Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00064
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1991
Copyright Date: 1917
Subject: Florida Entomological Society
Entomology -- Periodicals
Insects -- Florida
Insects -- Florida -- Periodicals
Insects -- Periodicals
General Note: Eigenfactor: Florida Entomologist: http://www.bioone.org/doi/full/10.1653/024.092.0401
 Record Information
Bibliographic ID: UF00098813
Volume ID: VID00064
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

(ISSN 0015-4040)


(An International Journal for the Americas)

Volume 74, No. 3 September, 1991


Hematophagous Strategies of the Cat Fleas (Siphonaptera: Pulicidae) .... 377
PILGRIM, R.L.C.-External Morphology of Flea Larvae (Siphonaptera) and Its
Significance in Taxonomy .......................................................... 386
Research Reports
MCPHERSON, J. E., S. L. KEFFER, AND S. J. TAYLOR-Taxonomic Status of
Melanolestes picipes and M. abdominalis ........................................... 396
WILLIAMS, L.A. D.-Acaricidal Activity ofFive Marine Algae Extracts on Female
Boophilus microplus (Acari: Ixodidae) .............................................. 404
son of Field Observations and Trapping of Papaya Fruit Fly in Papaya
Planting in Central America and Florida. ....................................... 408
PRICE, J. F., AND J. B. KRING-Dyscinetus morator (Coleoptera: Scarabaeidae)
Flight Activity, Food Plant Acceptance, Damage and Control in Caladium
............................... ......... .............................................. 4 15
SCHUSTER, J. C.-Petrejoides (Coleoptera: Passalidae): Four New Species from
Mesoamerica and Mexico with a Key to the Genus. ............................. 422
VAN SAUERS-MULLER, A.-An Overview of the Carambola Fruit Fly, Bactrocera
Species (Diptera: Tephritidae), Found Recently in Suriname ............... 432
WALTER, D. E., AND H. A. DENMARK-Use of Leaf Domatia on Wild Grape
(Vitis munsoniana) by Arthropods in Central Florida .......................... 440
BRUSHWEIN, J. R. AND J. D. CULIN-Modified Rearing and Maintenance Tech-
niques for Mantispa viridis (Neuroptera: Mantispidae) ....................... 446
GALLIART, P. L., AND K. C. SHAw-Role of Weight and Acoustic Parameters,
Including Nature of Chorusing, in the Mating Success of Males of the
Katydid, Amblycorypha parvipennis (Orthoptera: Tettigoniidae) ............ 453

Scientific Notes
GONZALEZ, J. M., AND J. E. LATTKE-A Paper Wasp Nest Inside A Plant
of Brocchinia hectioides (Bromeliaceae) ................................... 465
for Toxotrypana curvicauda (Diptera: Tephritidae) ..................... 466
Glass and Plastic McPhail Traps in the Capture of the South Amer-
ican Fruit Fly, Anastrepha fraterculus (Diptera: Tephritidae) in
B razil ............................................................................... 467
MASON, L. J., AND R. K. JANSSON-Disruption of Sex Pheromone Com-
munication in Cylas formicarius (Coleoptera: Apionidae) as a Poten-
tial M eans of Control ......................................................... 469

Continued on Back Cover

Published by The Florida Entomological Society

President ................................................................................. J. L. Knapp
President-Elect .................................................................... D. F. W illiams
Vice-President ............................................................................. J. E. Pefia
Secretary .................................................................................... D G H all
Treasurer ................................................................................ A C. Knapp
Other Members of the Executive Committee
J. E. Eger J. A. Hogsette J. R. Cassani J. H. Frank
J. R. McLaughlin S. Valles M. Lara

J. R. McLaughlin, USDA/ARS, Gainesville, FL ....................................... Editor
Associate Editors
Agricultural, Extension, & Regulatory Entomology
Ronald H. Cherry-Everglades Research & Education Center, Belle Glade, FL
Michael G. Waldvogel-North Carolina State University, Raleigh, NC
Stephen B. Bambara-North Carolina State University, Releigh, NC
Biological Control & Pathology
Ronald M. Weseloh-Connecticut Agricultural Experiment Sta., New Haven, CT
Book Reviews
J. Howard Frank-University of Florida, Gainesville
Chemical Ecology, Physiology, Biochemistry
Louis B. Bjostad-Colorado State University, Fort Collins, CO
Ecology & Behavior
John H. Brower-Stored Product Insects Research Laboratory, Savannah GA
Theodore E. Burk-Dept. of Biology, Creighton University, Omaha, NE
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. Pefia-Tropical Research and Education Center, Homestead, FL
Systematics, Morphology, and Evolution
Michael D. Hubbard-Florida A&M University, Tallahassee
Howard V. Weems, Jr.-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 September 30, 1991

Hinkle et al.: Cat Flea (Siphonaptera: Pulicidae) 377



Department of Entomology and Nematology
Institute of Food and Agricultural Sciences
University of Florida
Gainesville, Florida 32611

U.S. Dept. of Agriculture
Agricultural Research Service
Medical and Veterinary Entomology Research Laboratory
Gainesville, FL 32608


Hematophagy of the cat flea, Ctenocephalidesfelisfelis (Bouche), was investigated.
Blood feeding in the adult stages nearly doubled the weight of mixed-sex fleas. However,
within 12 h, the gained weight was lost. Protein mass tripled after feeding, but starvation
caused a reduction in protein with the percentage protein remaining constant (5%).
Both in vivo and in vitro rearing of cat fleas was successful in allowing flea survival,
feeding, fecal production, and reproduction. In vivo rearing, infesting cats with 50 fleas
per week, resulted in a mean of 332 fleas per cat. Because 68% were female, male
survival times on the host were shorter than those of females. Female fleas produced
1 egg per h, and combined sexes averaged 0.77 mg of feces per day. Average blood
ingestion for defecation was 6.97 j1 of blood. In vitro rearing resulted in lower egg
production (12%), feces production (50%), and ingestion of blood for defecation.
Two types of flea feces were found-spherules and coils. Within 24 h of first feeding,
almost all feces were spherules <0.07 mm in diameter. After 10 days of feeding, 60-70%
of the feces were coils. These adult feces are the natural larval diet of cat flea larvae.


Se investigo la hematofagia de la pulga gatuna, Ctenocephalides felis felis (Bouche).
Cuando los estados adults de ambos sexos se alimentaron de sangre, su peso se duplico.
Sin embargo, 12 horas despues, perdieron este peso. La cantidad de protein se triplico
despues de la alimentacion, pero la inanicion causo una reduction en la protein, y luego
el porcentaje de esta se mantuvo constant (5%).
Tanto la cria en vivo como en vitro de las pulgas de gatos, fue exitosa y permitio la
sobrevivemcia, alimentacion, production de material fecal y reproduccion. La cria en
vivo en gatos infestados con 50 pulgas por semana, result en un promedio de 332 pulgas
por gato. Cuando el 68% de la poblacion era femenino, la sobrevivencia de los machos
fue menor que la de las hembras.
Las pulgas hembras produjeron un huevo por h, y ambos sexos produjeron 0.77 mg
de feces por dia. El promedio de ingestion de sangre para defecacion fue de 6.97 ul de
sangre. La cria en vitro result en una production baja de huevos (12%), production de
feces (50%), e ingestion de sangre para defecacion. Se encontraron 2 tipos de feces de
pulga, esferoides y espiroides. Venticuatro h despues de la primera alimentacion, todas
las feces fueron esferoides < 0.07 mm de diametro. Diez dias despues, 60-70% de las
feces fueron espiroides. Estas feces de adults sirven de dieta natural para las larvas
de las pulgas.

