, whiteflies also cause
damage through honeydew production and virus transmission.
Because of the relatively long period which they spend as sessile insects, whiteflies
are often considered suitable objects for biological control. Their known natural enemies
include fungal pathogens, and arthropod predators and parasitoids.
Fungi occur principally in climates with high relative humidity. They usually attack
the host nymphs, but some species (e.g. Paecilomycesfumosoroseus) may attack all of
the stages (Fransen 1990). Host specificity varies; the genus Aschersonia is specific to
whiteflies whereas Verticillium, Poecilomyces and Aegerita are not.
Fransen (1990) reviewed the known species of Aschersonia and other whitefly-infect-
ing fungi and their host association. When they are grouped by host ranges (Fig. 1),
two extremes become evident. Twenty one species of whiteflies have only one recorded
fungus species attacking them, while 15 species of fungi are known from a single whitefly
host. On the other hand, a few well studied species have extensive host-pathogen asso-
ciations (T. vaporariorum and Dialeurodes citri are recorded as being attacked by 16
and 10 fungus spp. respectively and Aschersonia aleyrodis is recorded from 10 whitefly
species). Such relationships often indicate insufficient study of some of the organisms.
This is supported by the fact that, for the 1200 known species of whiteflies (Mound &
Halsey 1978), only 37 species of fungi have been described.
Fungi can be utilized by transferring infected branches from one plant to the next,
or by the more modern methods of mass producing fungal spores and applying them
with commercial spray equipment. To date, this has been limited to V. lecanii, where
strain selection and improvement have produced a marketable commercial product
(Ravensberg et al. 1990). Research leading to the commercial utilization of P.
fumosoroseus is in progress (Osborne et al. 1990).
Predators belong mainly to four insect orders: Heteroptera, Neuroptera, Diptera,
and Coleoptera, and to the mites (Gerling 1990b). They have not often been credited
with whitefly control, and their role has been little investigated. At present, in light of
recent field and laboratory observations, the role of indigenous predators is receiving
new attention and their potential as controlling agents is being reassessed (Alomar et
From the few studies that have been conducted, it is possible to outline characteris-
tics that may serve as guidelines for future attempts at biological control using pre-
1. Mobility: The adults are mobile and may travel considerable distances. Mobility of
immatures ranges from very limited (Acletoxenus) (Diptera:Drosophilidae) (Clausen &
Berry 1932) to extensive (Heteroptera) (Alomar et al. 1990).
2. Physiological characteristics: Mating and oviposition are often associated with com-
plex behaviors such as premating flights (Chrysoperla) (Duelli 1980) or ovipositional
diapause (Coccinellids) (Hagen 1962). Adults of some groups pass unfavorable periods
(adverse climate or absence of food) in diapause (Hagen et al. 1976), while others mig-
3. Prey relations: The prey range extends from oligophagy [coccinellids that feed
mainly on whiteflies, (Clausen & Berry 1932)] to polyphagy [some mirids feed on plants
and insects (Alomar et al. 1990)]. Some species are cannibalistic as larvae and/or adults,
and others prey upon parasitized whiteflies. Predators may change their dietary prefer-
ences or habits in response to changes in prey availability (Lawton et al. 1974). In the
case of whitefly predators, we found that Chrysoperla carnea which previously fed on
Biological Control Workshop-'91: Gerling 449
fungus spp .0
a=T. vaporariorum 4
c=Aschersonia aleyrodis 1 1 1 1 2 5 15 21
No. of cases
Fig. 1. Frequency distribution of whitefly-fungus host ranges (Data from Fransen
1990). High numbers of whitefly species attacked by the same fungus sp., or high
numbers of fungus species associated with a single whitefly species are relatively rare
(left side of graph). Conversely, numerous cases exist with a one whitefly-one fungus
association (right side of graph).
*Numbers of fungus or whitefly species per associated whitefly or fungus species.
non-whitefly prey because of its relative availability, shifted to include whiteflies in
their diet once these became abundant (Or 1986, Gerling personal observations).
These points indicate that predators constitute a very versatile group, and once
their particular characteristics are known, different species can be utilized under a
multitude of environmental conditions. In particular, predators may be efficient at high
prey densities, both because of adult requirements (Hagen et al. 1976), and because
their immatures can maximize their potential for development within a limited area.
Indeed, several species of predators have been effective as controlling agents, particu-
larly at high whitefly densities (Clausen 1978, Waterhouse & Norris 1989).
Predators have, so far, been used to control whiteflies especially in perennial ecolog-
ical systems (Clausen 1978, Waterhouse & Norris 1989). However, due to their mobility
and relative polyphagy they may also be useful in temporary agricultural systems, such
as annual vegetables and field crops, where an unstable, changing environment is not
as favorable for the establishment of specific parasitoids (Alomar et al. 1990, Dowell
1990). Consequently, their introduction and encouragement may contribute measurably
to manipulative (inundative and short-term inoculative) control. However, as pointed
out previously, the utilization of particular organisms requires running a considerable
number of specific tests (see Alomar et al. 1990 for a list) in order to establish their
characteristics in relation to the control requirements.
Parasitoids are credited with most of the success in the biocontrol of whiteflies.
Exploration for parasitoids has been conducted either in the presumed endemic range
of the host (Clausen & Berry 1932, Clausen 1978, Onillon 1990, Waterhouse & Norris
1989), or in locations where previous success was achieved (Clausen 1978, Onillon 1990).
A notable exception is the fortuitous control of Parabemisia myricae by Eretmocerus
sp. in California (Rose & DeBach 1982) where indigenous parasitoids provided excellent
450 Florida Entomologist 75(4) December, 1992
Recent reviews of whitefly parasitoids (Gerling 1990b, Onillon 1990, Waterhouse &
Norris 1989) were examined in order to rate the importance of five host/parasitoid traits
to the success of biological control programs:
1) Host instar attacked. 2) Plant-host range of the whitefly. 3) Mode of development
(arrhenotoky, autoparasitism). 4) Reproductive physiology (synovigenic vs.
proovigenic). 5) Host range of the parasitoid.
Of the above, success was not related to the first three criteria, whereas proovigenic
Amitus were particularly successful under high host densities.
The importance of the fifth trait, host-range, could not be rated unequivocally. Most
successful whitefly control was achieved by parasitoids with a limited range that in-
cluded, at most, a few congeneric species that might have evolved with their hosts in
the same region. However, excellent success in greenhouse whitefly control is attained
with Encarsia formosa that has a wide host range. Moreover, some whitefly species
(e.g. Bemisia tabaci and Trialeurodes vaporariorum) have a very wide geographical
range and are often attacked by different species of parasitoids which presumably have
moved on to them from local whitefly spp. (Gerling 1986); so much so that looking for
an originally evolved "narrow" association becomes meaningless. Such an increase in
parasitoid host range can yield excellent control, as has been shown by the fortuitous
biological control of Parabemisia myricae by Eretmocerus sp. in California (Rose &
DeBach 1982). Consequently host specificity in itself cannot be taken as a measure in
determining the chances for success of the parasitoids.
B. WHITEFLY-ENEMY INTERACTIONS
As pointed out by Price (1984), a population that exists at a low level may reach an
outbreak phase due either to the creation of an ecological advantage (e.g. increased
reproductive rates), or to the relaxation of an ecological constraint that has been intro-
duced into the population.
In many of the whitefly outbreaks that were controlled biologically, the ecological
constraints that had been relaxed, and which were re-established by man were the
natural enemies. Often multiple introductions of enemies were carried out (Clausen
1978, Onillon 1990) and one or more of these became established and reduced the
whitefly populations sufficiently. However, the outbreak-inducing ecological conditions
may also be climatic (man-made or natural), or due to the availability of new crops as
host plants, or the extension of the agricultural season through the introduction of
protected plants, or the use of agrochemicals. Some of these practices and conditions,
such as the use of insecticides, can and should be changed in order to facilitate biological
control, whereas others, such as the type of crop or the use of greenhouses, can not.
Many of the current acute whitefly outbreaks do not involve the transfer of a whitefly
species from a place in which it is under satisfactory natural control to a location where
its enemies are lacking. Rather, other natural or man-made ecological conditions can
usually be blamed for the outbreaks. Such outbreaks, e.g. of B. tabaci or T. vap-
orariorum may reach monumental proportions. The pests are often widely distributed
and their economic damage varies in accordance with the crop grown, the agricultural
practice, the geographic locality, and the duration of the species' residence in the area.
These invading whitefly populations may show considerable genetic variability and
an increasing host range-features that often accompany an out-break phase (Price
1984). The natural enemy complex of these whiteflies usually consists of some parasitoid
species that apparently migrate with them through very extensive ranges (e.g. Eret-
mocerus mundus is recorded from Spain in the west to Pakistan in the east, and from
Africa in the south to Turkemanistan in the north, without any intended human assist-
Biological Control Workshop-'91: Gerling 451
ance), and of other species that moved on to the pest from the local fauna (Gerling 1986).
In addition, local predators may adapt gradually to the newly abundant food source.
The efficiency of controlling the whitefly population by each of these local enemy
complexes is usually quite specific, incorporating, in addition to human activities, the
nature of the pest outbreak, and the attributes of the enemies and climate. At times,
situations occur in which whitefly control has been achieved by indigenous enemies (e.g.
Rose & DeBach 1982). Thus, both, endemic and introduced whitefly species may be
controlled by either locally occurring or introduced natural enemies (e.g. Dowell et al.
1979, Onillon 1990). Consequently, almost any place in which host-enemy associations
of whiteflies take place may be considered as a potential source of natural enemies to
be used for biological control. However, it should be kept in mind that since uncertain-
ties exist in the classification of both hosts and parasitoids, not all of the apparent
specific associations may prove to involve the host species that we are seeking.
APPROACHES TO BIOLOGICAL CONTROL
1. The Host-refuge Consideration
Murdoch (1990) in his suggestions for maintaining low host populations and fluctua-
tions, stressed the need for elimination of host refuges. These can take the form of: a)
Invulnerable host stages, b) refuges in time or c) refuges in space.
a. Differential vulnerability of insect stages to natural enemies is a common occur-
rence. In whiteflies, it is expressed through preference of many predators to feed on
eggs and young larvae which are easier to overcome and/or differ in their nutritive value
(Gerling 1990b). Likewise, parasitoids prefer to attack certain nymphal host instars
(e.g. many Encarsia prefer instars 3 and 4 and seldom attack the first two instars,
whereas Amitus may attack instars 1 and 2 and leave eggs and older instars unaffected
b. Temporal refuges often occur where pests become established in a new area or
crop, before natural enemies reach the new location in sufficient numbers. With whitef-
lies, this is especially acute in annual crops and greenhouses to which B. tabaci and T.
vaporariorum migrate each season and where their natural enemies must follow (Gerl-
ing et al. 1980, van Lenteren & Woets 1988).
c. Spatial refuges may take numerous forms. They may be associated with plant
architecture, as for example, when whiteflies cover certain parts of the plant whereas
the enemies search on others. This has been suggested as a factor limiting the effective-
ness of Encarsia formosa on Gerbera plants infested by the greenhouse whitefly (van
Lenteren pers. comm.). Climatic and other regional characteristics may also furnish
hosts with refuges from enemies, and finally, host density itself may act in that capacity.
Above a certain host density, parasitization may act in an inverse density dependent
manner because, 1. The physical environment on a heavily infested leaf is not conducive
to parasitoid activity due to the copious amounts of honeydew and exuviae present (van
Lenteren pers. comm., Gerling pers. obs.); and 2. There is no evidence for long-distance
attraction of parasitoids to infested leaves (Gerling 1990b). Thus, less prasitoids per
whitefly would arrive on densely infested leaves than on sparsely infested leaves. In-
verse density dependence would follow since the daily rate of oviposition by parasitoids
cannot exceed a fixed, species specific, level (Vet & van Lenteren 1981, Shimron 1991).
Arrestment alone cannot compensate for this shortcoming.
The practical field situations usually involve more than one kind of refuge and may
require unique solutions. The introduction of multiple species of natural enemies, each
having a somewhat different host preference and developmental niche, the use of
"banker plants" (Osborne et al. 1990), and the development of whitefly-resistant plants
are usually the most profitable approaches to reduce these refuges.
Florida Entomologist 75(4)
2. Plant Resistance
The fact that some plant species or varieties are more hospitable to insects than
others can be used to reduce pest refuge areas. Although complete resistance is rare,
partial resistance is common (DePonti et al. 1990). Differential susceptibility to whitefly
attack was demonstrated at the specific and varietal levels (DePonti et al. 1990, van
Lenteren & Noldus 1990). Thus, certain plant species or varieties will support only low
whitefly populations; this may suffice as a control method in itself, or at least facilitate
parasitoid activity by reducing the detrimental effects of high whitefly density.
To achieve maximum benefits, and to overcome some of the host-refuge problems,
biological control of whiteflies must be integrated within itself and with other control
methods. A strategy that combines different natural enemies, predators (especially
mites, coccinellids and Heteroptera) (Avilla et al. 1990, Gerling 1990b) and fungi with
parasitoids, can deal with most host stages and habitats in a complementary fashion.
However, the compatibility and complementary action of the different organisms must
be examined in order that they may be employed most effectively. For example,
Waterhouse and Norris (1989) reported that in the campaign against Aleurodicus dis-
persus the predator Nephaspis oculatus was effective at high host densities whereas
the parasitoids of the genus Encarsia were effective at low densities. Smith et al. (1964)
introduced several species of parasitoids against Aleurocanthus woglumi and each be-
came established in a different climatic zone. Fransen (1990) discussed the interactions
and utilization of fungi together with parasitoids in controlling whiteflies. Whereas
broad spectrum fungi like P. fumosoroseus and Beauveria bassiana may infect
parasitoids and predators, the integration of the whitefly specific Aschersonia aleyrodis
and Aschersonia 'Cuba red' with parasitoids gave better control than the use of
The use of natural enemies with insecticides can often be achieved through the
choice of the right materials and proper timing, as has been elaborated by Dowell
(1990). He concluded that the improvement of biological control of whiteflies, depends
upon the development of new control measures for pests currently under chemical
control. Recently, a relatively new group of insecticides, the Insect Growth Regulators
(IGRs) have come into use. These include molting inhibitors and chitin synthesis in-
hibitors. Their action is selective so that their use in whitefly management may be
valuable. Laboratory experiments with Buprofezin, a chitin synthesis inhibitor, con-
ducted with E. formosa and the greenhouse whitefly have shown that while the pest
was affected the development of the parasitoid and percent parasitization were not
(Garrido et al. 1984, Wilson & Anema 1988, van de Veire & Vacante 1989). Garrido et
al. (1984) have also shown that Cales noacki and its host Aleurothrixus floccosus re-
sponded similarly. Laboratory experiments in Israel demonstrated that Buprofezin is
effective mainly against first and second instar B. tabaci nymphs but that it also had
some detrimental effects on young larvae of Eretmocerus sp. and on pupae of Encarsia
desert (Sinai 1990).
Complementary field experiments with Buprofezin were conducted in cotton fields
for three years. Since the IGR affects mainly whitefly instars that are too young to
serve as hosts for parasitoids, the trials were run with the expectation that parasitoid/
host ratio (and consequently, percentage parasitization) would be higher in the treated
field than in the corresponding untreated plots. However, although, they resulted in
Biological Control Workshop-'91: Gerling
good whitefly control, only an occasional and limited increase in percent parasitism
occurred (Sinai 1990, Gerling, pers. obs.). These results might be partially explained
by the parasitoid mortality that has been demonstrated in the lab, and partially by other
factors, such as predators, whose activity had not been adequately monitored. However,
the overall success of control without decline in percent parasitism is indicative of the
potential of IGRs in whitefly pest management.
A LOOK TO THE FUTURE
One of the main limitations to cheap and efficient biological control is our inability
to preselect natural enemies and determine their future qualities as controlling agents
prior to their field utilization. Recent studies on host-enemy interaction, parasitoid
behavior and developmental strategies (e.g. van Alphen & Vet 1986, Waage & Godfray
1984, Walde & Murdoch 1988) as well as studies of practical natural enemy implementa-
tion (Dowell et al. 1979, van Lenteren 1986, van Lenteren & Woets 1988), explored in
depth desirable attributes of natural enemies. Such studies also led to the production
of a flow chart for natural enemy selection (van Lenteren 1986).
However, the guidelines provided in the flow chart are mainly relevant for
greenhouses, where the artificial conditions require continual interference and provide
relatively stable and predictable environmental conditions. In inoculative, out-of-door
control, where the relative involvement per crop plant is lower, and the risks of chang-
ing and/or unpredictable environmental conditions may be high, it is still simpler and
cheaper to utilize the "introduce and see what happens" method than to run a series of
tests on the host and its natural enemies. Moreover, the complexity of the ecological
conditions that may influence the efficiency of a natural enemy, hinder the formation of
a simple and effective method to predict the quality of a natural enemy that will be used
for whitefly control out-of-doors.
In the absence of a general testing method to estimate the quality of a natural
enemy, the role of guesswork can be reduced by studying the dynamics of whitefly-
enemy interactions and by learning from previous successes (and failures) in the biolog-
ical control of whiteflies. Such studies should be conducted in both the area of the pest's
origin and in the country in which the control took place. These should elucidate the
role of the physical and trophic components in the ecological system, and may shed light
on factors that facilitated the natural enemy's success. A step in this direction was taken
by Southwood & Reader (1976) when they studied the population dynamics of the
Aleurotrachelus (Aleurotuba) jelinekii, at times a pest in Britain.
It seems only natural that in present times, when man has learned to change natural
processes through genetic engineering, and to manipulate biological organisms in ways
that would have been unthinkable only a short time ago, extensive efforts should be
expanded towards the manipulation and improvement of natural enemies. The goal is
to obtain a natural enemy that we can recognize as being better adapted to control its
host under specified conditions, and that can be used with considerable chances for
success. At the same time, it should be resistant to insecticides that may be applied in
the agroecosystem. It is also necessary to improve the techniques of rearing, mass
producing, storing, and releasing enemies.
Many of the abovementioned desiderata have been considered in detail (van Lente-
ren 1986, Mackauer et al. 1990). Unfortunately, only limited successes have been ob-
tained so far, especially in the field of insecticide resistance and artificial rearing (Roush
1990). However, the pace at which science is advancing gives hope that much more can
be achieved if we, as consumers of natural enemy 'high tech', will make our priorities
better known and venture to experiment using the newly discovered techniques.
454 Florida Entomologist 75(4) December, 1992
I wish to thank Dr. Y. Rossler and Prof. A. Hefetz for the critical reading of the
manuscript, and the anonymous reviewers for numerous helpful comments.
ALOMAR, 0., C. CASTANE, R. CABARRA, AND R. ABLAJES. 1990. Mirid bugs -
another strategy for IPM on Mediterranean vegetable crops? IOBC/WPRS Bul-
letin 13(5): 6-9.
ALPHEN, J. J. M. VAN, AND L. E. M. VET. 1986. An evolutionary approach to host
finding and selection, pp. 23-61 in J. K. Waage and D. J. Greathead [eds.], Insect
parasitoids. Academic Press, London.
AVILLA, J., M. ANTIQUES, M. J. SARAUSA, AND R. ABLAJES. 1990. A review of the
biological control characteristics of Encarsia tricolor and their implication for
biological control. IOBC/WPRS Bulletin 13(5): 14-18.
BINK-MOENEN, R., AND L. A. MOUND. 1990. Whiteflies: diversity, biosystematics
and evolutionary patterns, pp. 1-11 in D. Gerling [ed.], Whiteflies: their
bionomics, pest status and management. Intercept, Andover, UK.
BYRNE, D. S., T. S. BELLOWS, AND M. P. PARELLA. 1990. Whiteflies as crop pests,
pp. 147-186 in D. Gerling [ed.], Whiteflies: their bionomics, pest status and man-
agement. Intercept, Andover, UK.
CLAUSEN, C. P. [ed.]. 1978. Introduced parasites and predators of arthropod pests
and weeds: a world review. USDA Agric. Handbook no. 480, Washington, DC.
CLAUSEN, C. P., AND P. A. BERRY. 1932. The citrus blackfly in Asia, and the impor-
tation of its natural enemies into tropical America. Technical Bulletin No. 320.
U.S. Department of Agriculture, Washington, DC.
DEBACH, P., AND M. ROSE. 1981. A new genus and species of Aphelinidae with some
synonimies, a rediagnosis of Aspidiotiphagus and a key to pertamerous and
heteromerous Prospaltellinae (Hymenoptera, Chalcidoidea, Aphelinidae). Proc.
Ent. Soc. Washington 83: 658-674.
DEPONTI, O. M. B., L. ROMANOW, AND M. J. BERLINGER. 1990. Whitefly-plant
relationships: plant resistance, pp. 91-106 in D. Gerling [ed.], Whiteflies: their
bionomics, pest status and management. Intercept, Andover, UK.
DOWELL, R. V. 1990. Integrating biological control of whiteflies into crop manage-
ment systems, pp. 315-336 in D. Gerling [ed.], Whiteflies, their bionomics, pest
status and management. Intercept, Andover, UK.
DOWELL, R. V., G. E. FITZPATRICK, AND J. A. REINERT. 1979. Biological control
of citrus blackfly in southern Florida. Env. Ent. 8: 595-597.
DUELLI, P. 1980. Preovipository migration flights in the green lacewing Chrysopa
carnea (Plannipenia Chrysopidae). Behav. Ecol. Sociobiol. 7: 239-246.
FRANSEN, J. J. 1990. Natural enemies of whiteflies: Fungi. pp. 187-210 in D. Gerling
[ed.], Whiteflies, their bionomics, pest status and management. Intercept, An-
GARRIDO, A., F. BEITIA, AND P. GRUENHOLZ. 1984. Effects of PP618 on immature
stages of Encarsia formosa and Cales noacki (Hymenoptera: Aphelinidae). pp.
305-311 in British Crop Protection Conference Pests and Diseases, Brighton,
GERLING, D. 1986. Natural enemies of Bemisia tabaci, biological characteristics and
potential a biological control agents: A review. Agric. Ecosyst. Environ. 17: 99-
GERLING, D. [ed.]. 1990a. Whiteflies: their bionomics, pest status and management.
Intercept, Andover UK.
GERLING, D. 1990b. Natural enemies of whiteflies: predators and parasitoids, pp. 147-
185 in D. Gerling [ed.], Whiteflies: their bionomics, pest status and management.
Intercept, Andover, UK.
GERLING, D., U. MOTRO, AND H. R. HOROWITZ. 1980. Dynamics of Bemisia tabaci
(Gennadius) (Homoptera: Aleyrodidae) attacking cotton in the coastal plain of
Israel. Bul. Ent. Res. 70: 213-219.
Biological Control Workshop-'91: Gerling
HAGEN, K. S. 1962. Biology and ecology of predaceous coccinellidae. Annu. Rev. Ent.
HAGEN, K. S., S. BOMBOSCH, AND J. A. MCMURTRY. 1976. The biology and impact
of predators, pp. 93-142 in C. B. Huffaker and S. Messenger [eds.], Theory and
practice of biological control. Academic Press, N.Y.
HAYAT, M. 1989. Taxonomic notes on Indian Encyrtidae (Hymenoptera: Chalcidoidea)
IV. Oriental Insects 23: 275-285.
LAWTON, J. H., J. R. BEDDINGTON, AND R. BONSER. 1974. Switching in inverteb-
rate predators, pp. 141-158 in M. H. Usher and H. M. Williamson [eds.], Ecolog-
ical stability. Chapman and Hall, London.
LENTEREN, J. C. VAN. 1986. Parasitoids in the greenhouse: successes with seasonal
inoculative release systems, pp. 342-374 in J. K. Waage and D. J. Greathead
[eds.], Insect parasitoids. Academic Press, London.
LENTEREN, J. C. VAN, AND L. P. J. J. NOLDUS. 1990. Whitefly-plant relationships:
Behavioural and ecological aspects, pp. 47-89 in D. Gerling [ed.], Whiteflies:
their bionomics, pest status and management. Intercept, Andover, UK.
LENTEREN, J. C. VAN, AND J. WOETS. 1988. Biological and integrated pest control
in greenhouses. Annu. Rev. Ent. 33: 239-269.
MACKAUER, M., L. E. EHLER, AND J. ROLAND, [eds.]. 1990. Critical Issues in
Biological Control. Intercept, Andover, UK.
MOUND, L. A., AND S. H. HALSEY. 1978. Whitefly of the World. A Systematic
Catalogue of the Aleyrodidae (Homoptera) with Host Plant and Natural Enemy
Data. John Wiley, London.
MURDOCH, W. W. 1990. The relevance of pest-enemy models to biological control, pp.
1-24 in M. Mackauer, L. E. Ehler, and J. Roland [eds.], Critical issues in biolog-
ical control. Intercept, Andover, UK.
ONILLON, J. C. 1990. The use of natural enemies for the biological control of whiteflies,
pp. 287-314 in D. Gerling [ed.], Whiteflies, their bionomics, pest status and
management. Intercept, Andover, UK.
OR, R. 1986. Examination of factors that limit or encourage contamination of the
whitefly Bemisia tabaci (Gennadius) (Aleyrodidae) on Cotton Plants in the Bet
Shean valley. M.Sc. thesis, Tel Aviv University (Hebrew with English sum-
mary), 51 pp.
OSBORNE, L., K. HOELMER, AND D. GERLING. 1990. Prospects for biological control
of Bemisia tabaci IOBC/WPRS Bulletin 13(5): 153-160.
PRICE, P. W. 1984. Insect Ecology (2nd ed.), Wiley, N.Y.
RAVENSBERG, W. J., M. MALAIS, AND VAN DER SCHAFF. 1990. Applications of Ver-
ticillium lecanii in tomatoes and cucumbers to control whitefly and thrips. IOBC/
WPRS Bulletin 13(5): 168-172.
ROSE, M., AND P. DEBACH. 1982. A native parasite of the bayberry whitefly. Citro-
graph 67(12): 272-276.
ROUSH, R. T. 1990. Genetic variation in natural enemies: critical issues for colonization
and biological control, pp. 263-288 in M. Mackauer, L. E. Ehler, and J. Roland
[eds.], Biological control. Intercept, Andover, UK.
SHIMRON, O. 1991. Processes and cues in host parasitization of Bemisia tabaci
(Aleyrodidae) by its parasitoids Eretmocerus sp. and Encarsia desert
(Hymenoptera). Ph.D. thesis, Tel Aviv University, (Hebrew with English sum-
mary), 147 pp.
SINAI, P. 1990. Differential effects of insects growth regulators on the whitefly Be-
misia tabaci and its parasitoids, as a basis for integrated pest management.
M.Sc. thesis, Tel Aviv University. (Hebrew with English summary), 47 pp.
SMITH, H., D. MALTBY, AND J. E. JIMENEZ. 1964. Biological control of the citrus
blackfly in Mexico. USDA Technical Bul. 1311, Washington, DC.
SOUTHWOOD, T. R. E., AND P. M. READER. 1976. Population census data and key
factor analysis for the viburnum whitefly, Aleurotrachelus jelinekii (Frauenf.),
on three bushes. J. Anim. Ecol. 45: 313-325.
VAN DE VEIRE, M., AND V. VACANTE. 1989. Buprofezin: a powerful help to integ-
rated control in greenhouse vegetables and ornamentals. Boletin de Sanidad Veg-
etal, 17: 425-435.
Florida Entomologist 75(4)
VET, L. E. M., AND J. C. VAN LENTEREN. 1981. The parasite-host relationship be-
tween Encarsia formosa (Hymenoptera: Aphelinidae) and Trialeurodes vap-
orariorum (Westw.) (Homoptera: Aleyrodidae). X. A comparison of three Encar-
sia spp. and one Eretmocerus sp. to estimate their potentialities in controlling
whitefly on tomatoes in greenhouses with a low temperature regime. Z. ang.
Ent. 91: 327-348.
VIGGIANI, G., AND P. MAZZONE. 1979. Contributi alla conoscenza morfobiologica
delle specie del complesso Encarsia Foertser Prospaltella Ashmead (Hym.:
Aphelinidae). 1. Un comment sull'attuale stato, con proposte sinonimiche e de-
scrizione di Encarsia silvestri n.sp. parassita di Bemisia citricola Gom.-Men.
(Hom. Aleyrodidae). Boll. Lab. Entomol. Agrar. "Filippo Silvestri" Portici 36:
WAAGE, J. K., AND H. C. J. GODFRAY. 1984. Reproductive strategies and population
ecology of insect parasitoids, pp. 449-470 in R. M. Sibly, and R. H. Smith [eds.],
Behavioural ecology: ecological consequences of adaptive behaviour. Blackwell,
WAAGE, J., AND D. GREATHEAD [eds.]. 1986. Insect parasitoids. Academic Press,
WALDE, S. J., AND W. W. MURDOCH. 1988. Spatial density dependence in
parasitoids. Annu. Rev. Ent. 33: 441-466.
WATERHOUSE, D. F., AND K. R. NORRIS. 1989. Biological control, Pacific prospects.
WILSON, D., AND B. P. ANEMA. 1988. Development of buprofezin for control of
whitefly Trialeurodes vaporariorum and Bemisia tabaci on glasshouse crops in
the Netherlands and the UK, pp. 175-180 in Brighton Crop Protection Conference
Pests and Diseases.
--- L - N
BIOLOGICAL CONTROL OF WHITEFLIES
WITH ENTOMOPATHOGENIC FUNGI
L. S. OSBORNE AND Z. LANDA
University of Florida, IFAS
2807 Binion Road
Apopka, FL 32703
This paper is a review of 3 fungi (Aschersonia aleyrodis Webber, Verticillium
lecanii (Zimmerman) Viegas and Paecilomyces fumosoroseus (Wize) Brown & Smith
that are being evaluated for the management of injurious polyphagous whiteflies and
was presented as part of a conference on the potential for biological control in the
Caribbean. The prospect for the utilization of biopesticides based on entomopathogenic
fungi is promising. The need to develop alternatives to conventional pesticides has
become apparent in recent years because two of the major whitefly pest species, Be-
misia tabaci (Gennadius) and Trialeurodes vaporariorum (Westwood), have developed
resistance to many of the insecticides used for their control.
Se revisan en este manuscrito 3 species de hongos: Aschersonia aleyrodis Webber,
Verticillium lecanii (Zimmerman) Viegas y Paecelomycesfumosoroseous (Wize) Brown
Biological Control Workshop-'91: Osborne & Landa 457
& Smith los cuales estan siendo evaluados para el manejo de moscas blancas polifagas.
La future ultilizacion de hongos como biopesticidas se consider muy prometedora. La
necesidad de desarrollar alternatives a los pesticides convencionales se hace cada vez
mas aparente dado que las dos species mas importantes de mosca blanca, Bemisia
tabaci (Gennadius) y Trialeurodes vaporariorum (Westwood) han desarrollado resisten-
cia a muchos de los insecticides que se utilizan para su control.
Chemical control of whiteflies (Homoptera: Aleyrodidae) is generally very difficult
because of their morphological and autecological characteristics (waxy substances as a
component of the cuticle, colonization of the underside of leaves, rapid development of
very dense populations, etc.). Furthermore, whiteflies represent a group of insects with
the ability to develop populations that are highly resistant to pesticides. Predators,
parasitoids and/or entomopathogenic microorganisms have been therefore intensively
studied for biological control of whiteflies and have been successfully utilized in biolog-
ical control against several whitefly species, especially the greenhouse whitefly. The
prospect for the utilization of biopesticides based on entomopathogenic fungi is promis-
ing. This paper is a review of 3 fungi that are being evaluated for the management of
injurious polyphagous whiteflies and was presented as part of a conference on the
potential for biological control in the Caribbean.
SWEETPOTATO WHITEFLY AND GREENHOUSE WHITEFLY
The sweetpotato whitefly, Bemisia tabaci (Gennadius), and greenhouse whitefly,
Trialeurodes vaporariorum (Westwood), are considered major pests of economically
important crops worldwide. B. tabaci feed on cotton, cucurbits, lettuce, soybean, to-
matoes and over 500 ornamentals, garden plants and weeds. In recent years, its impor-
tance as a pest of field crops has increased and it has also become one of the most
economically damaging pests of greenhouse crops (Osborne 1988, Osborne et al. 1990a,
1990b, 1990c, Hoelmer et al. 1991). T. vaporariorum is a serious pest of vegetables and
numerous ornamental plants grown in greenhouses and has been reported to feed on
about 250 species of plants throughout the world. Both species are also major pests of
urban vegetable gardens in many states (Greathead 1986, Vet et al. 1980).
The life cycles of both species consist of the egg, four nymphal instars (scales), and
the adult. The egg is spindle-shaped, mostly deposited on the underside of new foliage.
The first instar nymph hatches from the egg and crawls over the leaf surface until it
finds a suitable place to feed. This nymphal stage is oval with well-developed legs and
antennae and is pale green in color. The second and third instar nymphs are flattened,
scale-like in form, and generally transparent with legs and antennae non-functional. The
second through fourth nymphal stages are sedentary throughout their developmental
period. The fourth instar nymph is divided into three substages: the early fourth is
flattened and translucent to opaque-whitish (substage one); the transitional fourth is
thickened, opaque and ensheathed with wax (substage two); the pharate adult is similar
to substage two except that the red eyes of the developing adult are clearly visible and
the body becomes increasingly yellow as the adult develops inside (substage three). The
adult whitefly emerges from a slit in the exoskeleton of the fourth instar nymph (Lopez-
Avila 1986, Gill 1990).
Both adults and nymphs of whiteflies normally occur and feed on the underside of
leaves. Dense populations can adversely affect plant growth as well as deposit copious
amounts of honeydew. Often, a black mold grows on the honeydew which reduces
photosynthesis and especially the aesthetic quality of ornamental plants. Furthermore,
viral plant diseases are also commonly vectored by B. tabaci adults (Cohen 1990). In
458 Florida Entomologist 75(4) December, 1992
Florida, B. tabaci is also responsible for new vegetable and foliage disorders of unknown
etiology (Osborne et al. 1990a, 1990b, Yokomi et al. 1990, Hoelmer et al. 1991).
Control of both species of whitefly by conventional chemical means has become more
difficult in recent years. Many crops grown in greenhouses receive as many as two
applications of different insecticides per week to suppress populations. As a result,
highly resistant populations of both species are common worldwide with residual and
ground water contamination becoming major problems and concerns to the public. These
factors are forcing researchers to look for alternative means of control.
Greenhouse crops with controlled temperature, high relative humidity and reduced
solar radiation offer an excellent opportunity for pest control with entomogenous fungi
(van Lenteren & Woets 1988), especially with several species of deuteromycetes. Both
Aschersonia aleyrodis Webber and Verticillium lecanii (Zimmerman) Viegas represent
common fungal pathogens of whiteflies. Recent data indicate that Paecilomyces
fumosoroseus (Wize) Brown & Smith might play an important role among fungal biopes-
ticides used against whiteflies and other major pests of vegetable and ornamental plants
grown under greenhouse condition or in the field, especially with the humid conditions
of the Caribbean region (Osborne et al. 1990c).