Florida Entomologist 74(3)

September, 1991

Hematophagous strategies of fleas are of particular interest because during blood
feeding they can transmit plague (Yersinia pestis; Pollitzer 1960) and murine typhus
(Rickettsia mooseri; Farhang-Azad & Traub 1985). In addition, flea bites can cause
severe pruritus and dermatitis in domestic animals (Kissilett 1938). While scratching
and grooming irritated skin, the host can ingest fleas, resulting in transmission of the
dog tapeworm, Dipylidium caninum (Hamrick et al. 1983).
Cat fleas, Ctenocephalides felis felis (Bouche), have been collected from a wide
spectrum of hosts and have been found to successfully reproduce on a variety of blood
types, although many fleas have very narrow host ranges. Sustained infestations have
been reported on cats, dogs, humans, poultry, rodents, artiodactyl livestock, and various
wild mammals (Yeruham et al. 1982). Cat fleas feed to repletion in only ten minutes;
however rat fleas often feed for up to two hours if left undisturbed (Iqbal & Humphries
1982). Success in obtaining a blood meal is related to the host's grooming efficiency
(Waage & Nondo 1982).
Cat fleas and other hematophagous arthropods have refined abilities to locate verteb-
rate hosts (Rothschild & Ford 1973), behaviors to allow interaction with their hosts
(Marshall 1987), and modifications of mouthparts for blood sucking (Hocking 1971).
Fleas, like other hematophagous arthropods, locate their hosts by kairomones emitted
by the host, including carbon dioxide and other volatiles (Smith et al. 1970, Vale 1984).
There are many scents emitted by warm living bodies which allow biting insects to cue
in on them (Omer & Gillies 1971), and insects have modifications of their anatomy and
physiology which permit them to exploit these signals (Davis & Sokolove 1976). Numer-
ous short-distance chemical cues include heat and moisture (Burgess 1959), lactic acid
(Davis & Sokolove 1976), carbon dioxide (Gillies 1980), and other components (Khan &
Maibach 1966).
Fleas have a laterally flattened body shape with extensive spination that makes it
difficult to groom fleas from the pelage of the host (Karandikar & Munshi 1950). Species
with extensive spination, such as cat fleas, are more successful in remaining on the host
than those lacking genal combs (e.g. Orchopeas howardii; Amin & Sewell 1977).
Fleas have suctorial mouthparts with strong buccal and pharyngeal muscles, well
adapted for piercing the skin and sucking blood (Quick 1972, Sutcliffe & McIver 1984).
Daniel and Kingsolver (1983) reviewed the requirements for insect mouthparts capable
of piercing the vertebrate skin and sucking blood, with detailed analyses of concentration,
viscosity and the mechanical constraints of blood-feeding.
The proventricular armature of fleas (Coluzzi et al. 1982) is used to physically disrupt
the blood cells and release intracellular components for digestion by the midgut. Fleas
have digestive enzymes specifically adapted to handling blood (Prasad 1979). Levels of
the catheptic enzymes, lysosomal carboxy-peptidase, and aminopeptidase have been
determined for various times following the blood meal (Houseman & Downe 1983).
Reinhardt (1976) has prepared an extensive description of the physiological and mor-
phological changes in Xenopsylla cheopis, Echidnophaga gallinacea, and Tunga pene-
trans midguts precipitated by blood-feeding. The three flea species had two different
feeding strategies-intermittent feeding (temporary parasitic) and continuous feeding
(stationary parasitic). Not surprisingly, he found that midgut changes occurring in the
intermittent feeders are cyclic whereas the midgut morphology and physiology of continu-
ous feeders after a blood meal is rather constant.
The dramatic alopecia and pruritus of cat flea allergy dermatitis (Kissilett 1938) is
considered evidence of the early stages of coevolution in which the host and flea have
not yet achieved an optimal adaptation to one another (Girardin & Brossard 1985).
Usually it is in the insect's advantage not to irritate the host by feeding, so that it can
successfully complete the blood meal and escape to reproduce (Langley 1967). However,
flea saliva is irritating to the host. Adult cat fleas readily resume feeding after disruption


Hinkle et al.: Cat Flea (Siphonaptera: Pulicidae) 379

and their feces is the main natural diet of cat flea larvae. Consequently it is essential
that feces be deposited in the same location as the flea eggs. The irritation of the flea
bite may be designed to provoke scratching by the host, thus insuring that eggs and
feces will be deposited simultaneously.
The objective of this study was to investigate the hematophagous strategies of the
cat flea both in vivo and in vitro. Three main areas were investigated-the effect of
blood-feeding and intervals of starvation on adult weights and protein, the phenomenon
of prediuresis and the associated fecal production, and egg production.


In vivo feeding. Two adult cats, neutered prior to sexual naturity, were used as hosts
for the cat flea. The cats were housed separately in stainless steel cages (45 by 60 by
30 cm high) with screen floors (12 mm mesh). Cats were infested by placing about 50
fleas per week on them.
Flea eggs and feces were collected by removing a tray (43.5 by 61 by 6 cm deep)
under the cage and brushing the contents into a Petri dish. Eggs and feces were separated
from large debris by sieving (No. 10 U.S.A. Standard, 2.0 mm opening). Numbers of
flea eggs were counted daily for two days.
Daily flea feces production was quantified by measuring total hemoglobin using a
total hemoglobin test kit (Sigma Chemical, St. Louis, Missouri). Feces and other debris
from the tray was dissolved in Drabkin's reagent (5 ml) and filtered. A split beam
spectrophotometer (Lambda 6; Perkin-Elmer, Norwalk, Conn.) measured absorbance
at 540 nm. Amount of flea feces was determined using a standard absorbance curve for
known quantities of flea feces in Drabkin's reagent.
Numbers of fleas on the cats were quantified by serial combing with a fine metal
comb (12 teeth per cm). Between combings, cats were returned to cages to collect flea
eggs. Combing and egg collections were done until no eggs were recovered, indicating
all fleas were removed.
In vitro feeding. Cat fleas were fed citrated bovine blood using a device similar to
the one described by Wade & Georgi (1988). The fleas were confined in cages made from
a plastic vial (5 by 4.5 cm diam.). The bottom was cut off and the open end was covered
with nylon screen (300 um mesh). A hole (3 cm. diam) was cut in the vial lid and covered
with screen (500 um mesh). The vial was inverted so that the fleas would insert their
mouthparts through the upper fine screen for a blood meal, and eggs and feces would
fall through the lower screen.
A thin layer of cat fur was provided as a substrate inside the cage. The fur was
suspended against the upper screen with a wire screen (4 cm. diam. and 0.75 cm mesh).
The fur provided a substrate for the fleas to cling to while feeding, and posed minimal
obstruction for the eggs and feces to fall through the lower coarse screen.
The tubes containing bovine blood were plastic vials (15 dram) with a 0.5 cm hole
drilled in the bottom. Plexiglass tubing (1 cm, 0.25 cm inner diam.) was cemented to
the hole. The open end of the vial was covered with Parafilm and about 10 ml of bovine
blood was added through the Plexiglass tubing.
The blood temperature was maintained at 41C with an electrical heat band cable
(TPI model TPT-6, W. W. Grainger, Chicago, Ill.) connected to a digital temperature
controller (Model CN5000, Omega Engineering, Stamford, Conn.). The heat was distrib-
uted to the blood using a copper collar.
The flea feeding device (91 by 13 by 12.5 cm) had locations for eight cages. The blood
vials were placed on top of the cages so that the fleas could feed by inserting their
mouthparts through the screen and the Parafilm to the blood. An aluminum weighing

Florida Entomologist 74(3)

September, 1991

pan (7 cm diam.) was placed under the cage to collect eggs and feces that fell through
the lower screen.
To determine the egg and feces production, 10 fleas were placed in each cage and
fed blood continuously for 7 days. Blood was changed every 24 h. Eggs and feces were
collected, weighed, and counted daily. The cage was weighed and dead fleas within the
cage were counted daily. Because the fleas only began egg production on the third day,
the egg production per cage and per female data was calculated only for days 4-7.
To determine the effect of feeding and starvation, 100 fleas per cage were placed in
the eight cages. Six to 12 fleas were removed from each cage before feeding, immediately
after feeding, and at 12 and 24 h of starvation after feeding. The fleas were weighed in
groups of 48-90, and the test was replicated 10-12 times. Types of feces produced following
defined periods of blood feeding were examined microscopically, categorized, and quan-
Protein determinations. Flea eggs, larvae, pupae, and adults were weighed on a
semianalytical balance ( 0.01 mg) and ground in a borosilicate tissue grinder with 3.0
ml (6.0 ml for the two fecal samples and the bovine blood) phosphate buffer (pH. 8.0).
The homogenate was filtered through Whatman #1 paper. Using Bradford's (1976)
method of protein analysis, 0.1 ml of each homogenate was placed in 3.0 ml of reagent,
and absorbance of the reaction mixture was read by a spectrophotometer.
Statistical analysis. Data on protein levels, egg production, feces production, and
flea longevity were analyzed by analysis of variance (GLM) and means were separated
by Duncan's multiple range test (P = 0.05; SAS Institute 1988).


Body weight of mixed-sex adult fleas increased significantly after blood feeding
(Table 1). Before feeding, flea weights averaged 0.192 mg, and weights nearly doubled
after access to blood for 24 hours. Starvation for 12 and 24 hours resulted in the loss of
all gained weight, with mean weights lower than before feeding, likely due to the
expenditure of digestive enzymes.
Protein levels of adult fleas also increased after feeding, but to a greater extent that
body weight. Soluble protein content immediately after feeding more than tripled from
0.005 to 0.017 mg per flea. Starvation for 12 and 24 hours resulted in a loss of gained
protein, but protein remained at about double the level of unfed fleas. Protein for unfed
fleas averaged about 2.6% of body weight, and immediately after feeding was 5.0% of
body weight. Even after starvation for 12 and 24 h, the percentage of protein was 5.0
and 5.7%, respectively. Although the amount of protein decreased with starvation, other
body constituents were selectively lost during starvation. There is little doubt from the
effects of starvation on body weight and protein that fleas must feed at least every 12
h in order to survive and reproduce.


Mean SE
Hours since
Feeding status last feeding Live weight (mg) Protein (mg)

Unfed 0.192b 0.012 0.005b 0.001
Fed 0 0.343a 0.010 0.017a 0.002
Fed 12 0.181b 0.001 0.009b 0.001
Fed 24 0.175b 0.012 0.010b 0.001

Means within a column followed by the same letter are not significantly different.


Hinkle et al.: Cat Flea (Siphonaptera: Pulicidae) 381

Infesting cats with 50 fleas per week resulted in an infestation averaging more than
300 fleas per cat. About 68% of the fleas on the animals were female, indicating longer
longevity or survival of female fleas on the host. Osbrink & Rust (1984) found that the
lifespan of male fleas (7.2 days) placed in microcells on cats was shorter than that of
females (11.2 days). Average flea egg production per cat was 5,577 eggs per cat or 24
eggs per female. Evidently, a healthy female can produce about one egg per hour.
Osbrink & Rust (1984) reported that female fleas have six ovarioles in each of two
ovaries, and that fleas from cats had an average of six mature eggs in the abdomen.
They also found that egg production throughout the female's lifetime averaged 158.4 eggs.
In vitro production was based on about 10 fleas per cage. Egg production began on
the third day of the experiment and was markedly less than from fleas released on cats
with a daily production that averaged 14 eggs per cage. Assuming that 50% of the fleas
in the cages were female, daily egg production was almost 3 eggs per female. Therefore,
we concluded that in vivo production of fleas is about 8 times more efficient than in
vitro production.
Feces produced from fleas on cats averaged 254 mg per day or 0.77 mg per flea, in
contrast to significantly lower production in vitro (Table 2). Fleas on the artificial feeding
device excreted about 50% of the amount of feces excreted by fleas on cats. The lower
feces production in vitro indicates that the system is not yet perfected and flea feeding
is lower than that achieved by fleas on cats. The dry weight and specific gravity of blood
was used to estimate the total volume of blood consumed to produce feces. Fleas consume
about 6.97 ul of blood on a cat to produce feces compared to 2.28 ul in vitro.
Two types of adult flea feces are produce, spherules and coils. Spherules can be
graded by size as small (<0.07 mm) or large (0.10-0.25 mm). The small spherules are
usually cohesive and are stuck to each other like a beaded necklace. Although the cat
flea is not considered to have one, the configuration of small spherules is an argument
for the presence of a peritrophic membrane. The large spherules are shiny, with little
surface decoration, and discrete. The coils average 0.84 mm in length with a diameter
somewhat less than the big spherules. The furrows visible on the coils are likely due to
the configuration of the rectum, so as moisture is removed from the forming fecal bolus,
strata develop.
Almost all the spherules produced by fleas during the first 24 hours of feeding on a
host are small (Table 3). However, by the tenth day of feeding, most flea fecal material
is excreted as coils. The remaining fecal material is equally distributed as small and large


Mean + SE

in vivo (per cat) in vitro (per cage)

Numberoffleas 332.75 78.62 9.64 0.15
Number of females 226.75 58.87
Daily egg production 5,577.00 1,635.45 14.44 2.74
Eggs per female 23.96 0.83 2.92 0.56a
Daily feces production (mg) 253.63 + 59.61 2.60 0.87
Feces per flea (mg) 0.77 0.03 0.38 0.08
Dry weight of blood (g/liter) 109 120
Daily blood consumptionb (ul) 2,320 22
Blood consumption per fleab (ul) 6.97 2.28
"Assuming 50% of fleas were females.
bAmount of blood consumed to produce feces. Estimated based on the quantity of flea feces produced, and the
dry weight per liter and specific gravity of host's blood.

Florida Entomologist 74(3)


Types of feces produced
Days of
blood-feeding Spherules *0.07mm Spherules >0.10 mm Coils

1 89-90% <15% <5%
10 15-20% 15-20% 60-70%

spherules. The two types of fecal conformations produced are distinctive, and the timing
of their production is probably of significance, although the actual importance can only
be conjectured.
The spherules produced initially have a protein content of 7.4%. Subsequent feces
(mostly coils) has a higher protein content of 11.0%. Both protein levels are higher than
the 5% protein level of bovine blood that was fed to the fleas in this experiment.