HOST RANGE AND NATURAL OCCURRENCE OF
MAJOR ENTOMOPATHOGENIC FUNGI OF WHITEFLIES
The entomopathogenic fungus, A. aleyrodis, is the most frequently studied species
from about 50 taxa which represent the genus Aschersonia (Petch 1921, Procenko 1967,
Fransen 1987). Aschersonia aleyrodis along with A. flava, A. flavocitrina, A. goldiana,
A. placenta and A. viridans belong to the group Aleyrodiicolae, which includes specific
pathogens of whiteflies.
Aschersonia aleyrodis naturally occurs as a pathogen of several whitefly species in
the subtropical region of the western hemisphere. Natural epizootics were noted in
populations of Aleurocanthus woglumi Ashby, Aleurothrixus floccosus (Maskell), B.
tabaci, B. giffardi (Kotinsky), Dialeurodes citri (Ashmead), D. citrifolii (Morgan), Tet-
raleurodes acaciae (Quaintance), Trialeurodes abutiloneus (Haldeman) and T. vap-
orariorum (Petch 1921, Berger 1921, Mains 1959, Fransen 1987).
Verticillium lecanii is a cosmopolitan species first reported as a pathogen of the
scale insect Coccus viridis (Green) (Hall 1976, 1981) and was described under several
names (Cephalosporium lecanii, C. aphidicola) (Samson & Rombach 1985). Gams (1971)
had revised the taxonomy of the species and placed it into the genus Verticillium,
mainly because of the arrangement of its conidiogenous cells in regular whorls.
Verticillium lecanii is well known as an entomopathogenic fungus, but it is not
restricted to insect hosts. It has been predominately recorded as a pathogen of the
homopterans, particularly of aphids, whiteflies and coccids (Gams 1971, Hall 1976, 1980).
Less frequent are reports of hosts in other insect orders (e.g., Orthoptera, Hemiptera,
Thysanoptera, Coleoptera, Lepidoptera and Hymenoptera) (Gams 1971, Hall 1980,
McCoy et al. 1988). It is also known as a pathogen of non-insect hosts (e.g., arachnids,
either of tetranychid or eryiophid mites) (Gams 1971, Kanagaratnam et al. 1981) as well
as a hyperparasite of several phytopathogenic fungi, mostly rusts (e.g., Uromyces dian-
thi, U. appendiculatus, Puccinia graminis) and powdery mildew (Hall 1981, Deacon
1983). However, as a means of biological control it has been most often tested and/or
used against aphids, whiteflies and thrips in greenhouses.
Paecilomyces fumosoroseus is a cosmopolitan species reported as a pathogen of
many different insect hosts. The fungus was described under several names (Isaria
fumosoroseus Wize, Spicaria aphodii Vuill., S. cossus Portier & Sartory, P. hibernicus
Kennelly & Grimes, and P. isarioides Inagaki) (Samson & Rombach 1985). The genus
Biological Control Workshop-'91: Osborne & Landa
Paecilomyces was described by Bainier and is closely related to Penicillium. The
characteristics used to separate these two genera are that species of Paecilomyces lack
green colored colonies and they have short cylindrical phialides which taper into long
necks. The latest revision of this genus was by Samson (1974) in which he placed 31
species into two sections; Paecilomyces and Isarioidea. P. fumosoroseus belongs to the
section Isarioidea as do several of the entomopathogenic species.
Most of the host records for P. fumosoroseus are from Lepidoptera, Coleoptera, and
Diptera (Bajan 1973, Fargues & Robert 1985, Maniania & Fargues 1984, Poprawski et
al. 1985, Rodriguez-Rueda & Fargues 1980, Zimmerman 1986). The first report of P.
fumosoroseus being a pathogen of whiteflies was from China, where the fungus was
isolated from T. vaporariorum after natural epizootics occurred on this insect in
greenhouses in Beijing (Fang et al. 1983). These new isolates of P. fumosoroseus were
given the trinomial var. beijingensis. Recently, an isolate of P. fumosoroseus highly
virulent to sweetpotato whitefly and a broad range of other pests has been isolated in
Florida. This isolate has caused dramatic natural epizootic in populations of B. tabaci
in both greenhouses and open shade-cloth protected structures. This strain (PFR 97 -
collection of entomopathogenic fungi, University of Florida, CFREC Apopka, Dr. L.
S. Osborne) is currently being evaluated in the laboratory and under greenhouse condi-
tions (Osborne et al. 1990b, Osborne et al. 1990c).
MODE OF INFECTION OF MAJOR FUNGI PATHOGENIC TO WHITEFLIES
The most susceptible stages of whiteflies to infection by A. aleyrodis are the imma-
tures (scales), especially the first, second and third nymphal instars. All substages of
the 4th nymphal instar are less susceptible. First instar nymphs may be infected if
hatched from eggs treated by the conidial suspension of A. aleyrodis (Fransen et al.
1987). No infection caused by A. aleyrodis was noted on whitefly eggs and infection of
adults is rare (Landa 1982, Samson & McCoy 1983, Fransen 1987). Landa (1982) noted
an infection of greenhouse whitefly adults when A. aleyrodis was applied as a part of
an IPM program on greenhouse cucumbers. Infection of adults expresses itself as myce-
lial growth between the head and prothorax, with typical sporulation occurring later.
Recently, an infection of B. tabaci adults caused by A. aleyrodis (strain China 3, collec-
tion of entomopathogenic fungi, Dr. D. Boucias, University of Florida, Gainesville) was
obtained in the laboratory (Landa & Osborne, unpublished data).
The mode of infection of the various entomopathogenic deuteromycetes is basically
quite similar. A typical infection cycle is as follows: conidial attachment, germination,
penetration, vegetative growth and conidiogenesis. The infection process for A.
aleyrodis starts when the one-celled, fusiform conidia (Fig. 1) attaches to the surface
of the host's body. After a swelling and germination phase, the germ tube penetrates
the cuticle. The very early stage of infection is followed by visible color changes of the
infected host. Compared with a healthy host (usually whitish to yellowish green), an
infected nymph turns to a light yellow color and the body appears much less translucent
than uninfected ones. The first visible sign of infection is followed by external growth
of mycelia, which usually takes place all around the body of an infected host (Fig. 2).
Later, the fungus produces vigorous hyphal growth and mat-like pustules which cover
the entire surface of the host body (Fig. 3). The fungus covered host is white, but soon
turns to an orange-reddish color as conidiogenesis begins. Conidia of A. aleyrodis are
produced in slime by phialids which are arranged in cavities or pycnidia (Fig. 4). A
conidial mass is visible as orange-red slimy droplets on a surface of pustules (Berger
1921, Landa 1982, Samson & McCoy 1983, Ramakers & Samson 1984, Samson & Rom-
bach 1985, Fransen 1987). The intensive orange-reddish color of the conidial mass is
caused by presence of carotenoids (mostly beta-carotene) the synthesis of which is
460 Florida Entomologist 75(4) December, 1992
Fig. 1. Conidia of A. aleyrodis attached to the body surface of a greenhouse
whitefly (Trialeurodes vaporariorum) the nymph (SEM, 5000 x).
Fig. 2. Late 4th instar nymphs of greenhouse whitefly a) healthy b) infected with
A. aleyrodis, the fungal growth is visible alongside the body of the mummified dead
host (5 days after initiation of infection) (SEM, 95 x).
Biological Control Workshop-'91: Osborne & Landa 461
-,. !X .'"
.- ; ., o. -.. -.
Fig. 3. The pycnidium of A. aleyrodis formed in the mycelium mat which covers
the entire surface of the mummified nymph of greenhouse whitefly, the initial conidia
are produced in central slimy mass (7 days after initiation of infection) (SEM, 400 x).
1 1,9aa x1
Fig. 4. Large quantities of conidia present in pycnidia of A. aleyrodis at the end of
the infection cycle (SEM, 1000 x).
Florida Entomologist 75(4)
induced by light (Landa et al. 1989). Under optimal conditions, the first symptoms of
infection occur within 24 to 48 h, a vigorous hyphal growth may take place within 4 to
6 days and production of the conidial mass occurs usually from 7 to 9 days after initiation
of infection (Solovej & Kolcov 1976, Ramakers & Samson 1984, Landa 1982, 1984,
Fransen 1987, Rombach & Gillespie 1988).
Aschersonia aleyrodis is a common representative of the entomopathogenic mycof-
lora of plantation tree crops (e.g., citrus) in tropical and subtropical regions (Berger
1921, Petch 1921, Mains 1959), but the nature of the epizootics it produces is not yet
satisfactorily understood. Samson & McCoy (1983) suggest that production of conidia
in a slimy-mass indicates that the conidia are dispersed by free water (dew, rain) or by
animals. Landa (in press) noted the presence of mycophagous mites (Acalvolia sp.) on
leaves with infected nymphs of the common citrus whitefly D. citri. The number of
Acalvolia mites on the leaves correlated with the number of infected host nymphs.
These mites fed on the sporodochia of this fungus and were able to develop and repro-
duce having A. aleyrodis as its only food source. Also, Acalvolia mites lay eggs beside
whitefly eggs and may transfer conidia of A. aleyrodis to whitefly eggs when oviposit-
ing. All these data indicate that this mite may play an important role in A. aleyrodis
The range of the whitefly stages infected by V. lecanii and P. fumosoroseus is
broader than that infected by A. aleyrodis. Both of these fungi cause infections of all
immature stages and adults of both the greenhouse and sweetpotato whitefly. Further-
more, P. fumosoroseus infects whitefly eggs.
The infection process of either V. lecanii or P. fumosoroseus is not known in detail.
In general, these fungi are dispersed as conidia, either by air or water movement or by
other insects and/or mites. The growth of mycelium around an infected host and sub-
sequent sporulation of the fungus may result in further dispersal and secondary infec-
tions caused by these fungi. This spread is probably caused by the casual contact be-
tween infected and healthy individuals. This process plays an important role particularly
in the dense colonies of some pests (whiteflies, aphids, mites etc.). Natural epizootics
are common for both species in the field and greenhouse.
Once the conidia attaches to the host cuticle, it germinates and initial growth is
apparently saprophytic on the outside of the host (Samson & Rombach 1985). Infection
of the host is initiated by an invasion hypha growing through natural orifices and/or
between body segments. Verticillium lecanii forms a cotton-like whitish aerial
mycelium on the infected host. Conidiophores produced on the mycelium of V. lecanii
bear awl-shaped phialids arranged in a characteristic verticillate manner (Fig. 5). Con-
idia are of various shapes (cylindrical to ellipsoidal) and are aggregated in mucus (Fig.
6) (Gams 1971, Hall 1981, Samson & Rombach 1985).
Paecilomyces fumosoroseus produces colonies which are white at first but change
to shades of pink and become light gray when conidia are present. Conidiophores arise
from the basal growth or from aerial hyphae, and have verticillate branches which bear
whorls of 3 to 6 phialids which in turn bear cylindrical to fusiform conidia with rounded
ends. Conidia are born in long chains and fully sporulated colonies (or cultures on an
artificial media) have a powdery gray appearance (Fig. 7) (Samson 1974).
The infection cycles of V. lecanii and especially P. fumosoroseus are somewhat
faster than that of A. aleyrodis. The infection cycle of P. fumosoroseus is particularly
rapid. First, symptoms of infection caused by this fungus are apparent within 24 to 48
h after the conidia contact the insect. Recently, TEM and SEM studies revealed that
P. fumosoroseus conidia attached to the dorsum of the insect and hyphae are present
in the host hemocoel within 24 h. The mycelium is present on the dorsum of the whitefly
body within 48 hours and sporulation occurs within 72 hours (Storey, McCoy & Osborne,
unpublished data). Under optimal conditions, the first visual sign of infection may be
Biological Control Workshop-'91: Osborne & Landa
Fig. 5. Verticillium lecanii a detail of the ellipsoidal conidia arranged in the termi-
nal spherical heads of phialids with no mucilaginous substances on the surface (SEM, 2
Fig. 6. Verticillium lecanii conidia formed on the end of phialids of aerial
mycelium, and protected with a mucilaginous substance (SEM, 1000 x).
Florida Entomologist 75(4)
Fig. 7. Conidia of Paecilomyces fumosoroseus arranged in typical chains which are
formed on phialids on the aerial mycelium (SEM, 2000 x).
noted within 48 to 72 h (mycelial growth on the host surface) and maximum sporulation
may occur within 5 to 7 days (Hall 1981, Landa & Jiranova 1988).
Relative humidity plays the most important role among abiotic factors affecting the
infection cycle of the fungi discussed in this paper. Germination of A. aleyrodis conidia
is retarded if relative humidity declines under 98% and is usually impaired at humidities
lower than 90% (Fransen 1987). This appears to be similar for both V. lecanii and P.
fumosoroseus (Gillespie 1984). Under adverse conditions, the fungus may stay dormant
inside the dead host for several weeks and become active if conditions are suitable
(Landa 1982). Gillespie (1984) showed that successful conidiogenesis of several
deuteromycetes (Metturhi-iam anisopliae (Metschnikoff) Sorokin, Beauveria bassiana
(Balsamo) Vuillemin, V. lecanii, P. fumosoroseus) depends on high relative humidities
(95-100%), and it is not fully realized when relative humidity is below this range. As to
practical utilization of entomopathogenic fungi for whitefly control, it seems critical to
have at least a short period of high humidity during the initiation of the infection process
(conidial attachment to the host surface, conidial swelling, germination and penetration
of the host cuticle). In A. aleyrodis, this process is very rapid and at optimal conditions
may be realized within 12 to 24 h (Rombach & Gillespie 1988). Verticillium lecanii
requires high humidity, especially for conidial germination. During this phase, high
humidity is required for a period of at least 10 to 12 h (Hall 1981, Samson & Rombach
1985). Similarly, P. fumosoroseus requires, for this portion of the infection cycle, a
relative humidity above 95% (Osborne, unpublished data).
Compared with relative humidity, the temperature seems to be less of a limiting
factor. All three fungal species grow and multiply at temperatures between 15 and
30'C, with colony growth optimal between 23 and 25C. Germination of conidia and
Biological Control Workshop-'91: Osborne & Landa
growth of mycelium decline above 25C and cease above 32C (Gillespie 1984, Fransen
1987). Mycelial growth of V. lecanii ceases when it is maintained below 11C (Hall 1981).
Besides humidity and temperature, light also affects some aspects of fungal develop-
ment, particularly the last phase of the infection cycle conidiogenesis. Gillespie (1984)
noted a significant increase in the production of conidia when cultures of P.
fumosoroseus and V. lecanii were exposed to fluorescent light. The same effect was
noted when different strains of A. aleyrodis were cultured under different light condi-
tions. In this case, besides a significant increase in the production of conidia, the amount
of beta-carotene and the shape of the slime-conidial mass were significantly influenced
(Landa et al. 1989).
UTILIZATION OF A. ALEYRODIS AND V. LECANII IN BIOLOGICAL CONTROL
AGAINST WHITEFLIES IN PRACTICE
The utilization of A. aleyrodis as a biological control agent has a long history. First
of all, the introduction and colonization of A. aleyrodis into citrus plantations in Florida
is a classical example of successful biological control against D. citri (citrus whitefly)
and D. citrifolii (cloudy wing whitefly). The fungus was first utilized by making a
conidial suspension from the rinsate of leaves with naturally infected scales. A method
for production of A. aleyrodis conidia on semi-artificial media was developed later
(Berger 1921). Presently, both D. citri and D. citrifolii are naturally controlled by this
introduced fungus and regular chemical treatments are no longer needed (Samson &
McCoy 1983). Similar results have been obtained with A. aleyrodis in the USSR after
the accidental introduction of D. citri into the Azerbadijan region. Several strains of A.
aleyrodis (from China, Cuba, USA, India, Japan) were successfully introduced into
citrus groves by applying conidia produced on a media (Primak & Chiznik 1975,
Ponomorenko et al. 1975). Other successful introductions of A. aleyrodis to control
whiteflies on either citrus (Yen & Tsai 1969, Gao et al. 1985) or other host plants
outdoors have been reported (Chen & Chen 1986, Muraleedharan 1985).
As a consequence of the success in citrus, A. aleyrodis was tested in greenhouses
as a potential selective biological control agent against greenhouse whitefly, T. vap-
orariorum. Different strains of A. aleyrodis were tested as the only regulatory agent
or as a component of an IPM program in combination with the parasitoid Encarsia
formosa Gahan. Most of the experiments were conducted in the USSR (Osokina &
Izevskij 1976, Solovej & Kolcov 1976), Bulgaria (Spasova et al. 1980), Sweden (Ekbom
1979), Netherlands (Ramakers & Samson 1984, Fransen 1987) and Czechoslovakia
(Landa 1982, Landa & Jiranova 1988) with encouraging results. In general, the fungus
was applied as a conidial suspension (1.0 x 106 to 5.0 x 107 conidia per ml) and, in some
cases, infection of scales was as high as 85% (Landa & Jiranova 1988). The potential for
the use of A. aleyrodis to control whiteflies in greenhouses is high, especially when
applied in combination with the parasitoid E. formosa. Aschersonia aleyrodis is a selec-
tive pathogen which does not infect scales parasitized with E. formosa and the
parasitoid E. formosa does not parasitize scales which are in the early phases of the
infection cycle (Landa 1982, 1984, Fransen 1987). A combination of E. formosa and A.
aleyrodis is used experimentally in Czechoslovakia as a component of IPM for
greenhouse cucumber (Landa 1985, Landa & Jiranova 1989); however, there are no
biopesticides based on A. aleyrodis currently available.
In spite of fact that V. lecanii is a common pathogen of whiteflies worldwide, its
practical use is restricted and this fungus is used only against several pests in
greenhouses. Different biopesticides based on V. lecanii are utilized on greenhouse
crops to manage greenhouse whitefly, aphids and thrips in the Netherlands (Van der
Schaaf et al. 1989), Denmark (Borregaard 1991), USSR (Solovej & Sogojan 1982), Swe-
Florida Entomologist 75(4)
den (Ekbom 1979), U.K. (Hall & Burges 1979, Hall 1980, 1985), Czechoslovakia (Sam-
sinakova & Kalalova 1975, Landa & Jiranova 1989) and other countries. Verticillium
lecanii is utilized as a component of IPM programs for cucumbers (Hall 1985, Hussey
1985, Landa & Jiranova 1989) and for ornamental plants (Wardlow 1985, Borregaard
1991). In fact, commercial biopesticides based on conidia or blastospores of different V.
lecanii strains represent one of the few examples of entomopathogenic fungi already
being utilized commercially. Several commercial formulations based on conidia of V.
lecanii are registered in the Netherlands, Denmark and the U.K. (e.g., Mycotal -
formerly a product of Microbial Resources Ltd., U.K., now of Koppert B.V., Netherlands;
Vertalec Koppert B.V.; Microgermin A and F Chr. Hansens Bio Systems Denmark).
The blastospore formulations are used in Czechoslovakia, where V. lecanii is used as
part of their "IPM complete-service-system" and it is not commercialized as a biopes-
ticide for market distribution. Because of problems caused by some new pests that have
been recently introduced into European greenhouses (e.g., B. tabaci and the western
flower thrips Frankliniella occidentalis (Pergande)), the possibility of utilizing V.
lecanii to manage these pests in greenhouses is being intensively studied.
RECENT EXPERIENCES WITH P. FUMOSOROSEUS IN
BIOLOGICAL CONTROL OF B. TABACI
When compared with A. aleyrodis and V. lecanii, much less information is available
about P. fumosoroseus. Nevertheless, results obtained during the last two years indi-
cate that there is a high potential for the implementation of P. fumosoroseus into
practice as a fungal biopesticide against sweetpotato whitefly under greenhouse condi-
tions and possibly on field crops in both Florida and the Caribbean region.
As mentioned above, an extremely virulent strain of P. fumosoroseus was isolated
from naturally infected mealybugs in Florida in 1989 (PFR 97). After isolation, prelim-
inary experiments were conducted using conidial suspensions of PFR 97 against sweet-
potato whitefly populations established on ornamentals under greenhouse conditions.
The results of these experiments were positive and the University of Florida patented
(Osborne 1990) and then licensed the rights to this strain to W. R. Grace & Co., Connec-
ticut. Both parties began a cooperative research program to investigate the possibility
of formulating and utilizing P. fumosoroseus to control various pests of horticultural
plants. Regardless of the fact that this joint research project covers a broad spectrum
of potential host pests, B. tabaci has been studied extensively.
When a modified laboratory in-vitro bioassay was used for comparison (Landa &
Jiranova 1988, Osborne & Landa, unpublished data), the infection cycle of V. lecanii
was 1 to 3 days longer than PFR 97, and the infection cycle of A. aleyrodis was even
longer (4-7 days after PFR 97). The ability of the PFR 97 to infect whitefly eggs is
unique among the entomopathogenic fungi. When exposed to the conidial suspension,
eggs of sweetpotato whitefly are overgrown within 24 to 48 h and the presence of
conidial chains indicates that an egg has sufficient nutrients to allow for the whole
infection cycle. Furthermore, infection of adults that were treated directly or treated
prior to their emergence from the last nymphal stage was frequently observed. An in
vitro bioassay under laboratory conditions demonstrated that conidia applied within 24
h of adult eclosion could regularly infect partly or fully emerged adults. As to the
immatures, younger stages are more sensitive in terms of mortality, but the 4th instar
nymphs support significantly better fungal growth and sporulation. A procedure for
evaluation of the P. fumosoroseus strain was developed which allows us to compare
other strains of this entomopathogenic fungus and to evaluate basic qualitative paramet-
ers of a biopesticide containing P. fumosoroseus before use in experiments. This proce-
dure includes a viability test (standard test of germination of conidia) and an in vitro
Biological Control Workshop-'91: Osborne & Landa
bioassay using early (substage 1) and late 4th instar nymphs (substage 3 partly formed
As a consequence of results obtained in laboratory studies, the first field experiments
with PFR 97 were started in 1990, and other experiments were conducted in 1991. Most
of experiments focused on the management of sweetpotato whitefly populations on
different ornamental plants (e.g., Hibiscus rosa-sinensis, Mandevilla amabilis (Dip-
ladenia rosea), Euphorbia pulcherrima (Poinsettia), and Crossandra infundibulifor-
mis) with different strategies for the utilization of the fungus being tested.
In most of the experiments, conidial suspensions were applied to the plant surface,
especially the underside of leaves, and the results have been very encouraging.
Paecilomyces fumosoroseus significantly reduces populations of sweetpotato whitefly
when applied at weekly intervals. Under normal greenhouse conditions, the establish-
ment of infection is detectable within 7-10 days after the first treatment. After applica-
tion of PFR 97 conidia, a significant increase in the number of dead scales without
visible infection was noted, but when placed into control wet chambers (high relative
humidity), the presence of fungus was usually observed. In contrast to laboratory obser-
vations, infection of whitefly adults is less frequent. Paecilomycesfumosoroseus is very
compatible with some beneficial, especially with Eretmocerus sp. a common parasitoid
of B. tabaci and with Delphastus pusillus (LeConte) a common predator of various
whiteflies in the Florida region. As a part of the PFR 97 project, additional studies
have been conducted. The fungus was tested using one of the common in vitro methods
and found to exhibit a high tolerance to a broad spectrum of fungicides which are
frequently used in protection of greenhouse foliage and ornamental plants: a high level
of tolerance was observed (Osborne & Hoelmer 1990). Tolerance of PFR 97 to most of
the fungicides is similar to that of V. lecanii (Ledieu 1985), and both of these pathogens
and the fungicides may be utilized if applications are scheduled so as to minimize nega-
tive interactions. The integration of V. lecanii and some fungicides is already recom-
mended for special IPM programs (Gardner et al. 1984). Also, the possibility of utilizing
PFR 97 against other greenhouse pests (e.g., aphids, thrips, broad mites, two-spotted
spider mites, leaf miners, mealybugs, etc.) is currently being evaluated in the laboratory
and in university and grower greenhouses. All experiments with PFR are linked to the
production and formulation of this fungus into a standard product to be commercialized
by W. R. Grace & Co.
The three pathogens discussed in this review are frequently found attacking insects
in the greenhouse and in the field. Because these fungi have been found in many differ-
ent environments, it is quite possible that each of these pathogens could be utilized in
Florida and the Caribbean for the management of whiteflies.
Recent trends in greenhouse pest management lead us to believe that future IPM
programs for arthropod pests will rely on various entomopathogenic organisms. These
trends include the elevation of secondary pests to primary pest status when pesticide
pressure is removed once biological control programs are implemented for other pests.
Another complicating factor is the importation of new pest species without adequate
biological controls available, as occurred with leafminers, thrips, and more recently for
B. tabaci. The colonization of exotic pests has a major destabilizing influence on the
IPM practices that are currently being used. Chemicals are utilized to maintain control
of these new pests until methods compatible with established programs can be found.
Viruses, bacteria and fungi are, for the most part, influenced much less than predators
and parasitoids by many of the agricultural chemicals used in these disturbed situations,
but the development of these pathogens into control tactics will take time.
Florida Entomologist 75(4)
The usefulness of various predators and parasitoids is reduced in many ornamental
crops because of the very low aesthetic injury thresholds that currently exist. The
organisms that will be most readily accepted by growers will probably be those that
can be mass released on a regular basis as biological insecticides. Pathogens fit these
criteria much better than do many arthropods with the possible exception of certain
Another complicating factor with the development of any type of control program
for ornamental plants is simply the diversity of plant species and varieties grown. Each
plant type has its own complex of pests and sensitivities to pesticides. In general,
monocultures do not exist in the ornamental industry. A few growers specialize in
growing a few plant types, but the common practice is to produce a wide range of
products which are often grown on the same or adjacent benches in a greenhouse. This
practice of mixed cultures makes it very difficult to treat one crop with a pesticide and
not affect non-target plant material.
We feel that there are excellent opportunities for the use of fungi to control whiteflies
in Florida and the Caribbean. The choice of which pathogen to utilize in a specific
situation can't be answered with our current knowledge; however, we think that each
will find place in arthropod management programs developed in the future.
This is Florida Agricultural Experiment Station Series No. N-00591.
BAJAN, C. 1973. Paecilomyces fumoso-roseus (Wize) pathogenic agent of the Col-
orado beetle (Leptinotarsa decemlineata Say). Ekologia Polska 24: 705-712.
BERGER, E. W. 1921. Natural enemies of scale insects and whiteflies in Florida. The
Quarterly Bull. St. Plant Board 5: 141-154.
BORREGRAARD, S. 1991. Current situation of biological control in ornamentals in De-
nmark. Sting, Newsletter on Biological Control in greenhouses, January 1991: 6-
CHEN, Z. AND Q. CHEN. 1986. An entomogenous fungus of the whitefly Aleuro-
trachelus camelia Kawana. Acta Mycologica Sinica 5: 37-43.
COHEN, S. 1990. Epidemiology of whitefly-transmitted viruses, pp. 211-226 in Gerl-
ing, D. [ed.], Whiteflies: their bionomics, pest status and management.
DEACON, J. W. 1983. Microbial control of plant pests and diseases, pp. 31-42 in As-
pects of microbiology 7. Van Nostrand Reinhold Co. Ltd., Wokingham, England.
EKBOM, B. 1979. Investigation on the potential of a parasitic fungus (Verticillium
lecanii) for biological control of the greenhouse whitefly (Trialeurodes vap-
orariorum). Swedish J. Agric. Res. 136-138.
FANG, Q. X., Y. Z. GONG, Y. Y. ZHOU, Y. M. Hu, AND S. F. YANG. 1983. Paeci-
lomyces fumosoroseus var. beijingensis n. var. Acta Mycologica Sinica 2: 168-
FARGUES, J., AND P. H. ROBERT. 1985. Persistence des conidiospores des hyphomy-
cetes entomopathog6nes Beauveria bassiana (Bals.) Vuill., Metarhizium anisop-
liae (Metch.) Sor., Nomuraea rileyi (F.) Samson et Paecilomyces fumosoroseus
Wize dans le sol, en conditions controlees. Agronomie 5: 73-80.
FRANSEN, J. J. 1987. Aschersonia aleyrodis as a microbial control agent of
greenhouse whitefly. Doctoral Thesis, Univ. Wageningen, 167 pp.
FRANSEN, J. J., C. WINKELMANN, AND J. C. VAN LENTEREN. 1987. The differen-
tial mortality at various life stages of the greenhouse whitefly, Trialeurodes
vaporariorum (Homoptera:Aleyrodidae), by infection with the fungus Ascher-
sonia aleyrodis (Deuteromycotina: Coelomycetes). J. Invert. Pathology 50: 158-
Biological Control Workshop-'91: Osborne & Landa
GAMS, W. [ed.]. 1971. Cephalosporium artige Schimmelpilze (Hyphomycetes). Gus-
tav Fischer, Stuttgart.
GAO, R. X., Z. QUYANG, Z. X. GAO, AND J. X. ZHENG. 1985. A preliminary report
on the application of Aschersonia aleyrodis for the control of citrus whitefly.
Chinese J. Biol. Control 1: 45-46.
GARDNER, W. A., R. D. OETTING, AND G. K. STOREY. 1984. Scheduling of Verticil-
lium lecanii and Benomyl applications to maintain aphid (Homoptera:Aphidae)
control on chrysanthemums in greenhouses. J. Econ. Entomol. 77: 514-518.
GILL, R. J. 1990. The morphology of whiteflies, pp. 13-46 in Gerling, D. [ed.], White-
flies: their bionomics, pest status and management. Athenaeum Press, Newcastle
GILLESPIE, A. T. 1984. The potential of entomogenous fungi to control glasshouse
pests and brown planthoppers of rice. Ph.D. Thesis, University of Southampton,
GREATHEAD, A. H. 1986. Bemisia tabaci host plants, in Cock, M.J.W. [ed.] Bemisia
tabaci a literature survey. Chameleon Press Limited, London, 17-26.
HALL, R. A. 1976. A bioassay of the pathogenicity of Verticillium lecanii on the aphid,
Macrosiphoniella sanborni. J. Invert. Pathology 27, 41-48.
HALL, R. A. 1980. Comparison of laboratory infection of aphids by Metarhizium
anisopliae and Verticillium lecanii. Ann. Appl. Biol. 95: 159-162.
HALL, R. A. 1981. The fungus Verticillium lecanii as a microbial insecticide against
aphids and scales, pp. 483-498, in Burges, H. D. [ed.], Microbial control of pests
and plant diseases. Academic Press, London.
HALL, R. A. 1985. Whitefly control with fungi, pp. 116-118 in Hussey, N. W., and
N.E.A. Scopes [eds.], Biological pest control the glasshouse experiences. Cor-
nell University Press, Ithaca, N.Y.
HALL, R. A., AND H. D. BURGES. 1979. Control of aphids in glasshouses with the
fungus Verticillium lecanii. Ann. Appl. Biology 93: 235-246.
HOELMER, K. A., L. S. OSBORNE, AND R. K. YOKOMI. 1991. Foliage disorders in
Florida associated with feeding by sweetpotato whitefly, Bemisia tabaci. Florida
Entomol. 74: 162-166.
HUSSEY, N. W. 1985. Integrated programmes for specific crops cucumbers, pp. 175-
179 in Hussey, N. W., and N.E.A. Scopes [eds.], Biological pest control the
glasshouse experiences. Cornell University Press, Ithaca, N.Y.
KANAGARATNAM, P., R. A. HALL, AND H. D. BURGES. 1981. New or unusual re-
cords of plant diseases and pests. Plant Pathol. 30: 117-118.
LANDA, Z. 1982. Integrated control against greenhouse whitefly. Ph.D. Thesis, Ag-
ricultural University, Prague, 204 pp.
LANDA, Z. 1984. Protection against greenhouse whitefly Trialeurodes vaporariorum
in IPM programs for greenhouse cucumbers. Sbornik UVTIZ, Zahradnictvi 11:
215-228 (in Czechoslovakian, with English translation).
LANDA, Z. 1985. An IPM program for greenhouse vegetable crops. Zahradnictvi 12:
LANDA, Z., AND R. JIRANOVA. 1988. A laboratory improvement of en-
tomopathogenicfungi and use of selected strains in IPM programme for
greenhouse cucumbers. Annual Report, Agric. University Ceske Budejovice, 95
LANDA, Z., AND R. JIRANOVA. 1989. Entomopathogenic fungi as an additional selec-
tive pest suppressing agents of greenhouse whitefly populations on greenhouse
cucumbers. Proc. Conf. "Biopesticides -theory and practice". Ceske Budejovice,
LANDA, Z., A. JEGOROV, V. MATHA, AND J. NOVAK. 1989. Light induced production
of carotenoids by the entomogenous fungus Aschersonia aleyrodis. Proc. Conf.
"Biopesticides theory and practice. Ceske Budejovice, 110-119.
LEDIEU, M. 1985. Evaluation of side-effects of pesticides by the Glasshouse Crops
Research Institute, pp. 153-161 in Hussey, N. W., and N.E.A. Scopes [eds.],
Biological pest control the glasshouse experience. Cornell Univ. Press, Ithaca,
470 Florida Entomologist 75(4) December, 1992
LENTEREN, J. C. VAN, AND J. WOETS. 1988. Biological and integrated pest control
on greenhouses. Annual Review of Entomol. 33: 239-269.
LOPEZ-AVILA, A. 1986. Taxonomy and biology of Bemisia tabaci, pp. 3-12 in Cock,
M.J.W. [ed.], Bemisia tabaci a literature survey. Chameleon Press Limited,
MAINS, E. B. 1959. North American species of Aschersonia parasitic on Aleyrodidae.
J. Invert. Pathol. 1: 43-47.
MAINANIA, N. K., AND J. FARGUES. 1984. Specifit6 des Hyphomycetes en-
tomopathog6nes pour les larves de lepidopt6res Noctuidae. Entomophaga 29:
McCoY, C. W., R. A. SAMSON, AND D. G. BOUCIAS. 1988. Entomogenous fungi, pp.
151-236 in Ignoffo, C. M. [ed.], CRC Handbook of Natural Pesticides. CRC
Press, Boca Raton, Vol.5, A.
MURALEEDHARAN, N. 1985. New disease of the cardamon whitefly. Trop. Pest Man-
agement 31: 234-235.
OSBORNE, L. S. 1988. The not so sweet sweetpotato whitefly. Florida Foliage 14(5): 8-
OSBORNE, L. S. 1990. Biological control of whiteflies and other pests with a fungal
pathogen. United States Patent Number 4,942,030.