Blood consumption and weight gain is quite easy to quantify in ticks and other larger
arthropods that take a distinct blood meal (Koch & Sauer 1984). For cat fleas and other
continuous feeders, like some fleas and mosquitoes, which exhibit prediuresis, it is
inherently difficult to quantify blood consumption and weight gain. Nevertheless, our
initial weights (0.19 mg) for cat fleas are similar to Joseph's (1976) weight of female
adult C.felis orientis (0.18 mg) prior to feeding. He documented that her weight increased
to 0.26 mg over 47 min. of feeding on a human, while excreting 0.88 mg rectal fluid.
Joseph (1976) also found distinct sexual dimorphism in the weights of these fleas, with
females weighing 2.5 times more than males.
Like the Anopheles mosquito (Briegel & Rezzonico 1985), cat fleas have prediuretic
excretion, i.e., excretion of serum or serum and erythrocytes while blood is ingested,
leading to protein concentration in the midgut. Fleas demonstrate an elegant adaptation
to this system in that the larval fleas are provided with both the spherules of essentially
unhydrolized blood and then later with the coils of partially digested blood. The signifi-
cance of the two types in the diet of larval fleas has yet to be determined. It may be
that the smaller spherules provide a manageable meal for the newly-hatched flea larvae
while the larger coils provide sufficient nutrition and are of an appropriate size for
consumption by later instars.
Adult flea feces is essentially dried blood; therefore, because flea larvae feed on this
dried blood, they, as well as the adults, could be considered hematophagous parasites
that depend on the host and conspecifics for their nutrition. Considering the intimate
relationship of the larval flea and the host animal, it would not be too surprising to find
endoparasitism in some fleas. One unique species of flea is known to exhibit this lifestyle
where all stages of development actually occur on or in the host (Williams 1986a).
In species where the larva is free-living, it would be important for the adult to
provide adequate nutrition for the development of the larva. By passing through its
digestive tract more blood that is less digested, the adult augments the nutrient content
of the larval diet. Unlike many other holometabolous, hematophagous arthropods, the
male flea is exclusively a blood-feeder. This might be a form of parental care whereby
the male contributes to the feeding of the young. Both sexes produce larval food to
insure that feces are available for the developing larvae.
The cat flea was unable to successfully develop as a larva without feeding on adult
flea feces or dried blood (Strenger 1973, Moser 1989). Other organic materials (Bruce

September, 1991

Hinkle et al.: Cat Flea (Siphonaptera: Pulicidae) 383

1948) are suitable for larval nutrition, but it is almost impossible to collect eggs from
their contaminated environment without some fecal material adhering to the shell (Hud-
son & Prince 1958).
The discrepancy between volumes of blood consumed by males or females is of
particular interest in the consideration of paternal input in larval nutrition. Iqbal &
Humphries (1982) found that the male flea imbibes less than 15% of the blood volume
that the female does (0.8 X 10" compared with 6.21 X 104 mm3) in a single blood meal.
It would be revealing to compare the nutritional components of feces from the two sexes
to determine if the female actually produces most larval food. Also of interest are the
effects of temperature of the blood on consumption and the fecundity of blood-fed females
(Davis et al. 1983).
The integrity and appearance of flea feces may be indicative of a peritrophic mem-
brane. While the typical view is that a peritrophic membrane is produced in insects
which consume rough materials, to protect the delicate intima of the midgut, many
blood-sucking insects possess peritrophic membranes, as well. It has been demonstrated
in mosquitoes that the space between the midgut epithelium and the peritrophic mem-
brane is a compartment for optimal digestion, with a concentration of digestive enzymes
(Graf & Briegel 1982). A peritrophic membrane has also been found in Rhodnius (Bil-
lingsley & Downe 1983).
The blood of most vertebrates is quite similar in composition (Dittmer 1961). This
explains why some fleas have such wide host ranges; they are able to adequately exploit
the resources of a variety of blood types. For instance, Bacot (1914) found that rat fleas
could survive for extended periods fed solely on human blood, but that they were unable
to reproduce. Cat fleas can reproduce on calves, but reproductive failure based on
ovariole regression is 25% compared to <10% on cats (Williams 1986b). In general, it
has been found that hematophagous arthropods have higher fecundity when fed on their
normal hosts (Mather & DeFoliart 1983).
In conclusion, it appears that while fleas may have initially been scavengers in animal
nests and burrows, they have subsequently evolved nearly complete hematophagy
coupled with a range of host association from the perfunctory up to and including en-
doparasitism (Prasad 1987). The cat flea on the surface appears to be poorly adapted to
its hosts because of irritation, pruritis, and allergic responses. In reality, the irritation
is an elegant adaptation to insure larval hematophagy. The irritation causes host groom-
ing that dislodges both eggs and adult flea feces into the same environment. The reali-
zation that cat flea larvae are also hematophagous will certainly enhance our understand-
ing of flea ecology and behavior.


This is Florida Agricultural Experiment Station Journal Series number R-01821.


AMIN, O. M., AND R. G. SEWELL. 1977. Comb variations in the squirrel and chipmunk
fleas, Orchopeas h. howardii (Baker) and Megabothris acerbus (Jordan)
(Siphonaptera), with notes on the significance of pronotal comb patterns. Amer.
Midl. Nat. 98(1): 207-212.
BACOT, A. 1914. A study of the bionomics of the common rat fleas and other species
associated with human habitation, with special reference to the influence of tem-
perature and humidity at various periods in the life history of the insects. J. Hyg.
13 (Plague Suppl. 3): 447-654.
BILLINGSLEY, P. F. AND A.E.R. DOWNE. 1983. Ultrastructural changes in posterior
midgut cells associated with blood feeding in adult female Rhodnius prolixus Stal
(Hemiptera: Reduviidae). Canadian J. Zool. 61(11): 2574-2586.

384 Florida Entomologist 74(3) September, 1991

BRADFORD, M. M. 1976. Rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Analytical
Biochemistry 72: 248-254.
BRIEGEL, H., AND L. REZZONICO. 1985. Concentration of host blood protein during
feeding by anopheline mosquitoes (Diptera: Culicidae). J. Med. Entomol. 22(6):
BRUCE, W. N. 1948. Studies on the biological requirements of the cat flea. Ann. Ent.
Soc. Amer. 41: 345-352.
BURGESS, L. 1959. Probing behaviour of Aedes aegypti (L.) in response to heat and
moisture. Nature, Lond. 184: 1968-1969.
COLUZZI, M., A. CONCETTI, AND F. ASCOLI. 1982. Effect of cibarial armature of
mosquitoes (Diptera: Culicidae) on blood-meal haemolysis. J. Insect Physiol.
28(10): 885-888.
DANIEL, T. L., AND J. G. KINGSOLVER. 1983. Feeding strategy and the mechanics
of blood sucking in insects. J. Theor. Biol. 105(4): 661-672.
1983. Laboratory blood feeding of Culicoides mississippiensis (Diptera:
Ceratopogonidae) through a reinforced silicone membrane. J. Med. Entomol.
20(2): 177-182.
DAVIS, E. E., AND P. G. SOKOLOVE. 1976. Lactic acid-sensitive receptors on the
antennae of the mosquito, Aedes aegypti. J. Comp. Physiol. 105: 43-54.
DITTMER, D. S. 1961. Blood and Other Body Fluids. Federation of American Societies
for Experimental Biology, Washington, D.C.
FARHANG-AZAD, A., AND R. TRAUB. 1985. Transmission of murine typhus rickettsiae
by Xenopsylla cheopis, with notes on experimental infection and effects of tem-
perature. Am. J. Trop.Med. Hyg. 34(3): 555-563.
GILLIES, M. T. 1980. The role of carbon dioxide in host-finding by mosquitoes (Diptera:
Culicidae): a review. Bull. Ent. Res. 70: 525-532.
GIRARDIN, P., AND M. BROSSARD. 1985. Development of delayed-type hypersensitiv-
ity by rabbits infested with Ixodes ricinus L. females. Ann. Parasitol. Hum.
Comp. 60(3): 299-309.
GRAF, R., AND H. BRIEGEL. 1982. Comparison between aminopeptidase and trypsin
activity in blood-fed females ofAedes aegypti. Rev. Suisse Zool. 89(4): 845-850.
1983. Two cases of dipylidiasis (dog tapeworm infection) in children: update on
an old problem. Pediatrics 72(1): 114-117.
HOCKING, B. 1971. Blood-sucking behavior of terrestrial arthropods. Ann. Rev. Ent.
16: 1-26.
HOUSEMAN, J. G., AND A.E.R. DOWNE. 1983. Activity cycles and the control of four
digestive proteinases in the posterior midgut of Rhodnius prolixus Stal (Hemipt-
era: Reduviidae). J. Insect Physiol. 29(2): 141-148.
HUDSON, B. W., AND F. M. PRINCE. 1958. A method for large-scale rearing of the
cat flea, Ctenocephalides felis felis (Bouche). Bull. World Health Organization
19(6): 1126-1129.
IQBAL, Q. J., AND D. A. HUMPHRIES. 1982. Feeding behaviour of the rat flea Nosop-
syllusfasciatus (Bosc.) Pakistan J. Zool. 14(1): 71-74.
JOSEPH, S. A. 1976. Observations on the feeding habits of Ctenocephalidesfelis orientis
Jordan, 1925 on human host. Cheiron 5(2): 73-77.
KARANDIKAR, K. R., AND D. M. MUNSHI. 1950. Life-history and bionomics of the
cat flea, Ctenocephalides felis Bouche. J. Bombay Nat. Hist. Soc. 49: 169-177.
KHAN, A. A., AND H. I. MAIBACH. 1966. Quantitation of effect of several stimuli on
landing and probing by Aedes aegypti. J. Econ. Ent. 59: 902-905.
KISSILETT, A. 1938. The dog flea as a causative agent in summer eczema in dogs. J.
Am. Vet. Med. Assoc. 93: 21.

Hinkle et al.: Cat Flea (Siphonaptera: Pulicidae) 385

KOCH, H. G., AND J. R. SAUER. 1984. Quantity of blood ingested by four species of
hard ticks (Acari: Ixodidae) fed on domestic dogs. Ann. Entomol. Soc. Am. 77(2):
LANGLEY, P. A. 1967. The control of digestion in the tsetse fly, Glossina morsitans:
a comparison between field flies and flies reared in captivity. J. Insect. Physiol.
13: 477-486.
MARSHALL, A. G. 1987. Nutritional ecology of ectoparasitic insects, pp. 721-739 in
Slansky, F., Jr. & J. G. Rodriguez [eds.], Nutritional ecology of insects, mites,
spiders and related invertebrates: an overview. Wiley, New York.
MATHER, T. N., AND G. R. DEFOLIART. 1983. Effect of host blood source on the
gonotrophic cycle of Aedes triseriatus. Am. J. Trop. Med. Hyg. 32(1): 189-193.
MOSER, B. A. 1989. The larval cat flea, Ctenocephalidesfelisfelis (Bouche): nutrition,
instar determination, and response to methoprene and diflubenzuron, two insect
growth regulators. M. S. Thesis. Univ. of Florida. 112 pp.
OMER, S. M., AND M. T. GILLIES. 1971. Loss of response to carbon dioxide in palpec-
tomized female mosquitoes. Entomologia Exp. Appl. 14: 251-252.
OSBRINK, W.L.A., AND M. K. RUST. 1984. Fecundity and longevity of the adult cat
flea, Ctenocephalides felis (Siphonaptera: Pulicidae). J. Med. Entomol. 21(6):
POLLITZER, R. 1960. A review of recent literature on plague. Bull. World Health
Organization 23: 313-400.
PRASAD, R. S. 1979. A comparative study of the digestive enzymes of two vectors of
plague. J. Com. Dis. 11(2): 94-96.
PRASAD, R. S. 1987. Host dependency among haematophagous insects: a case study
on flea-host association. Proc. India Acad. Sci. 96(4): 349-360.
QUICK, C. B. 1972. Allergic flea dermatitis. Southwest. Vet. Winter, 1972, pp. 119-122.
REINHARDT, C. A. 1976. Ultrastructural comparison of the midgut epithelia of fleas
with different behavior patterns (Xenopsylla cheopis, Echidnophaga gallinacea,
Tunga penetrans, Siphonaptera: Pulicidae). Acta Trop. 33(2): 105-132.
ROTHSCHILD, M., AND B. FORD. 1973. Factors influencing the breeding of the rabbit
flea (Spilopsyllus cuniculi): a spring-time accelerator and a kairomone in nestling
rabbit urine (with notes on Cediopsylla simplex, another "hormone bound"
species). J. Zool. (Lond.) 170: 87-137.
SAS Institute. 1988. SAS/STAT guide for the personal computer. SAS Institute, Cary,
MAYER, B. J. SMITTLE, AND A. HOFBAUER. 1970. L-lactic acid as a factor in
the attraction of Aedes aegypti (Diptera: Culicidae) to human hosts. Ann. Ent.
Soc. Amer. 63: 760-770.
STRENGER, A. 1973. Zur Ernaehrungsbiologie der Larve von Ctenocephalides felis
felis B. Zool. Jb. Syst. 100: 64-80.
SUTCLIFFE, J. F., AND S. B. MCIVER. 1984. Mechanics of blood-feeding in black flies
(Diptera: Simuliidae). J. Morphol. 180(2): 125-144.
VALE, G. A. 1984. The responses of Glossina (Glossinidae) and other Diptera to odour
plumes in the field. Bull. Entomol. Res. 74(1): 143-152.
WAAGE, J. K., AND J. NONDO. 1982. Host behaviour and mosquito feeding success:
an experimental study. Trans. R. Soc. Trop. Med. Hyg. 76(1): 119-122.
WADE, S. E, AND J. R. GEORGI. 1988. Survival and reproduction of artificially fed
cat fleas, Ctenocephalides felis Bouche (Siphonaptera: Pulicidae). J. Med. En-
tomol. 25: 186-190.
WILLIAMS, B. 1986a. Mandibular glands in the endoparasitic larva of Uropsylla tas-
manica Rothschild (Siphonaptera: Pygiopsyllidae). Int. J. Insect Morphol. Em-
bryol. 15(4): 263-268.
WILLIAMS, B. 1986b. One jump ahead of the flea. New Scientist 31: July.
YERUHAM, I., A. HADANI, A. SKLAR, AND K. RAUCHBACH. 1982. Infestation of
calves with the cat flea Ctenocephalides felis felis (Bouche, 1835). Refuah Vet.
39(1-2): 8-10.

386 Florida Entomologist 74(3) September, 1991


Department of Zoology, University of Canterbury
Christchurch, New Zeland


More than 180 species and subspecies of larvae from all except 4 small families of
fleas, and including at least 30% of known genera/subgenera, were examined with phase-
contrast and/or scanning electron microscopy. A much greater variety of detailed form
was revealed than appears in the literature. Especially significant are: numbers and
patterns of setae and their topographical relations to sensilla on the thorax and abdomen;
form of the setae, of which at least 9 types can be distinguished on the head, thorax
and abdomen; form, arrangement and surface texture of sensory papillae on the antennal
mound. Some of these features appear to be characteristic of, and restricted to, genera,
subfamilies or higher taxonomic categories. Others, such as the shape of the mandible
and their numbers of teeth, show considerable variation which does not clearly follow
recognized classifications based on adult fleas; these may be more related to factors in
the biology of specific larvae.