OSBORNE, L. S., AND K. A. HOELMER. 1990. Potential for using Paecilomyces
fumosoroseus to control Sweetpotato whitefly, Bemisia tabaci, pp. 79-80 in
Yokomi, R. K., Narayanan, K. R., Schuster, D. J. [eds.], Sweetpotato whitefly
mediated.vegetable disorders in Florida. Proc. Workshop TR&EC Univ. Florida,
Homestead, February 1-2.
OSBORNE, L. S., G. K. STOREY, C. W. McCOY, ANDJ. F. WALTER. 1990a. Potential
for controlling the sweetpotato whitefly, Bemisia tabaci, with the fungus
Paecilomycesfumosoroseus. Proc. 5th Internat. Colloquium on Invert. Pathology
and Biol. Control, Adelaide, Australia, August 20-24, 386-390 pp.
OSBORNE, L. S., K. A. HOELMER, AND R. K. YOKOMI. 1990b. Foliage disorders in
Florida associated with feeding by the sweetpotato whitefly, Bemisia tabaci, pp.
49-52 in Yokomi, R. K., K. R. Narayanan, and D. J. Schuster [eds.], Sweetpotato
whitefly mediated vegetable disorders in Florida. Proc. Workshop TR&EC Univ.
Florida, Homestead, February 1-2.
OSBORNE, L. S., K. A. HOELMER, AND D. GERLING. 1990c. Prospects for biological
control of Bemisia tabaci. SROP/WPRS Bull. XIII/5: 153-160.
OSOKINA, G. A., AND S. S. IZEVSKIJ. 1976. A test on the control of greenhouse
whitefly. Zashchita Rastenij 2: 28-29.
PETCH, T. 1921. Studies in entomogenous fungi; II. The genera Hypocrella and As-
chersonia. Royal Bot. Garden, Peradeniya Annals 7: 167-278.
PONOMORENKO, N. G., H. A. PRILEPSKAJA, M. Y. MURVANIDZE, AND L. A. STOL-
JAROVA. 1975. Aschersonia against whiteflies. Zashchita Rastenij 6: 44.
POPRAWSKI, T. J., M. MARCHAL, AND P. H. ROBERT. 1985. Comparative susceptibil-
ity of Otiorhynchus sulcatus and Sitona lineatus (Coleoptera: Curculionidae)
early stages to five entomopathogenic Hyphomycetes. Environ. Entomol. 14:
PRIMAK, T. A., AND R. I. CHIZHIK. 1975. The basis for possible use of Aschersonia
aleyrodis in the control of glasshouse whitefly. Zashch. Rast., Mez. Temat Nauk.
PROCENKO, E. P. 1967. Fungi of the genus Aschersonia. Sb. Karant. Rast., Kolos,
RAMAKERS, P.M.J., AND R. A. SAMSON. 1984. Aschersonia aleyrodis, a fungal
pathogen of whitefly. II. Application as a biological insecticide in glasshouses. J.
Appl. Entomol. 97: 1-8.
RODRIGUEZ-RUEDA, D., AND J. FARGUES. 1980. Pathogenicity of entomopathogenic
hyphomycetes, Paecilomyces fumosoroseus and Nomuraea rileyi, to eggs of
Noctuids, Mamestra brassicae and Spodoptera littoralis. J. Invert. Pathol. 36:
Biological Control Workshop-'91: Osborne & Landa
ROMBACH, M. C., AND A. T. GILLESPIE. 1988. Entomogenous Hyphomycetes for
insect and mite control on greenhouse crop. Biocontrol News & Information
SAMSINAKOVA, A., AND S. KALALOVA. 1975. Artificial infection of scale insects with
entomophagous fungi, Verticillium lecanii and Aspergillus candidus. En-
tomophaga 20: 361.
SAMSON, R. A. 1974. Paecilomyces and some allied hyphomycetes. Studies in Mycol-
ogy 6: 1-43.
SAMSON, R. A., AND C. W. McCoY. 1983. Aschersonia aleyrodis, a fungal pathogen
of whitefly. I. Scanning electron microscopy of the development on the citrus
whitefly. J. Appl. Entomol. 96: 380-386.
SAMSON, R. A., AND M. C. ROMBACH. 1985. Biology of the Verticillium and Ascher-
sonia, pp. 34-42 in Hussey, N. W., and N.E.A. Scopes [eds.], Biological pest
control the glasshouse experience. Cornell Univ. Ithaca Press, N.Y.
SCHAFF, D. A. VAN DER, M. MALAIS, AND W. J. RAVENSBERG. 1989. Experiments
with Verticillium lecanii in cucumber against whitefly and thrips. Proc. Conf.
"Biopesticides theory and practice." Ceske Budejovice, Sept. 1989: 191-195.
SOLOVEJ, E. F., AND P. D. KOLCOV. 1976. The action of entomogenous fungi of the
genus Aschersonia on the greenhouse whitefly. Mikologija i Fitopatologija 10:
SOLOVEJ, E. F., AND L. N. SOGOJAN. 1982. Fungus against the glasshouse whitefly,
biological control of Trialeurodes vaporariorum. Zashch. Rastenij, Kiev 5: 28-32.
SPASOVA, P., E. KHRISTOVA, AND E. S. ELENKOV. 1980. Pathogenicity of various
species of fungi of genus Aschersonia to larvae of the greenhouse whitefly
(Trialeurodes vaporariorum Westw.) on tomatoes and cucumbers. Gradinarska
i Lozarska Nauka 17: 70-71.
VET, L.E.M., J. C. VAN LENTEREN, AND J. WOETS. 1980. The parasite-host re-
lationship between Encarsia formosa (Hymenoptera: Aphelinidae) and
Trialeurodes vaporariorum(Homoptera: Aleyrodidae). IX. A review of the
biological control of the greenhouse whitefly with suggestions for future research.
Zeit. Angew. Entomologie 90: 26-51.
WARDLOW, L. R. 1985. Integrated programmes for specific crops Chrysanthemums,
pp. 180-185 in Hussey, N. W., and N.E.A. Scopes [eds.], Biological pest control
the glasshouse experiences. Cornell Univ. Press, Ithaca, N.Y.
YEN, D. F., AND Y. T. TSAI. 1969. Entomogenous fungi of citrus Homoptera in
Taiwan. Plant Protection Bulletin, Taiwan 11: 1-10.
YOKOMI, R. K., K. A. HOELMER, AND L. S. OSBORNE. 1990. Relationship between
the sweetpotato whitefly and the squash silverleaf disorder. Phytopathology 80:
ZIMMERMANN, G. 1986. The 'Galleria bait method' for detection of entomopathogenic
fungi in soil. J. Appl. Entomol. 12: 213-215.
Fig. 1. Conidia of A. a.eyrodis attached to the hody surface of a greenhouse
whitefly (TrialteHrodes taporarinru.m) the nymph (SEM, 5000 x).
Fig . ,ate I'" instar nymphs of greenhouse whiteily a) healthy b) intfeted with
.IL ft,'yrodiis, the fungal growth is visible alongsIie the body tf the mummified dead
host (5 days after initiation of infection) (SEM. 95 x ).
Fig. 3, The ipynidium of A. aleyrodis formed in the mycelium mat which covers
the entire surface of Lhe mummified nymph of greenhouse whitefly, the initial conidia
are produced in central slimy mass (7 days after initiation of infection) (SEM, 40K) x ).
Fig. 4. Large quantities of conidia present in pycnidia of A, arrleydis at the end ot
the infection cycle (SSEM, 1000) x).
Fig. F.. Verfivillm kcaDii a detail of the ellipsoidal cvonifia arranged in the termi-
tial -pherieal heads of phiali&-, with no mucilaginous subtstriics o~n tht- .-urfavIe fSNM, 2
RN4 x ).
Fig. 6. VIrrnii con cdia formed on the end of phiakids of are~iaI
mycelium, and protected with a mucilaginous substance (E.M. 1cMNI x ).
Florida Entomologist 75(4)
TECHNIQUES FOR EVALUATION OF
PREDATORS OF HOMOPTERA
JOHN J. OBRYCKI
Department of Entomology,
Iowa State University, Ames, Iowa 50011
The experimental evaluation of predation in biological control can be difficult, but is
necessary to determine the impact of predators on pest population dynamics. These
evaluative techniques document the role of existing predators, quantify the impact of
previous releases, and also identify pest life stages or seasonal periods that may be
targeted for additional releases. An overview of evaluation methods appropriate for
Caribbean Basin biological control programs for Homopteran pests is presented. Several
examples of interactions between predators and other types of natural enemies are
discussed to provide a basis for interpretation of results from an evaluation study.
La evaluacion de depredacion en el control biologico puede ser dificil, pero es
necesario determinar el impact de los predadores en la dinamica poblacional de las
plagas. Estas tecnicas de evaluacion documentan el papel de los predadores existentes,
cuantifican el impact de liberaciones previas, y tambien identifican los estados de vida
de las plagas o las estaciones en las cuales pueden efectuarse liberaciones
adicionales. Se present una sintesis de la evaluation de metodos apropriados para el
control biologico de plagas de homopteros en la Region del Caribe. Se discuten various
ejemplos de interaccciones entire predadores y otro tipo de enemigos naturales con el
fin de dar una base para la interpretation de resultados en un studio de evaluacion.
In classical biological control programs, evaluation studies are an often neglected,
but critical, research component, that demonstrate the effectiveness of natural enemies.
These post-release studies are required to document the ecological and economic bene-
fits of a biological control program and can also identify pest life stages that may require
additional releases (DeBach et al. 1976). Evaluation of the impact of arthropod predators
may be more difficult than for parasitoids or pathogens because prey remains may be
difficult to locate and identify or prey may be totally consumed (Whitcomb & Godfrey
1991, Greenstone 1990, Hagen et al. 1976).
In this paper, I present a brief overview of prey specificity and methods to evaluate
entomophages, discuss examples of evaluation methods that have been employed with
particular predator-prey systems, and finally, outline some particular areas that should
be considered when conducting predator evaluation studies. The goal of this paper is to
provide the participants in this workshop with the background to assess the applicability
of these techniques in Caribbean Basin biological control programs directed against
One key area that relates directly to the use of predators in biological control is prey
specificity. This topic has been discussed for at least 40 years in relation to the establish-
ment of predators in classical biological control programs. For example, after conducting
research in Bermuda on scale-feeding coccinellids, Thompson (1951) pointed out that,
although insect predators generally have wider prey ranges than insect parasitoids,
predators demonstrate a high degree of prey specificity. Predatory feeding behavior is
Biological Control Workshop-'91: Obrycki
generally adaptive, not random, and the understanding of prey specificity continues to
be a critical area in biological control research (Gilbert 1990, Greenstone 1990, Hagen
1987, Tauber & Tauber 1987). Prey specificity presents a dual problem for biological
control programs that involve predators. On the one hand, the interaction between the
predator and target pest must be known, on the other hand, the nutritional require-
ments for a predator may include, not only the target prey, but alternate prey/food
sources (e.g., Obrycki & Orr 1990, Ruberson et al. 1986).
As pointed out by Luck et al. (1988), clear definition of the objectives of an evaluation
study will aid in the selection of the most appropriate techniques. There are strengths
and weaknesses with each technique used to evaluate predators, and generally more
than one method is required to adequately assess the impact of a natural enemy (Kiritani
& Dempster 1973). Experimental techniques have been the most useful means to meas-
ure the effectiveness of entomophagous species. These techniques were described by
Smith & DeBach (1942), summarized by DeBach & Bartlett (1964), DeBach & Huffaker
(1971), and DeBach et al. (1976), and were recently reviewed by Luck et al. (1988).
These reviews should be consulted for detailed descriptions of the strengths and weak-
nesses of the various evaluation techniques.
Six experimental evaluation techniques (following terminology of Luck et al. 1988)
suitable for measuring the impact of a predatory species are:
(1) The introduction or addition of a predator into a new environment, that includes
a quantitative comparison of the density of the prey before and after introduction of the
(2) Exclusion or inclusion studies using mechanical barriers; for example, compara-
tive open and closed cages or screen cages with variously sized mesh can be used to
exclude predators or enclose known densities of predators and prey (see Obrycki et al.
(3) Removal techniques involving selective insecticides to eliminate predators, ants
to interfere with predator activity, or hand-removal of predators. These techniques
severely reduce densities of predators, but do not completely exclude them.
(4) Prey enrichment studies that place prey (e.g. eggs) in the field to estimate
predation rates with minimal habitat modification.
(5) Direct observation of the target pest in the field to record predation under
natural conditions. This technique provides the identity of predators and estimate of
(6) Use of highly specific serological tests or marking techniques with rare elements
or radioactive isotopes to provide chemical evidence of predation. These procedures
accurately identify what predator is feeding on a prey species or a specific life stage of
a target pest.
Given these techniques, I will discuss several examples of evaluations of the role of
predators in different agricultural systems. These examples include a range of objec-
tives and approaches by several researchers to demonstrate different ways to use these
The first example involves predation by Hippodamia convergens and H. sinuata
(Coleoptera: Coccinellidae) on grain sorghum infested with greenbugs Schizaphis
graminum (Homoptera: Aphididae) (Kring & Gilstrap 1984). Using three types of field
cages, Kring et al. (1985) demonstrated the importance of these two coccinellid species
in reducing late-season greenbug infestations. This direct measurement of greenbug
predation documented suppression of greenbug densities due to the activity of the two
Hippodamia species. Previously, because of observations of high numbers of parasitized
greenbugs, aphid parasitoids were believed to be the major suppressive agent in this
system. Subsequent research showed that an early-season infestation of corn leaf aphid
(Rhopalosiphum maidis), which is not an economic pest, led to higher densities of
Hippodamia spp. in greenbug-infested sorghum (Kring & Gilstrap 1986).
474 Florida Entomologist 75(4) December, 1992
O'Neil and coworkers, although not working with predators of Homopterans, have
made valuable contributions in the determination of field predation rates and quantifica-
tion of the relationship between predatory searching behaviors and plant growth
dynamics (O'Neil & Stimac 1988, O'Neil & Wiedenmann 1987). Using field cage studies,
they determined that several arthropod predators in soybeans maintain a low constant
predation rate relative to increasing prey densities. On the basis of their studies, they
caution about the interpretation of laboratory functional response studies conducted at
high prey densities. They point out that field studies of predation need to examine the
activity of predators at low, as well as high, prey densities as predators may function
at low prey densities to prevent pest outbreaks.
My third example considers serological techniques that are used to determine which
predators are feeding upon target pests in the field (Greenstone 1990). Recent develop-
ments using monoclonal antibodies have made these assays more sensitive and are also
the basis for a specific immunodot assay that is rapid, inexpensive and easily interpreted
(Stuart & Greenstone 1990, Greenstone & Morgan 1989). Cooperative research efforts
may be required because of equipment costs and the necessity to integrate these
serological studies with detailed ecological field work.
As I have emphasized in this paper, several evaluation techniques are required to
adequately assess the role of insect predators. Even if several techniques are used to
evaluate the impact of predators, the interpretation of results of an evaluation study
may be complex. For example, a recent study by McConnell & Kring (1990) quantified
predation and dislodgement rates of greenbugs by adult Coccinella septempunctata
(Coleoptera: Coccinellidae). They concluded that an estimate of the efficacy of C. sep-
tempunctata based solely on prey consumption would underestimate the effect of this
predator on greenbug densities due to dislodgement. In the field, greenbug adults and
nymphs were four to five times more likely to be dislodged, and thus exposed to addi-
tional abiotic and biotic mortality factors, than to be consumed by C. septempunctata.
Additionally, T. J. Kring (Dept. Entomology, Univ. of Arkansas, personal communica-
tion) has shown that C. septempunctata will consume parasitized greenbugs at higher
rates than nonparasitized individuals. Thus, in an evaluation of the impact of C. septem-
punctata on greenbug population dynamics, dislodgement and predation of parasitized
individuals need to be considered to accurately measure the predator's impact.
Not only do predators interact with parasitoids, but entomopathogens may also
directly infect predators or indirectly influence predators through prey (Vinson 1990).
This relationship may or may not be detrimental. For example, the microsporidium
Nosema pyrausta, which infects the European corn borer, Ostrinia nubilalis, has no
measurable effect on the lacewing predator Chrysoperla carnea (Sajap & Lewis 1989).
On the basis of these studies, the authors concluded that lacewing predation and infec-
tion by N. pyrausta were compatible mortality factors for suppression of 0. nubilalis.
As shown by these examples, interactions of biotic mortality factors are common in
biological control. The interpretation and analysis of these interacting mortality factors
have been a recurring theme in insect population ecology (e.g., Morris 1965, Varley et
al. 1973, Jones 1982). Recently, a new method, based upon human actuarial tables, to
examine interacting mortality factors has been presented by Carey (1989). This ap-
proach allows one to separately analyze mutually exclusive causes of mortality in a life
table; i.e., pathogens, parasitoids, and predators.
In summary, Luck et al. (1988) have proposed a logical framework for designing a
workable evaluation program for predators of Homopterans in a Caribbean Basin biolog-
ical control project,
1. Develop a sampling scheme suitable for both the predator and prey.
2. Use an insecticide disruption technique if prey has demonstrated insecticide re-
Biological Control Workshop-'91: Obrycki 475
3. Conduct exclusion experiments that include control cages to monitor the micro-
habitat effects of the barriers.
4. Enhance prey levels in the field, and make direct observations of the predators
attacking the target pest.
I thank Harold Browning, University of Florida, Lake Alfred Experiment Station,
for the invitation to speak at this workshop, the Caribbean Basin Administrative Group
for financial support to attend this workshop, and Clay McCoy, University of Florida,
Lake Alfred Experiment Station, for his patience. This is Journal Paper J- 14613 of the
Iowa Agriculture and Home Economics Experiment Station, Ames, Project 2755.