Con excepcion de 4 families, se examinaron mas de 180 species y subespecies en
estado larvario de pulgas. Se incluyo el 30% de los generos/subgeneros conocidos y se
examinaron con el microscopic de escanografia electronic con fase de contrast. Se
encontro una gran variedad de formas en comparacion con lo que aparece en la literature.
Se encontraron diferencias significativas en: numerous y patrons de setas y sus relaciones
topograficas con la sensilla del torax y abdomen; forma de las setas, de las cuales, al
menos 9 tipos pueden ser distinguidos en la cabeza, torax y abdomen; la forma y distribu-
cion, y la textura de la papilla sensorial en la antena. Algunos de estos rasgos parecen
ser caracteristicos o restringidos a un genero, subfamilia o categories taxonomicas altas.
Otras, tales como la forma de la mandibula y los numerous de molares, mostraron una
variacion considerable, la cual no sigue exactamente las clasificaciones reconocidas, las
cuales se basan en descripciones de las pulgas adults; estas pueden estar mas re-
lacionadas con factors de la biologia de las larvas.

The desirability of being able to identify flea larvae in the absence of the corresponding
adults) is not one of mere academic interest. It is of intense practical value to be able
to categorize a host's nest as to whether or not it contains certain flea species which
are potential vectors of bubonic plague or other diseases. This will be particularly useful
in the case of abandoned nests, or of nests examined in the temporary absence of the
hosts. It also makes allowance for seasonality factors in surveying the fauna of a nest,
i.e. in respect of "summer" and "winter" fleas, whereby the larvae may be present in
the near or complete absence of the adult flea(s). As well, a taxonomy of their larvae
can provide a base for extending studies on the biology of fleas; it could also add data
for consideration along with adult morphology in elucidating the taxonomy and phylogeny
of the order.
The life-cycle of a flea comprises: egg larva (with usually 3 instars) pupa adult.
About 5% of adult fleas are parasitic on birds, the remainder on mammals; they are

Pilgrim: External Morphology of Flea Larvae

found closely associated with the hosts or in their nests. Over 2200 species and subspecies
are known; they are classified on the basis of adult morphology in 15 families, grouped
into 5 superfamilies (Smit 1982). The families are of very disparate size, ranging from
Malacopsyllidae with 2 monotypic genera to Ctenophthalmidae with more than 630
species/subspecies and Ceratophyllidae with over 750.
The larvae are, with very rare exceptions, active scavengers living non-parasitically
in nest debris of the hosts. They are 1.5-10 mm long, legless, eyeless, maggot-like with
a firmly sclerotised head capsule. The head bears prominent antennae, well sclerotised
labrum, mandibles and maxillae, and a weakly developed labium. The body is usually
considered to comprise 3 thoracic and 10 abdominal segments, the last of which bears
a pair of anal struts projecting backwards and downwards from the swollen anal mounds.
Numerous long and short setae are present on the head capsule and on the body segments;
many setae on the body arise from alveoli in well-defined plate-like sclerotisations (Fig. 1).
This study was initiated to seek a key to the taxa of flea larvae over as wide a range
as possible, and to provide a basis for suggesting phylogenetic relationships within the
order as exhibited by the larvae. Before these aims could be realized, it was found
necessary to provide for a more consistent terminology for the larval morphology, since
there is much confusion between published accounts. Some features in the morphology
have been described in considerable detail in the literature for a number of species,
from many genera. These published accounts were based on light microscope examination
of slide-mounted specimens, and some patterns of taxonomic significance have emerged.
Re-examination of such material in the present study, together with supporting evidence
from scanning electron microscopy, now shows that some of these features have been
incorrectly interpreted. As well, attention is here drawn to a number of structures
which had been inadequately studied, dismissed as not significant, or simply overlooked;
they can now be shown to have considerable taxonomic value.
This account is based on examination of more than 3000 specimens from over 180
spp./sspp. of larvae in 84 genera/subgenera, taken from all superfamilies (sensu Smit
1982). The larvae of several small families remain unknown (Stephanocircidae, Ancitrop-
syllidae, Xiphiopsyllidae, Malacopsyllidae) as do the larvae of a few subfamilies/tribes
among the larger families. Generalizations made now must be subject to modification
as more widely representative collections are examined; almost every feature so far
described has been found to have some exception or enhancement when a fresh taxon
is studied. Uropsylla tasmanica Rothschild, 1905 (Pygiopsyllidae) is specifically excluded
from the generalizations which follow: its larva is so grossly modified in association with
its parasitic mode of life that almost every feature in its morphology would require a
separate statement.


1. Head: most published accounts describe the majority of setae behind the antennae
as distributed in 2 rows, anterior and posterior. These 'rows' are not very impressive,
several setae being variously displaced from anything resembling a linear arrangement;
and schools of writers are not in complete agreement as to the designation of the
lowermost setae. I propose to abandon the 'row' concept and to redefine the setal
nomenclature in terms of regions of the head capsule on which the setae lie; see Fig. 1.
Major distinctions are immediately evident between Pulicoidea and all remaining super-
families: the non-Pulicoidea possess, on each side of the head, 2 additional parietal setae,
a post-clypeal seta and an anterior genal seta beyond those found in Pulicoidea. (These
are basic patterns within which some variations occur, such as an extra pair of para-gular
setae in Stenoponia, and reduced station in Tunga.)
Many head capsule setae show important differences in length between different
groups, but the parietal setae are especially useful taxonomically in the non-Pulicoidea:

Florida Entomologist 74(3)

Th I

September, 1991

Ab 6


D plate

pa- 7.


V plate




Pilgrim: External Morphology of Flea Larvae


thus, at least 1, often 2 (but very rarely all 3) are long, the others short or minute (a
few pm; hence frequently overlooked). A great variation also occurs with respect both
to the topographic arrangement of these 3 setae within the parietal region, and to their
respective length, i.e. as to precisely which one(s) of the 3 may be long.
2. Body: thoracic segments I-III and abdominal segments 1-10 all bear very minute
microsetae towards their anterior limits; these have not been found to have any taxonomic
significance. The 2 major rows of setae on segments Th I Ab 9 are of immense value
in taxonomy (Fig. 1): (a) a posterior row of usually long setae, arising from alveoli set
in the sclerites of the segments, i.e. on the single dorsal (D), and the paired dorso-lateral
(DL), ventro-lateral (VL), and ventral (V) plates; (b) an anterior row of much shorter
setae, 1, 2 or 3 pairs of which have their alveoli on the D plate, the others mostly on
the cuticle in front of or between the sclerotised plates.
It is important, in descriptions, to relate the posterior row setae to the plates on
which they stand instead of merely citing their total number: thus whereas DL and VL
usually bear only 1 seta each, D may have 1-4 pairs and each V may have 1-3 setae.
Two different larvae may thus have the same total number of setae in a posterior row
but their distribution between D and V may be markedly different, and this forms an
excellent taxonomic character.
Station of the 10th abdominal segment is strikingly different from that of the other
body segments; together with the form and station of the anal mounds it provides much
information for taxonomic differentiation. As such, it has been well and intensively
described by many writers, though the nomenclature of some features requires reconsid-
eration in terms of their homologies with more anterior segments.
Both on the head and on the body segments, patterns of station appear to be stable
within a given taxon, apart from 'normal biological variations,' and are highly diagnostic.

Fig. 1. Diagrammatic representation of head, first thoracic segment (Th I) and sixth
abdominal segment (Ab 6) of typical non-Pulicoid larva (A dorsal; B lateral), and head
of typical Pulicoid larva (C). Sensilla of head not shown. Sizes of setae vary greatly
within each group. Abbreviations of setae:
a.gen. ..... anterior genal (non-Pulicoidea)
a, ..... dorsal plate seta, anterior row
Da ..... dorsal plate setae, posterior row
DL ..... dorso-lateral plate seta, posterior row
fr. ..... frontal
gen. ..... genal (Pulicoidea)
i-ant. ..... inter-antennal
1.cl. ..... lateral clypeal
l.occ. ..... lateral occipital
m.cl. ..... medial clypeal
m.occ. ..... medial occipital
p-ant. ..... post-antennal
p-cl. ..... post-clypeal (non-Pulicoidea)
p.gen. ..... posterior genal (non-Pulicoidea)
p-man. ..... post-mandibular
p-max. ..... post-maxillary
pa-gul. ..... para-gular
par. ..... parietal (1 in Pulicoidea; 3 in non-Pulicoidea)
s-gen. ..... sub-genal
V, V1-2 ..... ventral plate setae, posterior row
VL ..... ventro-lateral plate seta, posterior row
p ..... microseta(e)

Florida Entomologist 74(3)

September, 1991

0 0


Xenopsyl inae
loss of


mos t
supe families

a2 lateral to










Fig. 2. Pattern of setae and sensilla on dorsal plate (D) of abdominal segments Ab
1-6 (left half only shown). Some suggested phylogenetic relations are indicated.

Notes: 1. Examples have been examined from all superfamilies; in a few Hystrichop-
syllidae and Ctenophthalmidae (both Hystrichopsylloidea), seta a, lies
behind the sensillum.
2. Within the Ceratophylloidea, seta a, is medial or posterior to the sensillum
in some Ischnopsyllidae.
3. Pulicidae examined:
Xenopsyllinae: Synosternus, Xenopsylla (note X. hirtipes Rothschild,
1913 has both setae D1 and D2).
Pulicinae: Echidnophaga, Pulex.



Pilgrim: External Morphology of Flea Larvae

Contrary to some published opinions, I find that the patterns retain their characteristics
throughout the larval instars, the main changes being a progressive relative increase
in size of the setae at each moult. Even the pattern of parietal and occipital setae does
not become disturbed in the presence of the egg tooth in the first instar.


Most accounts refer to differences in length of various setae, but there are also great
variations in their shape and structure. The majority of setae are acicular, tapering
regularly to a fine tip. In several families, e.g. Hystrichopsyllidae, Pygiopsyllidae, at
least some body setae are obtuse; they are much stouter and their tip is blunt; they
may be short (Hystrichopsylla s.1.) or very long (Notiopsylla) and they are present not
only on the large larvae of these genera but also on the quite small Pagipsylla. Whereas
most head capsule setae are acicular, in Chiastopsylla they are attenuate, narrowing
rather abruptly to a finer almost filamentous nature at about mid-length. Ischnopsyllidae
larvae so far examined (3 genera, 7 species) have all their long body setae capitate, or
swollen at the tip. Similar, but perhaps only analogous, spatulate setae occur in many
Pygiopsyllidae, the terminal swelling tending to be more gradually developed from the
shaft. In both these types, the bulbous swellings are thin-walled and may perhaps be
devoted to the exchange of water or solutes with the substrate, but their function has
not been investigated.
In the above types of setae, the shaft may be quite smooth even at high magnification
in the SEM; in the following types the shaft is highly structured, the modifications being
visible at low power in light microscopy. Branched setae, with a few spine-like processes
irregularly arranged occur in Pygiopsylla; spiculate setae, with a series of finer processes
along one side of the shaft, are widespread in Pygiopsyllidae and they occur in Chiastop-
sylla and in Ischnopsyllidae as well. In pilose setae the processes densely clothe the
shaft, the tip of which may be truncated; they are very common in Pygiopsyllidae.
Individual setae may show a combination of types, and larvae may bear more than
1 type of seta: in Ischnopsyllidae, capitate setae towards the posterior end of the abdomen
have spiculate shafts; in Pygiopsyllidae the head capsule setae are acicular, but the body
segments variously bear acicular, obtuse, spatulate, branched, spiculate, and pilose
In Rhopalopsyllus the body setae appear acicular in light microscopy but at high
magnifications (SEM, x 2000) a helically wound series of fine grooves is visible producing
a striated appearance. This has not been seen in comparable SEM examination of any
other acicular setae and may be an idiosyncratic feature of Rhopalopsyllus (but note
that no other member of the Rhopalopsyllinae has been studied).


Clearly-defined, translucent, circular areas (ca. 4 Jm diameter), presumably sensory,
are visible on many parts of the body and the head, including the mouth appendages.
On the head capsule their patterns have not yet been completely elucidated though

Spilopsyllinae (birds): Actenopsylla, Ornithopsylla.
Archaeopsyllinae: Archaeopsylla, Ctenocephalides.
Spilopsyllinae (mammals): Cediopsylla, Euhoplopsyllus, Spilopsyllus.

Florida Entomologist 74(3)

September, 1991

there appear to be consistently more sensilla in the non-Pulicoid genera. On the body
segments they are confined to the dorsal plates (Fig. 1), apart from 1 on the ventral
face of each anal strut. In both the thoracic and the abdominal segments the sensilla
are closely associated, topographically, with setae of the anterior and posterior rows.
The combination of arrangements of these setae and the sensilla is taxonomically
useful and may also point to phylogenetic considerations. For example, in most super-
families the basic pattern over the first 6 abdominal segments is, on each half dorsal
plate (Fig. 2): a sensillum lying laterally of the single anterior seta (a1) and a sensillum
lying antero-medially of each of the 2 posterior setae (D1 and D2). It is characteristic of
all Ceratophylloidea so far examined (except some Ischopsyllidae) that the anterior seta
lies laterad of its associated sensillum. Within the Pulicoidea, the basic pattern is present
in Pulicinae and in those Spilopsyllinae parasitizing birds. In Xenopsyllinae it appears
that seta Di is absent from the posterior row (except in X. hirtipes Rothschild, 1913,
which may be primitive in at least this respect); conversely, it is seta D2 which is missing
in Archaeopsyllinae and the lagomorph-infesting genera of Spilopsyllinae.
This would suggest the Spilopsyllinae be separated into 2 groups, which neatly
parallel the host preferences: (1) those whose adults are parasites of birds (Actenopsylla,
Ornithopsylla) and (2) those whose adults are parasites oflagomorph mammals (Cediop-
sylla, Euhooplopsyllus, Spilopsyllua [Hoplopsyllus has not been examined]). This sep-
aration is supported by other differences: in group (1) the antennal mound papillae are
rounded (type A of Bacot & Ridewood 1914), and the integument bears small scales
with minute posteriorly directed spiny tips--this combination of characters also obtains
throughout the Xenopsyllinae, Pulicinae, and Archaeopsyllinae. In Spilopsyllinae group
(2), however, the antennal mound papillae are pointed (type B) and the integumentary
scales have long spines, as great as the length of the scale proper in Spilopsyllus.


One of the features which has commonly been incorrectly described is the series of
papillae on the mound at the base of the antennal shaft (Fig. 1). These were first referred
to in detail by Bacot & Ridewood (1914) who stated that there were 3 large papillae,
alternating with "3 (rarely 2") smaller ones; with few exceptions, subsequent authors
have followed this account, describing and even illustrating 6 papillae. Examination with
SEM shows that there are only 5 papillae on each side of the head and that variations
from this number are extremely rare abnormalities. The papillae lie postero-laterally
on the mound in an arc of about 1/3-1/2 the circumference. The 2 sizes of papillae are
of roughly the same shape, the larger (a-papillae) being a few to many times the bulk
of the smaller (p-papillae). On each side of the head they are here numbered, commencing
from the most anterior:
t1-P1--a-P2--3 (Fig. 3).