CAREY, J. R. 1989. The multiple decrement life table: a unifying framework for cause-
of-death analysis in ecology. Oecologia 78: 131-137.
DEBACH, P., AND B. R. BARTLETT. 1964. Methods of colonization, recovery and
evaluation, pp 402-426 in P. DeBach [ed.], Biological control of insect pests and
weeds. Reinhold, New York.
DEBACH, P., AND C. B. HUFFAKER. 1971. Experimental techniques for evaluation
of the effectiveness of natural enemies, pp. 113-140 in C. B. Huffaker [ed.],
Biological control. Plenum Press, New York.
DEBACH, P., C. B. HUFFAKER, AND A. W. MACPHEE. 1976. Evaluation of the im-
pact of natural enemies, pp. 255-285 in C. B. Huffaker and P. S. Messenger
[eds.], Theory and practice of biological control. Academic Press, New York.
GILBERT, F. 1990. Size, phylogeny and life-history in the evolution of feeding speciali-
zation in insect predators, pp. 101-124 in F. S. Gilbert [ed.], Insect life cycles:
Genetics, evolution and coordination. Springer-Verlag, London.
GREENSTONE, M. H. 1990. Foreign exploration for predators: A proposed new
methodology. Environ. Entomol. 18: 195-200.
GREENSTONE, M. H., AND C. E. MORGAN. 1989. Predation on Heliothis zea
(Lepidoptera: Noctuidae): An instar-specific ELISA assay for stomach analysis.
Ann. Entomol. Soc. Am. 82: 45-49.
HAGEN, K. S. 1987. Nutritional ecology of terrestrial predators, pp. 533-577 in F.
Slansky and J. G. Rodriguez [eds.], Nutritional ecology of insects, mites, spiders,
and related invertebrates. John Wiley & Sons, New York.
HAGEN, K. S., S. BOMBOSCH, AND J. A. MCMURTRY. 1976. The biology and impact
of predators, pp. 93-142 in C. B. Huffaker and P. S. Messenger [eds.], Theory
and practice of biological control. Academic Press, New York.
JONES, D. 1982. Predators and parasites of temporary row crop pests: Agents of irre-
placeable mortality or scavengers acting prior to other mortality factors? En-
tomophaga 27: 245-266.
KIRITANI, K., AND J. DEMPSTER. 1973. Different approaches to the quantitative
evaluation of natural enemies. J. Appl. Ecol. 10: 323-340.
KRING, T. J., AND F. E. GILSTRAP. 1984. Efficacy of the natural enemies of grain
sorghum aphids (Homoptera: Aphididae). J. Kansas Entomol. Soc. 57: 460-467.
KRING, T. J., AND F. E. GILSTRAP. 1986. Beneficial role of corn leaf aphid,
Rhopalosiphum maidis (Fitch) (Homoptera: Aphididae), in maintaining Hip-
podamia spp. (Coleoptera: Coccinellidae) in grain sorghum. Crop Prot. 5: 125-
KRING, T. J., F. E. GILSTRAP, AND G. J. MICHELS, JR. 1985. Role of indigenous
coccinellids in regulating greenbugs (Homoptera: Aphididae) on Texas grain sor-
ghum. J. Econ. Entomol. 78: 269-273.
LUCK, R. F., B. M. SHEPARD, AND P. E. KENMORE. 1988. Experimental methods
for evaluating arthropod natural enemies. Annu. Rev. Entomol. 33: 367-391.
Florida Entomologist 75(4)
McCONNELL, J. A., AND T. J. KRING. 1990. Predation and dislodgement of
Schizaphis graminum (Homoptera: Aphididae), by adult Coccinella septempunc-
tata (Coleoptera: Coccinellidae). Environ. Entomol. 19: 1798-1802.
MORRIS, R. F. 1965. Contemporaneous mortality factors in population dynamics.
Canadian Entomol. 97: 1173-1184.
OBRYCKI, J. J., AND C. J. ORR. 1990. Suitability of three prey species for Nearctic
populations of Coccinella septempunctata, Hippodamia variegata, and Propylea
quatuordecimpunctata (Coleoptera: Coccinellidae). J. Econ. Entomol. 83: 1292-
OBRYCKI, J. J., M. J. TAUBER, AND W. M. TINGEY. 1983. Predator and parasitoid
interaction with aphid-resistant potatoes to reduce aphid densities: A two-year
field study. J. Econ. Entomol. 76: 456-462.
O'NEIL, R. J., AND J. L. STIMAC. 1988. Measurement and analysis of arthropod pre-
dation on velvetbean caterpillar, Anticarsia gemmatalis, Hubner, in soybeans.
Environ. Entomol. 17: 821-826.
O'NEIL, R. J., AND R. N. WIEDENMANN. 1987. Adaptations of arthropod predators
to agricultural systems. Florida Entomol. 70: 40-48.
RURERSON, J. R., M. J. TAUBER, AND C. A. TAUBER. 1986. Plant feeding by
Podisus maculiventris (Heteroptera: Pentatomidae): Effect on survival, develop-
ment, and preoviposition period. Environ. Entomol. 15: 894-897.
SAJAP, A. S., AND L. C. LEWIS. 1989. Impact of Nosema pyrausta (Microsporida:
Nosematidae) on a predator, Chrysoperhi carnea (Neuroptera: Chrysopidae).
Environ. Entomol. 18: 172-176.
SMITH, H. S., AND P. DEBACH. 1942. The measurement of the effect of entomophag-
ous insects on population densities of their hosts. J. Econ. Entomol. 35: 845-849.
STUART, M. K., AND M. H. GREENSTONE. 1990. Beyond ELISA: A rapid, sensitive,
specific immunodot assay for identification of predator stomach contents. Ann.
Entomol. Soc. Am. 83: 1101-1107.
TAUBER, C. A., AND M. J. TAUBER. 1987. Food specificity in predacious insects: a
comparative ecophysiological and genetic study. Evol. Ecol. 1: 175-186.
THOMPSON, W. R. 1951. The specificity of host relationships in predaceous insects.
Canadian Entomol. 83: 262-269.
VARLEY, G. C., G. R. GRADWELL, AND M. P. HASSELL. 1973. Insect population
ecology. University of California Press, Berkeley.
VINSON, S. B. 1990. Potential impact of microbial insecticides on beneficial arthropods
in the terrestrial environment, pp. 43-64. in M. Laird, L. A. Lacey, and E. W.
Davidson [eds.], Safety of microbial insecticides. CRC Press, Boca Raton, Fl.
WHITCOMB, W. H., AND K. E. GODFREY. 1991. The use of predators in insect control,
pp. 215-241 in D. Pimental [ed.], CRC Handbook of pest management in agricul-
ture, 2nd Ed., Vol II. CRC Press, Boca Raton, Fl.
Biological Control Workshop-'91: Baker et al. 477
OVERVIEW OF BIOLOGICAL CONTROL OF LEPIDOPTERA
IN THE CARIBBEAN
P. S. BAKER, A. KHAN, A. I. MOHYUDDIN, AND J. K. WAAGE
International Institute of Biological Control (IIBC)
(An Institute of CAB International)
Gordon Street, Curepe,
TRINIDAD & TOBAGO
Some previous biological control programs in the Caribbean are listed. Three cases
are considered in detail: 1) Biocontrol of the sugarcane borer (Diatraea saccharalis) has
been a notable success, with effective control in several countries in the Caribbean. The
braconid parasitoid Cotesiaflavipes has been shown to be particularly adaptable to new
hosts and future studies should exploit this quality still further. 2) Biocontrol of the
diamondback moth on the other hand has been generally less successful, mostly due to
the very low economic threshold required of crucifers but also because the parasitoids
that have been tried up to now do not seem to work as well in the hot tropics as in more
temperate areas. 3) Biocontrol of the pigeonpea pod borer is still in its infancy but is a
suitable case for treatment.
The past history of lepidopteran biocontrol in the Caribbean suggests that prolonged
effort over many years yields results. It is difficult to finance years of sustained effort;
the challenge for the future is to coordinate the endeavors of several Caribbean organi-
zations in order to focus them on a few key pest problems.
Se mencionan algunos programs de control biologico en el Caribe. Tres casos se
consideran en detalle. 1) El exito notable del control biologico del barrenador de la cafia
de azucar Diatraea saccharalis en la region del Caribe. Se ha demostrado que el
parasitoide braconido Cotesiaflavipes se puede adaptar a nuevos hospederos, y studios
futures deben explotar esta cualidad. 2) El control biologico de la polilla diamante ha
tenido menos exito, debido a el bajo nivel de dalo economic que se require en las
cruciferas, y tambien porque los parasitoides que han sido probados hasta ahora, no
parecen ser efectivos tanto en el tropico caliente como en las areas templadas. 3) El
control biologico del barrenador del gandul esta en vias de desarrollo pero puede
The purpose of this paper is to review some features of lepidopteran biological
control in the region and to draw some inferences about where trends may be leading.
It is not intended to be comprehensive and it will be biased towards those areas where
IIBC (formerly CIBC) has expertise or interests.
Past attempts in the English speaking Caribbean were reviewed by Cock (1985) to
which the reader is referred for more detail. He mentions around 50 Caribbean pest
genera, of which 16 are Lepidoptera. A list of non-crucifer lepidopteran biocontrol
attempts appears in Table 1 (for more on crucifer biocontrol, see Alam 1992).
Instead of reviewing past attempts we will deal with three cases in more detail
which cover a past success, a present initiative and a future challenge.
Florida Entomologist 75(4)
TABLE 1. BIOCONTROL OF SOME LEPIDOPTERAN SPECIES IN THE ENGLISH-SPEAK-
ING CARIBBEAN (AFTER COCK 1985).
Sugarcane borers successes
Sweet Potato Leafroller
Successful control claimed in Guyana by
manipulation of a bacterium
Small parasitoid releases, no establishment
Small predator releases, no establishment
Small parasitoid releases, no establishment
Small parasitoid releases, no establishment
Many parasitoid releases, establishment of
several species, notable successes
Small parasitoid releases, no establishment
Small to moderate releases, no establishment
Small parasitoid releases, no establishment
Arrowroot Leafroller Small parasitoid releases, one species
(Calpodes ethlius) established
Armyworms (Spodoptera & Heliothis) 20 species shipped to Barbados (1968-1976),
numerous releases, two established, Telenomus remus claimed as a success. 18 species
released in Trinidad (1973-1982), none established.
BIOCONTROL OF SUGARCANE BORERS (DIATRAEA SPP.)
Biocontrol of Diatraea saccharalis (F.) is the most outstanding success story of
biocontrol in the Caribbean and in some ways it is an unlikely one. Diatraea spp. are
indigenous, having adapted to the Polynesian sugarcane from local wild grasses. Some
native parasitoids followed them to the introduced plant, but others did not, probably
because of the greater protection to D. saccharalis afforded by the thick stemmed cane.
Attempts at biocontrol have been going on for over 60 years, and at least 54 species
of parasitoids (12 tachinids, 42 hymenopterans) have been studied as possible candidates
Trichogramma spp. egg parasitoids were first released massively in 1921 in Guyana.
Bates (1954) carried out mass release experiments with controls for five successive
years but could find no differences between release and control plots. Later Metcalfe
& Breniere (1969) concluded that this was not effective for sugarcane; nevertheless T.
minutum, T. fasciatum, Telenomus alecto and Trichogrammatoidea eldanae were still
being released in Guyana in 1990 against D. centrella apparently without effect
(Quashie-Williams, pers. comm.).
Tachinid and braconid parasitoids have been more successful however, and a sum-
mary of successful establishments appears in Table 2. There have also been similar
successes in S. America, the most recent in Venezuela where the release of Cotesia
flavipes reduced damage to well below the economic threshold.
These successes have been mostly against D. saccharalis, and in many cases it is
now D. centrella which is the main pest because of its ability to encapsulate the
parasitoid's eggs within the hemocoel. There is anecdotal evidence however that C.
Biological Control Workshop-'91: Baker et al. 479
TABLE 2. SUMMARY OF SUCCESSFUL ESTABLISHMENTS OF PARASITOIDS OF DIA-
TRAEA SPP. (AFTER COCK 1985).
Results Species established
Antigua Lixophaga diatraeae
Sugar production now stopped
Barbados Lixophaga diatraeae
Together give good control Apantelesflavipes
Dominica Lixophaga diatraeae
Good control; Paratheresia claripalpis
P. claripalpis more important
Guyana Metagonistylum minenese
Good control of D. saccharalis
St Kitts Lixophaga diatraeae
Effective until 1960 Apantelesflavipes
St Lucia Metagonistylum minenese
Trinidad Apanteles flavipes
Fairly good control
flavipes has overcome this problem in some areas. Other parasitoids such as Allorhogas
pyralophagus and Pediobius furvus, are being tried in Guyana.
The success of C. flavipes against D. saccharalis is particularly noteworthy. Diat-
raea spp. was not its original host. C. flavipes is a native of the oriental region, between
32 N and 220 S in S.E. Asia (Mohyuddin 1971) where it has been reported from a large
number of pyralid and noctuid graminaceous stem borers. Mohyuddin et al. (1981)
studied the strain of C. flavipes established in Pakistan on the maize borer Chilo partel-
lus. By analyzing behavior and host preference, they found that this strain is adapted
to maize and sorghum borers and is not attracted to even its preferred host, C. partel-
lus, when fed on sugarcane. The attraction is elicited by kairomones present in the
mandibular glands as well as frass of the host larvae.
Mohyuddin introduced sugarcane-adapted strains of C. flavipes from Indonesia and
elsewhere. Laboratory tests confirmed its preference for sugarcane, it was released and
quickly became established on the sugarcane borers Chilo infuscatellus and Acigona
steniellus. The various strains were crossed in the laboratory and interbred freely,
producing viable off-spring, confirming that they are the same species.
Such host plant preferences can be changed by laboratory selection. Parasitoid
'emales having preference for C. partellus in maize were reared on the sugarcane borer
7. infuscatellus in sugarcane. The females showing preference for sugarcane were
;elected, in an olfactometer, over successive generations and were used to produce the
iext generation. With the first generation, 6% of females showed preference for sugar-
!ane, 52% for maize and the rest showed no preference. Selection for sugarcane pro-
luced a 45% preference for sugar after 4 generations. Strains of C. flavipes from differ-
ant regions in Asia apparently have different responses to both stemborer hosts and
host plants of various pyralids and noctuids; a current IIBC-ICIPE program for C.
partellus in E. Africa will take this research further.
480 Florida Entomologist 75(4) December, 1992
In the Caribbean, where C. flavipes switched to Diatraea saccharalis, there may
well have been natural selection of both its host finding mechanism and its ability to
avoid encapsulation. It seems likely that such host switching is more common than
previously realized and there remains the tantalizing potential to modify the host pref-
erence and specificity of C. flavipes still further, to develop strains more effective
against D. centrella and other species. To do this however, we need a considerable
research effort and some insight into why C. flavipes is sometimes not so successful
(e.g. in Jamaica).
BIOCONTROL OF THE DIAMONDBACK MOTH, PLUTELLA XYLOSTELLA (L.)
This pest of truly global significance probably originated from the Mediterranean
region. It attacks crucifers which are generally temperate crops, and where cultivation
has spread to the tropics the moth has followed. One of the main problems to overcome
in controlling this pest is the very low economic threshold, which can be less than 1
larva per plant. This has meant heavy use of insecticides which has provoked severe
resistance problems. Bacillus thuringensis (BT) can be effective against the pest, but
resistance is appearing to this as well (e.g. in Hawaiian watercress cultivation (Biocon-
trol News & Information 1991)).
There have been some successes with the use of parasitoids. IIBC achieved almost
total control of the pest in the Cape Verde Islands with Cotesia plutellae. The introduc-
tion of Diadegma semiclausum into highland Brassica crops in Malaysia, Taiwan and
Indonesia has given excellent results, displacing C. plutellae and usually eliminating
the need for pesticides. Attainment of this control has been greatly slowed by continuing
pesticide use, and only the adoption of BT has permitted biocontrol to develop to its
full potential-in the Malaysian Cameron Highlands this took 12 years. Generally
though, in the hot humid tropics classical biocontrol of the moth has had limited success.
Both D. semiclausum and C. plutellae are not very effective under such extreme con-
ditions and the prevailing opinion is that future biocontrol of the moth will have to be
set firmly within a program for its integrated pest management. Pesticides, chemical
or biological, will be required to complement any improved use of indigenous or intro-
Because of the effect of pesticides on Plutella biocontrol, IIBC surveyed global
pesticide tolerance in C. plutellae and chose the most tolerant strain (from Malaysia)
for introduction into Honduras, where it is now believed to be established at low densi-
ties (R. Cave, pers. comm.). Some pesticide tolerance has also been found in the Carib-
bean, but this can not be called proper resistance. A recent attempt to create fenitroth-
ion-resistant C. plutellae at IIBC, achieved only a 2-3 fold increase in tolerance after
12 generations of selection from a field strain from a heavily sprayed region (Ke, Waage
& Moore, unpublished data).
The use of parasitoids in the genera Cotesia, Diadegma and Tetrastichus is receiving
renewed attention. A new Global Working Group on Plutella has been established in
the IOBC, and its priority will be to resolve the confused taxonomic status of these
parasitoids. Recent work by IIBC and the Escuela Agricola Panamericana in Honduras,
for example, has shown that D. semiclausum and the New World D. insulare will mate
but produce nearly all male progeny, underlining the need for careful study of para-
sitoids before introduction.
With the support of the IOBC, IIBC and the Asian Vegetable Research and Devel-
opment Centre are promoting a new exploratory program to seek strains and species
from the hot lowland regions of the pest's region of origin, to obtain strains more
tolerant of extreme tropical conditions. Thus although P. xylostella parasitoids are
established on a number of Caribbean islands, due largely to the indefatigable efforts
Biological Control Workshop-'91: Baker et al. 481
of Munir Alam of CARDI, much work remains to be done on realizing their full potential
through their integration with pesticide use and other control methods. Given the likeli-
hood of rapid development of resistance to IGRs and BT if used indiscriminately, this
research is urgent. There is also a need to look at new biological pesticides. IIBC, with
the assistance of CARDI has recently isolated three fungal agents from P. xylostella
in the Caribbean. IIBC will soon begin to investigate their use in formulations especially
designed for application to vegetables under hot dry conditions, using technology de-
veloped in IIBC's biopesticide program for the desert locust.
THE PIGEONPEA POD BORER, ANCYLOSTOMIA STERCOREA (ZELL.)
This borer is the most serious pest of pigeonpea in most Caribbean Islands; the
percentage of damaged peas frequently passes 50%. Pod life tables from cohorts of
marked buds show that major mortality occurs to potential pods during the flowering
stage, due probably to a physiological process. Nevertheless, appreciable mortality
occurs during the pod stage as well, though it is difficult to isolate the causes of this
mortality and the role that the borer plays in it, or conversely the role that pod drop
might play in borer mortality.
Bennett (1960) details six parasitoids attacking A. stercorea in Trinidad. Five are
larval, the sixth is egg-larval. Insecticides are not very effective because of the cryptic
nature of the pest. IIBC has been carrying out some basic studies to understand more
about the life history of the pest and its interactions with the parasitoid complex. Life
table studies suggest that this complex is not controlling the borer very efficiently; for
every hundred eggs laid, between 60 and 70 adults are produced. Mortality is relatively
uniform through the larval stages and Bracon thurberiphagae appears to be the most
important biotic factor.
The problem is similar to that of sugarcane, an indigenous borer on an introduced
plant which has a natural enemy complex that is not controlling the pest. Innundation
by Trichogramma is unlikely to be effective because some property of the eggs seems
to deter attack. Chemical control is difficult, cultural control not effective. The obvious
route to follow is to look for parasitoids on closely related species in other countries;
the life table studies suggest there is plenty of room for another parasitoid to cause
extra mortality during larval and pupal development. With the experience gained from
strain selection of C. flavipes, we feel that future research on this problem is a worth-
There are other control possibilities however where biocontrol should not be forgot-
ten. This is in the area of plant breeding to produce resistant strains, or possibly strains
that shed fewer flowers or varieties with an extended productive season. Let us look
at these possibilities in turn.
Resistant strains might be developed to mechanically hinder entry of the young
larva, or more likely, to augmenting deterrent chemicals or decrease moth attractants.
This last mentioned factor is apparently the case for the differential attraction of
Helicoverpa armigera attraction to cultivars of pigeonpea (Rembold & Tober 1985). It
is very likely that the parasitoids are also attracted by plant substances themselves or
plant chemicals that have been altered or released by the feeding of the moth larvae.
(Turlings & Tumlinson 1991). In any resistant variety trials, it would be very important
to make sure that such synomones were still manufactured by the plant so that the
control by natural enemies would not decline on the introduction of a new variety.
If a strain could be produced to augment numbers of pods surviving to maturity,
would this be of use or would the moth population merely increase in proportion? The
available evidence suggests that moth population increase would be unlikely because of
the density dependent mode of action of the natural enemies. From basic life-table data,
Florida Entomologist 75(4)
it should be possible to model the effect that increased pod production would have on
If a strain of pigeonpea had a longer flowering period, the effect that this would
have on moth populations could also be modelled. Available evidence suggests that the
parasitoids are building up too late in the season to have a marked effect on moth
populations, so a prolonged season could be beneficial to parasitoid populations.
A possible scenario for future control of this pest might be: (a) Short term: use
insecticides which have been screened for minimal effect on the parasitoid complex
(already being undertaken). (b) Mid-term: look for foreign parasitoids, screen them in
an olfactometer or wind-tunnel against borers on different varieties of the plant; if they
are effective, release them and study them through life table analysis. (c) Long term:
breed resistant varieties, including testing for attractiveness to parasitoids.
There have been many attempts to analyze biocontrol projects to discover basic
principles governing success and failure. We do not propose to repeat these exercises,
only to point to some areas we think are important.
From the history of biological control in the Caribbean (Table 1) it is clear that
"small" is not beautiful. The greatest success, in sugarcane, has come from an outstand-
ing effort over many years. Although there are many examples throughout the history
of biocontrol where a few releases of a poorly studied parasitoid taken from one locality
have been successful, the history of biocontrol of lepidopteran pests in the Caribbean
belies this approach. It is no accident that most progress has been in the most important
cash crop of the region. Even here however, the reasons for success are not readily
apparent. Recent evidence (Pashley et al. 1990) suggesting that D. saccharalis may be
two sibling species should surely give biocontrol workers cause for thought. Is this why
C. flavipes has been so successful in some cases but not others? Or is it due to liability
of the virus which the parasitoid female injects with her eggs to suppress the immune
system? Biocontrol workers usually do not have the time or resources to consider these
points and consequently the problem continues to be addressed by releases of various
indistinguishable parasitoid "strains" with the hope that one will overcome the problem.
The situation is analogous to insecticide companies which used to select chemicals by
trial and error without trying to understand the physiology of the insect. They have
changed their strategy; so should we.
The greatest challenge for small Caribbean islands is to make biocontrol work on
small scale horticultural production. Fletcher et al. (1989) have shown that farmers in
Trinidad get most of their advice from chemical companies. Extension services are
limited and poorly financed on many islands and although pesticide regulations exist,
there is often no means of enforcing them efficiently. In recent years CARDI has had
notable successes with small farmer systems, though whether these can be sustained
once support is removed has yet to be ascertained. With the many other socioeconomic
problems afflicting such countries, pesticide abuse is viewed by some as a relatively
minor problem and governments often lack the commitment to punish offenders. The
only way that things are likely to change is if export commodities are refused entry to
the US and elsewhere because of pesticide contamination. However in many cases
vegetable production is mainly for local or regional consumption and there are no strong
reasons why current practices should change markedly. Perhaps biopesticides will re-
place chemicals but it is likely that these too will be overused and induce resistance.
Future biocontrol ventures are likely to be most successful where there is strong
export potential. With the decline of sugar and the concomitant diversification into
other crops, there is a danger that research efforts will fragment into many small
Biological Control Workshop-'91: Baker et al. 483
projects that may deal with serious pest problems but will be too small to be able to
pay for development of a sound biocontrol solution.
There is a need for a regional approach to rationalization of agricultural production
so that long-term biocontrol and IPM solutions can be developed and implemented with
sufficient funds. Rice in Guyana is an obvious candidate for biocontrol, fruit crops such
as mango and citrus are another, drought-resistant crops such as pigeonpea could be a
third. We should make strenuous efforts to coordinate the activities of the many re-
search organizations in the region and to identify a few key pests on which to focus our
attention; only in this way will we maximize chances of furthering the cause of biological
From the technical viewpoint, future projects should involve (1) extensive searches
in countries of origin for natural enemies; (2) considerable effort to taxonomically charac-
terize the material obtained and (3) after quarantine, a comprehensive breeding and
testing phase which would involve olfactometers or a wind-tunnel to select and behavior-
ally classify candidate strains. Such work would be expensive but the increased level
of success should make it cost effective.
At the very least, even if a project is not successful, we should be able to provide a
scientifically argued reason for its failure so that we do not continue to repeat the
mistakes of the past.
ALAM, M. 1992. Use of parasitoids for the biological control of cabbage pests in the
Caribbean. Florida Ent. 75: 493-505.
BATES, J. F. 1954. The status of the moth borer in British Guyana. Proc. British West
Indies Sugarcane Technol. 1954, 126-136.
BENNETT, F. D. 1960. Parasites of Ancylostomia stercorea (Zell.), (Pyralidae,
Lepidoptera) a pod borer attacking pigeon pea in Trinidad. Bull. Ent. Res. 50:
BIOCONTROL NEWS AND INFORMATION. 1991. Conference report. 12(1): 6.
COCK, M. J. W. 1985. A review of biological control of pests in the Commonwealth
Caribbean and Bermuda up to 1982. CIBC Tech. Comm. 9. CAB, UK.
FLETCHER, L. M., R. H. SINGH, AND M. FARROE. 1989. The role of agricultural
input suppliers in the transfer of agricultural technology in Trinidad and Tobago.
2nd Annual Seminar on Agricultural Research, Trinidad and Tobago November
METCALFE, J. R., AND J. BRENIERE. 1969. Egg parasites (Trichogramma spp.) for
control of sugar cane moth borers, pp. 80-115 in J. R. Williams, J. R. Metcalfe,
R. W. Mungomery, and R. Mathes [eds.], Pests of sugarcane. Elsevier, Amster-
dam, London, New York.
MOHYUDDIN, A. I. 1971. Comparative biology and ecology of Apanteles flavipes
(Cameron) and Apanteles sesamiae Cam. as parasites of graminaceous borers.
Bull. Ent. Res. 61: 33-39.
MOHYUDDIN, A. I., C. INAYATULLAH, AND E. G. KING. 1981. Host selection and
strain occurrence in Apanteles flavipes (Cameron) (Hymenoptera:Braconidae)
and its bearing on biological control of graminaceous stem-borers (Lepidopt-
era:Pyralidae) Bull. Ent. Res. 71: 575-581.
PASHLEY, D. P., T. N. HARDY, A. M. HAMMOND, AND J. A. MIHM. 1990. Genetic
evidence for sibling species within the sugarcane borer (Lepidoptera: Pyralidae).
Ann. Entomol. Soc. Am. 83(6): 1048-1053.
REMBOLD, H., AND H. TOBER. 1985. Kairomones as pigeonpea resistance factors
against Heliothis armigera. Insect Sci. Applic. 6: 249-252.
TURLINGS, T. C. J., AND J. H. TUMLINSON. 1991. Do parasitoids use herbivore-in-
duced plant chemical defenses to locate hosts? Florida Ent. 74(1): 42-50.
Florida Entomologist 75(4)
APPLICATION OF BIOTECHNOLOGY FOR IMPROVEMENT
OF BACILLUS THURINGIENSIS BASED PRODUCTS AND
THEIR USE FOR CONTROL OF LEPIDOPTERAN PESTS
IN THE CARIBBEAN
WENDY D. GELERNTER
San Diego, CA 92121 USA
Increased use of products based on Bacillus thuringiensis (Bt) in the Caribbean has
led to increased awareness of the advantages, as well as the limitations of these biolog-
ical insecticides. To improve the performance of Bt based products, recombinant DNA
technology has been utilized to transfer the Bt delta endotoxin gene to microbes and
plants. When the Bt protein is expressed in these recombinant organisms, improved
delivery, persistence and insecticidal activity have been demonstrated. In this paper,
the status of current research efforts to improve Bt will be reviewed, and future pros-
pects for the development of products based on genetically engineered organisms will
El incremento en el uso de products con Bacillus thuringiensis (Bt) en el Caribe,
ha conducido a un incremento de las ventajas y desventajas de estos insecticides
biologicos. Para mejorar la actuacion de los products con base de Bt, la tecnologia de
DNA recombinado ha sido utilizada con el fin de transferir el gene de la endotoxina Bt
delta a microbios y plants. Cuando la protein BT se expresa en estos organismos
recombinados, se ha demostrado un aumento de persistencia y actividad insecticide.
En este manuscrito, se discuten y revisan el estado de los esfuerzos de investigation
actuales para mejorar Bt y se discuten los futures projects para el desarrolllo de
products basados en organismos mejorados geneticamente.
The use of insecticide products based on the naturally occurring bacterium, Bacillus
thuringiensis (Bt) has increased rapidly during the last decade in the Caribbean region
and worldwide. This is largely due to international demand for products, such as those
based on Bt, that are effective but are also safe for consumers and workers, that com-
plement rather than destroy natural enemy complexes, and that effectively control
insects that have developed resistance to synthetic chemical insecticides. This demand
has accelerated the adoption of Bacillus thuringiensis products, particularly in high
value or environmentally sensitive "niche" markets such fresh market vegetables or
forestry. Yet Bt based products also have several important limitations including a
highly specific host range, short residual activity, and a unique mode of action which
requires ingestion by the target insect. These limitations appear to outweigh the advan-
tages of Bt in the majority of markets-particularly where profitability is lower and/or
there is less demand for environmentally compatible products. To expand the use of Bt
based products in agriculture, several projects are currently underway that utilize re-
combinant DNA technology, or biotechnology to improve the delivery, persistence and
insecticidal activity of naturally occurring Bt.
In this paper, Bt biology and current use patterns for Bt based products in the
Caribbean will be described. In addition, a review of progress in the development of
Biological Control Workshop-'91: Gelernter 485
| TOXIN GENE
Fig. 1. Schematic diagram of a mature Bacillus thuringiensis cell. Contents of the
cell are not drawn to scale.
improved recombinant Bt products, and a discussion of issues surrounding their use in
the Caribbean against lepidopteran pests will be presented.
BIOLOGY OF BACILLUS THURINGIENSIS
Bacillus thuringiensis is a naturally occurring, Gram positive bacterium, which is
most frequently found in soil. Upon maturation, each Bt cell produces a spore and a
proteinaceous crystal or crystals that are stomach poisons for specific insects (Fig. 1).
When Bt is commercially produced, it is grown in submerged liquid culture in large-scale
fermentors where, at the completion of its growth cycle, the cells lyse or burst and
release spores and protein crystals into the liquid medium. These spores and crystals
are harvested and concentrated, and serve as the active ingredient of commercial Bt
formulations (Rowe & Margaritis 1987).
In most interactions between Bt and insects, it is the protein crystal which is respon-
sible for Bt insecticidal activity. Classified as stomach poisons, these Bt proteins or delta
endotoxins are highly potent, with as little as 25 grams of protein required per hectare
to achieve acceptable levels of insect control (Mycogen Corporation, unpublished data).
When insects ingest foliage treated with Bt, the delta endotoxin crystal is rapidly dissol-
ved in the insect midgut and delta endotoxin molecules bind to specific receptors on the
microvillar membranes of midgut epithelial cells. The first gross symptom observed is
an almost immediate (within one hour) feeding inhibition response due to paralysis of
the gut. Susceptible insects rarely recover their appetites, and starvation is certainly
a contributing factor in Bt induced mortality. However, it is the gradual disintegration
of the midgut epithelium, followed by a lethal mixing of hemocoel and gut contents, that
is ultimately responsible for the death of the target insect, usually 1-7 days after inges-
tion. (Heimpel & Angus 1959, Knowles & Ellar 1987). Although ingestion of the delta
endotoxin crystals is a requirement for insect mortality, the presence of Bt spores may
also be required for death to occur in certain insects (Heimpel & Angus 1959).
When Bt was first described in 1901, it was isolated from diseased silkworm (Bombyx
mori) larvae. Since that time, thousands of Bt isolates have been described, with the
majority having specific activity towards lepidopteran larvae such as the cabbage looper
Florida Entomologist 75(4)
(Trichoplusia ni), the diamondback moth (Plutella xylostella) and the imported cab-
bageworm (Pieris rapae). All Bt isolates are currently grouped into over 30 different
varieties in a classification system based on serotyping of Bt flagellar proteins, insect
host range, and biochemistry (Dulmage 1982). Commercial products available today are
most commonly based on the lepidopteran active Bt variety kurstaki, although products
based on the mosquitocidal Bt variety israelensis and on the beetle active Bt varieties
san diego and tenebrionis are also available worldwide. In recent years, Bt isolates with
specific activity for unique hosts such as nematodes (Edwards et al. 1990) have been
described, and there is every reason to believe that many more isolates, with even more
diverse activities, remain to be discovered.
An important feature of each Bt isolate is its relatively narrow and specific host
range. Thus, Bt variety kurstaki is highly active against larvae of several noctuid pests
including the cabbage looper, but has very limited activity against the closely related
corn earworm, Helicoverpa zea. By the same token, Bt varieties san diego and teneb-
rionis are quite active against chrysomelid beetles such as the Colorado potato beetle,
Leptinotarsa decemlineata, but have no effect on other chrysomelids such as the corn
rootworm, Diabrotica longicornis. The basis for the specificity exhibited by each Bt
isolate has long eluded insect pathologists. The most recent hypotheses rely primarily
on the structure of the delta endotoxin and its relationship to the type and number of
specific binding sites on the surface of the insect midgut epithelium. Other factors
involved in Bt specificity include the internal environment of the insect gut, which may
influence the activation of Bt protein crystals to delta endotoxin molecules (Johnson et
al. 1990), as well as the interactions between multiple delta endotoxin molecules and
between the delta endotoxin and Bt spores (Heimpel & Angus 1959).
USE OF COMMERCIAL BT PRODUCTS IN THE CARIBBEAN
Commercial formulations of Bacillus thuringiensis have been available since the
1950's, and represent the most successfully commercialized group of biological insec-
ticides available today. Relatively high levels of potency for specific pest insects, coupled
with environmental, mammalian and non-target safety, have recently led to dramatic
increases in sales of Bt based products throughout the world (McKemy 1990) .
To better understand specific trends in the use of Bt in the Caribbean region, a
survey was developed and distributed to 53 key researchers, growers, distributors, and
extension agents in 15 countries. Twenty six responses were received (from Belize,
Columbia, Costa Rica, Dominican Republic, Guatemala, Haiti, Honduras, Jamaica, Mar-
tinique, Panama, Puerto Rico, Trinidad, Venezuela, Virgin Islands and South Florida
U.S.), and form the basis of the information presented in this section.
The large majority of Bt products utilized in the Caribbean are targeted for lepidop-
terous pests, and are based on Bt variety kurstaki according to survey participants.
Products most commonly marketed in the Caribbean include Dipel (Abbott Labs),
Thuricide and Javelin (Sandoz Corporation), and Bactospeine (Duphar). Among sur-
vey participants, there was little or no familiarity with mosquito and black fly active
products such as Vectobac (Abbott Laboratories), Skeetal (Novo Laboratories), and
Teknar (Sandoz Corporation) or with beetle specific products such as M-One (Myco-
gen Corporation) and Trident (Sandoz Corporation).
Survey participants indicated that of all biological control methods available (includ-
ing fungi, nematodes, beneficial insects, and baculoviruses), Bt based products had the
most potential for success in the Caribbean. In their opinion, the most attractive fea-
tures of Bt include (in priority order):
1) can be used to avoid development of resistance to synthetic insecticides
2) increased food, worker and environmental safety
Biological Control Workshop-'91: Gelernter
TABLE 1. CARIBBEAN CROPS AND INSECTS, MOST COMMONLY TREATED WITH BT
BASED PRODUCTS. INSECTS ARE RANKED ACCORDING TO THE FRE-
QUENCY WITH WHICH THEY ARE TREATED WITH BT BASED PRODUCTS
BY SURVEY PARTICIPANTS.
Crop Insect (common name) Insect (scientific name)
cucurbit crops melonworm Diaphania hyalinata
(melon, cucumber, armyworms Spodoptera spp.
squash pumpkin) squash vine borer Melittia cucurbitae
cabbage looper Trichoplusia ni
Heliothis, Helicoverpa spp. Heliothis, Helicoverpa spp.
cole crops (cabbage, diamondback moth Plutella xylostella
broccoli, cauliflower, cabbage looper Trichoplusia ni
greens, Chinese imported cabbageworm Pieris rapae
vegetables) beet armyworm Spodoptera exigua
fall armyworm Spodopteraffrugiperda
tomatoes tomato fruitworm Helicoverpa zea
armyworms Spodoptera spp.
leaf caterpillar Mocis spp.
cabbage looper Trichoplusia ni
corn corn earworm Helicoverpa zea
fall armyworm Spodopterafrugiperda
beet armyworm Spodoptera exigua
ornamentals banana moth Opogona sacchari
cassava cassava horn worm Erinnyis ello
3) lack of toxic pesticide residues
4) can be used to preserve beneficial insects
Consumer demand and restrictive government regulations were not considered impor-
tant factors in the adoption of Bt products.
When asked to list insects and crops where Bt based products are used, insects
resistant to synthetic chemical insecticides (e.g., the diamondback moth, the beet ar-
myworm, Spodoptera exigua, the cabbage looper and Heliothis and Helicoverpa species)
on fresh market produce were most frequently mentioned (Table 1). It is interesting to
note that although these resistant pests occur on a variety of crops including cotton,
sorghum and grains, Bt products were most frequently used on higher value crops such
as fresh market produce and ornamentals.
Survey participants cited several reasons why Bt based products are not more widely
adopted in the Caribbean including (in priority order):
1) narrow host range
2) relatively high cost of Bt based products compared to conventional insecticides
3) lack of efficacy compared to chemical insecticides
4) increased management inputs (scouting to determine correct timing of applica-
tions, improved application equipment to provide better foliar coverage) re-
5) short residual activity
6) inability to target internally feeding (cryptic) insects
Florida Entomologist 75(4)
APPLICATION OF BIOTECHNOLOGY IN THE IMPROVEMENT OF
BT BASED PRODUCTS
To resolve problems such as the above which are associated with the use of Bt based
insecticides, several research programs have focused on the use of recombinant DNA
technology, or biotechnology. For most isolates of Bt, the delta endotoxin gene is coded
for on extrachromosomal plasmids, making it possible for genetic engineers to locate,
manipulate and transfer the gene to other organisms including bacteria, viruses and
plants. In this way, scientists hope to improve Bt delivery, persistence and insecticidal
activity (Table 2).
Because Bt based products must be ingested to be effective, insects that feed inter-
nally within the plant or the roots cannot be controlled by conventional Bt foliar appli-
cations. In addition, even those insects which feed on the outer plant surface are often
difficult to control well because precise application timing (when the majority of the
insects are newly hatched and therefore most sensitive) and excellent foliar coverage
are required for optimum control with Bt products. Delivery of delta endotoxins can be
improved, however, if microorganisms associated with the plant, or the plant itself, are
engineered to produce Bt endotoxins. For example, larvae of the European corn borer,
Ostrinia nubilalis, most commonly feed within the stalk, the ear or the stem of corn
and other plants. Although these insects are quite susceptible to Bt variety kurstaki,
they are impossible to target with foliar applications of Bt once they have bored inside
the plant. To improve delivery of Bt toxins inside the plant, Crop Genetics International
Corporation has developed the InCideT biopesticide technology which is based on the
endophytic (vascular system colonizing) bacterium, Clavibacter xyli subsp. cynodontis;
this microorganism has been engineered to produce a Bt variety kurstaki endotoxin.
When corn seed is treated with the engineered C. xyli, the bacteria multiply and pro-
duce the delta endotoxin inside the growing corn plant. Several outdoor field trials have
been conducted with the InCide technology which demonstrate that corn plants inocu-
lated with the recombinant C. xyli had significantly less damage by the European corn
borer than untreated plants. Expanded field tests and attempts to increase delta endo-
toxin expression levels in C. xyli are continuing during 1991 (Beach 1990). By geneti-
cally modifying root colonizing bacteria such as Pseudomonas fluorescens to produce
Bt variety kurstaki endotoxin, Monsanto scientists have developed a different delivery
system-one that targets soil insects such as cutworms (Obucowicz et al. 1987). In
another example of improved delivery systems, companies such as Monsanto and Plant
Genetic Systems are currently developing recombinant plants that have been en-
gineered to produce various Bt endotoxins in plant tissues. Successful outdoor field
tests with transgenic tobacco and tomatoes have effectively targeted such cryptic pests
as tomato pinworm, Keiferia lycopersicella and tobacco budworm, Heliothis virescens
(Fischoff e. al. 1987, Gasser & Fraley 1989, Honee et al. 1989, Vaeck et al. 1987).
For conventional Bt based products, the active ingredient or delta endotoxin is
present as a "naked" or unprotected protein crystal. Not surprisingly, the endotoxin
protein crystal breaks down rapidly on the surface of the crop plant-usually within 1-2
days-as the result of harsh or inactivating light, microbial and plant enzymes, pH
extremes, moisture and high temperatures (Gelernter 1990). Because of their short
residual activity, Bt products must be applied more frequently and with greater atten-
m- > 0 m
43) ^ mD (1)i ce f '
0 41 c 3
V ~ ~ ~D
E ~ =, =
_3 |.3 z.
a) t )
CO m <
-S 2 g.
II 11111 E
N.- O oO 0"
si )iM. ~
*P ^ s-
S- s -te
S gg Mo-
0*'c ce gi
0- S I-
(- S c
0 o s Q|-
0^ q *
*+ t <
E:^ ) ,
.3 6(1`m~ ,
t4- *,,^ a
Biological Control Workshop-'91: Gelernter
"- -4 .
so- 'a u
Florida Entomologist 75(4)
tion to insect developmental stage than conventional synthetic chemical products. Sev-
eral of the technologies described above, including development of Bt expressing plants
and plant colonizing bacteria, address the issue of improved persistence as well as
enhanced delivery of delta endotoxins. A somewhat different approach, the CellCap"T
encapsulation system developed by Mycogen Corporation, utilizes dead bacterial cell
walls to protect delta endotoxins from environmental degradation. In this system, Bt
delta endotoxin genes are transferred from Bt to a non pathogenic isolate of
Pseudomonas fluorescens. When the recombinant cells are grown in fermentors, the
endotoxin is produced and forms a crystal within the cell. Unlike Bt, the recombinant
Pseudomonas cells do not break apart at the completion of the growth cycle. Instead,
while still in the fermentation tank, the intact cells are subjected to a chemical treatment
which kills and fixes the Pseudomonas cell wall, causing it to become thicker and more
rigid through cross linking of cell wall components. The rigid wall of the dead
Pseudomonas cell then serves as a protective biological microcapsule for the enclosed
toxin (Barnes & Cummings 1987). Replicated small plot and large scale grower trials
conducted over the past three years have demonstrated that the CellCap encapsulation
system results in a two fold improvement in residual activity of the Bt delta endotoxins,
compared to that for the nonencapsulated endotoxins present in conventional Bt formu-
lations. Mycogen has developed several products based on the CellCap encapsulation
system including MVP Bioinsecticide, based on an encapsulated delta endotoxin from
Bt variety kurstaki and targeted against caterpillar larvae, and M-TrakTM Bioinsecticide
which controls beetle pests such as the Colorado potato beetle and is based on an
encapsulated delta endotoxin from Bt variety san diego. Because these engineered
products contain only dead microorganisms, concerns regarding the release and spread
of living recombinants are not considered pertinent by regulatory agencies. In 1991,
MVP and M-Trak became the first genetically engineered biopesticides to be registered
for use by the United States Environmental Protection Agency.
Improved Insecticidal Activity:
Bt based products are most effective against newly hatched and small larvae. How-
ever, as insects mature, they grow increasingly more able to tolerate or overcome
intoxication by Bt (Zehnder & Gelernter 1989, Bauer 1990). Insect specific
baculoviruses, on the other hand, are capable of killing large lepidopteran larvae but
their ingestion does not result in the same feeding inhibition response as Bt; this can
result in insect feeding damage to the crop in the 2 to 10 days elapsing between virus
ingestion and mortality. By developing recombinant baculoviruses that have been en-
gineered to produce Bt delta endotoxin genes, it might be possible to achieve the best
of both worlds-the efficacy of baculoviruses against large larvae, and the effects of Bt
induced feeding inhibition and midgut disruption. Two research groups (Martens et al.
1990, Merryweather et al. 1990) have recently successfully engineered the Autographa
californica nuclear polyhedrosis virus to produce Bt delta endotoxins from Bt varieties
aizawai and Bt kurstaki. However, protein expression levels in the recombinant viruses
are not optimized at this early stage in development, and it therefore still too early to
accurately assess their efficacy.
By transferring the delta endotoxin gene from the caterpillar active Bt variety
aizawai into the beetle active Bt variety tenebrionis, Crickmore et al. (1990) have
demonstrated improved activity on both caterpillar, mosquito and beetle pests with the
transgenic bacterium, indicating that the delta endotoxin proteins may interact syner-
Biological Control Workshop-'91: Gelernter
gistically. Similarly, Crickmore et al. (1990) have transferred the Bt variety tenebrionis
endotoxin gene into mosquito active Bt variety israelensis, producing cells that have
beetle and mosquito activity, as well as unexpected lepidopteran activity. Carlton et al.
(1990) have taken a different approach towards improving the host range of Bt isolates
where recombinant DNA technology is not utilized. In this case, the natural process of
bacterial conjugation is utilized to transfer an entire Bt plasmid from one Bt to another.
The resulting "genetically manipulated" or "transconjugant" organism now produces
two different delta endotoxins-one from each "parent". This technique has allowed the
biopesticide company Ecogen to successfully develop products with expanded host
ranges. For example, the product Foil is based upon a transconjugant Bt strain that
produces two delta endotoxin crystals; one beetle active, and the other caterpillar active
(Carlton et al. 1990).
ISSUES SURROUNDING THE USE OF BIOTECHNOLOGY FOR
IMPROVING BT BASED PRODUCTS
Recombinant DNA technology allows scientists to develop significantly improved
biological insecticides. With this technology, the quality, quantity and diversity of effica-
cious biological insecticides available should dramatically increase over the next 20
years. Yet there are several issues which deserve continuing scrutiny and discussion
as we progress towards this goal. These include:
Setting Realistic Expectations On Timing Of New Product Introductions
Although great progress has been made in the past five years towards development
of improved Bt based insecticides, there are several technical and regulatory hurdles
which will limit the number of genetically engineered products commercialized over the
five to ten years. For example, many of the projects described above have not yet
produced organisms with sufficiently high yields of delta endotoxin, and field efficacy
has been demonstrated in only a few cases. Additionally, the United States and most
European countries have not yet registered any products based on living recombinant
microorganisms, and have only recently begun to develop guidelines for their testing
and registration. Until these hurdles are overcome, the result will be a relatively slower
paced, more gradual schedule of product introductions during the next several years.
Setting Realistic Expectations On Performance Of New Products
Too often, new technologies are positioned as a panacea. In the case of agricultural
"biotech" products, it is important to establish that the performance of biological pesti-
cides, whether they are engineered or not, will probably never quite equal the broad
spectrum, fast acting toxicity of synthetic chemical insecticides. By their very nature,
biopesticides will be more selective, and more sensitive to environmental conditions
than their synthetic chemical counterparts. Thus, growers must still be trained in the
proper use of these new products and must understand that increased management and
scouting inputs are required for best results.
The tendency to over-use new, effective products has led to many incidences of
insect resistance to formerly effective chemical products (Roush & Tabashnik 1990).
This unfortunate trend was recently documented for certain conventional Bt products
(Shelton & Wyman 1992, Tabashnik et al. 1991) which are used up to 25 times per year
Florida Entomologist 75(4)
for control of insecticide resistant diamondback moth larvae. Under this type of selection
pressure, diamondback moth tolerance or resistance to Bt variety kurstaki has now
been documented in at least two states within the U.S. (Florida and Hawaii). It is
probable that Bt toxins will become more widely distributed and utilized in the near
future. This, combined with the fact that persistence of toxins expressed in living recom-
binant plants or microorganisms is greatly enhanced, will increase the likelihood of the
development of Bt resistance. Efforts to better understand the basis of Bt activity and
resistance, and development of programs which incorporate other cultural and biological
control measures into programs that currently rely heavily on Bt based products are
essential if we are committed to preserving the use of Bt delta endotoxins for the future.
The information supplied by the 26 participants in the Caribbean Basin Biological
Control Survey is gratefully acknowledged.
BARNES, A. C., AND S. E. CUMMINGS. 1987. U.S. Patent No. 4695455.
BAUER, L. S. 1990. Response of the cottonwood leaf beetle to Bacillus thuringiensis
var. san diego. Environ. Entomol. 19: 428-431.
BEACH, R. M. 1990. Application for an experimental use permit to ship and use a
pesticide for experimental purposes only. Permit number 58788-EUP-4 for In-
CideT 586. Crop Genetics International, Hanover Maryland.
CARLTON, B. C., C. GAWRON-BURKE, AND T. B. JOHNSON. 1990. Exploiting the
genetic diversity of Bacillus thuringiensis for the creation of new bioinsecticides.
pp. 18-22 in Proceedings, Fifth International Colloquium on Invertebrate Pathol-
ogy and Microbial Control, August 20-24, 1990, Adelaide, Australia. Society for
CRICKMMORE, N., C. NICHOLLS, D. J. EARP, T. C. HODGMAN, AND D. J. ELLAR.
1990. The construction of Bacillus thuringiensis strains expressing novel en-
tomocidal delta endotoxin combinations. Biochem. J. 270: 133-136.
DULMAGE, H. T. 1982. Distribution of Bacillus thuringiensis in nature. pp. 209-237
in E. Kurstak [ed.], Microbial and Viral Pesticides. Marcel Dekker, Inc., New
EDWARDS, D. L., J. PAYNE, AND G. G. SOARES. 1990. U.S. Patent no. 4,948,734.
FISCHOFF, D. A., K. S. BOWDISCH, F. J. PERLAK, P. G. MARRONE, S. H. MCCOR-
MICK, J. G. NIEDERMEYER, D. A. DEAN, K. KUSANO-KRETZMER, E. J.
MAYER, D. E. ROCHESTER, S. G. ROGERS, AND R. T. FRALEY. 1987. Insect
tolerant transgenic tomato plants. Bio/Technology 5: 807-813.
GASSER, C. S., AND R. T. FRALEY. 1989. Genetically engineering plants for crop
improvement. Science 244: 1293-1299.
GELERNTER, W. D. 1990. Targeting insecticide-resistant markets: new developments
in microbial based products, pp. 105-117 in M. B. Green, W. K. Moberg and H.
LeBaron [eds], Managing Resistance to Agrochemicals: From Fundamental Re-
search to Practical Strategies. Series 4221. Washington, D.C. American Chemi-
HEIMPEL, A. M., AND T. A. ANGUS. 1959. The site of action of crystalliferous bacteria
in Lepidoptera larvae. J. Insect Pathol. 1: 152-170.
HONEE, G., T. VAN DER SALM, AND B. VISSER. 1989. Insect resistant transgenic
tomato plants. J. Cell Biochem Suppl. 0 (13 Part D).
JOHNSON, D. E., G. L. BROOKHART, K. J. KRAMER, B. D. BARNETT, AND W. H.
MCGAUGHEY. 1990. Resistance to Bacillus thuringiensis by the Indian meal
moth, Plodia interpunctella: a comparison of midgut proteinases from susceptible
and resistant larvae. J. Invert. Pathol. 55: 235-244.