In the great majority of species, the 5 papillae are regularly spaced in their row,
but certain variations from this appear to have taxonomic significance (Fig. 3). In Ischnop-
syllidae (3 genera examined) a wide gap occurs between al and Pl; a similar gap occurs
between pi and 2 in Ctenophthalmus (11 spp./sspp., 3 subgenera examined) but has not
otherwise been seen in the very large family Ctenophthalmidae or elsewhere. Among
the Pygiopsyllidae, species of Notiopsylla have very large papillae and in all but one
(N. corynetes Smit, 1979) the row becomes bent away from the usual arc, with P2 and
to lie much closer to the antennal shaft.
Differentiation of the shape of the papillae was made by Bacot & Ridewood who
distinguished A-type (rounded tip) and B-type (pointed tip); Karadina (1964) further
suggested a sub-division of B-type, separating straight-sided from convex-sided cones.

Pilgrim: External Morphology of Flea Larvae




\ o

B\A A0^

1- oy

^ ^



/ \
1 0o
/ o\""

Fig. 3. Diagrammatic representation of structures on (left) antennal mound, show-
ing some variations in patterns. The emphasis is on relative positions; no attempt is
made to depict the various shapes of the papillae.
A. most common arrangement in all superfamilies.
B. Notiopsylla (Pygiopsyllidae); 5 spp./sspp., but not N. corynetes.
C. Ctenophthalmus (Ctenophthalmidae); 11 spp./sspp. examined in 3 subgenera.
D. Myodopsylla, and other Ischnopsyllidae.

Examination of a now much wider range of material shows that these categories are
too restrictive and fail to include all shapes adequately.
Immediately following o% is a structure, here termed y, which is approximately the
same diameter as the base of an a-papilla (Fig. 3). It is probably what has been mis-
interpreted by many authors, including Bacot & Ridewood, as a 6th papilla, though
no writer has commented that only a small papilla would be expected if the alternation
of sizes were preserved. SEM studies show y to be a slightly saucer-shaped depression in-
side which is a flat plate-like structure with a small pit at the center. The 6 structures, ai-y,
are presumably sensory (the papillae show minute pores at high magnification of the
SEM) and all of them show, by transparency in light microscopy, a widely open connection
through the cuticle into the head cavity.


Florida Entomologist 74(3)

Two further structures occur regularly on the antenna mound (Fig. 3): one, termed 8,
lies adjacent to y, thus directly posterior to the antennal shaft; and e, whose position
varies from dorsal (closest to the mid-line) to, less commonly, quite close to 8, e.g. in
some Ischnopsyllidae and Leptopsyllinae. These 2 structures appear to be identical.
They are best seen in the sclerotised ring of the antennal mound. In optical section they
appear as excavations of the cuticle, but from the fact that they do not usually show in
SEM, they clearly do not penetrate to the outside. Their inner openings can be shown,
in optical section, to bear a slight flange protruding into the cavity of the head. Their
occasional appearance in SEM as slight dimples is interpreted as due to inward collapsing
of the thin roof during preparation of the specimen. No sign of a penetrating pore has
been seen in either 8 or e at up to 5000x magnification, and it is suggested that these
structures are not sensory, but possibly are the sites of attachment of muscles operating
the antennae; they would therefore be apodemal insertions of such muscles.


A great variety of form is now found to occur with regard to almost every structure
in the external morphology of flea larvae. Bearing in mind that the larvae are, with
extremely rare exceptions, free-living, scavenging detritus-feeders, it is timely to begin
to examine the significance of some of these variations. Attention has been drawn,
above, to the range of types of setae; at least those with swollen tips must have a
functional significance.
But perhaps one of the most conspicuous variations lies in the shape of the mandibles.
Many forms have been described in the literature, ranging from a single tiny tooth at
the tip of the mandible (Hystrichopsylla talpae (Curtis, 1826) ), to a few coarse teeth
(many Pulicidae), several finer teeth (many Ceratophyllidae), a row of blunt cusps
(Rhadinopsylla), or a series of finger-like teeth arranged around the expanded end of
the mandible (Typhloceras). Most of these variations occur sporadically through the
order, without obvious significance at a high taxonomic level; they would seem to be
related, rather, to differences in diet or methods of obtaining/ingesting food, about which
very little is known. Yet several of these mandible types may occur among larvae living
in the same nest and presumably ingesting components of the same nest debris. It is
not established, however, that all larvae eat the same food items, and investigations
might well seek to distinguish whether differing portions of the debris are eaten by the
different larvae. For example, larvae of 4 species of fleas are found in the nest litter of
the 'Mountain Beaver' Aplodontia rufa (Rafinesque, 1817) in the northwestern Nearctic
and their mandibles show some remarkable differences (Pilgrim, unpublished): Hys-
trichopsylla schefferi Chapin, 1919 has 6-7 sharp teeth arranged around the end of the
mandible (as do other Nearctic species of this genus, in contrast to the European Palaearc-
tic species); Paratyphloceras oregonensis Ewing, 1940 has a broad mandible with a stout
blunt tip and 6-9 low cusps along the concave margin; the closely related Trichopsylloides
oregonensis Ewing, 1938 resembles the latter, but is smaller; Dolichopsyllus stylosus
(Baker, 1904) has a single stout sharp tooth set on a very inflated base whose 2 setae
are now enormously enlarged.
It is difficult to believe that these mandibles are all devoted to the ingestion of the
same component of the debris; an examination of gut contents, together with observations
and experiments on live larvae might well show food-ingesting discrimination on the
part of these species. Comparable situations can be found in many other instances. Some
mandibles appear, a priori, to be well suited for scraping at inert or immobile substrata,
others more fitted for impaling living prey; a thorough study of the interrelations of
nest fauna and flora could shed some light on these aspects. As well, examination of the
nest fauna as a whole should indicate predator-prey relations, which could be important

September, 1991

Pilgrim: External Morphology of Flea Larvae

in the natural control of flea populations via larval mortality. Little work has been done
on this aspect, though considerable effort has gone into examining physical factors
obtaining in larval environments. It is acknowledged, of course, that a number of species
of larvae have been successfully reared in laboratories on 'artificial' diets including added
yeasts, blood or blood derivatives, such diets or pabula being 'fauna-free'. In not all
attempts, however, has it been possible to rear all instars and the failure might be due
in some measure to the absence of a dietary requisite.
With regard to the debris factor, it might be interesting to compare conditions-phys-
ical and biotic-in nests of mammals with those of birds. Admittedly there are wide
differences between nest conditions within these 2 groups of hosts, but the question
arises as to whether the litter in a bird's nest might not show qualitative distinctions
from that of a mammal's, bearing in mind that a bird's nest is usually occupied by the
host for a short time (the breeding period only): the sudden accumulation of organic
detritus, including a high nitrogen content, must surely affect the environment in which
the larvae find themselves. In contrast, the mammal's nest is often one of long-term
occupancy with host-breeding superimposed on it at intervals; the changes) in physical
conditions may be much less marked at those times and the larvae subject to less drastic
changes. The physical conditions in the 2 cases are probably accompanied by differences
in the in-fauna/flora between bird and mammal nest.
In this context, it would be interesting to compare the detailed morphology of larvae
of closely related taxa, in which some are inhabitants of birds' nests whereas their close
relatives live in mammals' nests. For example, most species of Xenopsylla are mammal
parasites but X. gratiosaJ. & R., 192, X. moucheti Smit, 1958 and X. trispinis Waterston,
1911 are found on birds. Similarly, 2 genera of Ceratophyllidae include both mammal
and bird parasites (Smit 1983): within Ceratophyllus s.l., the subgenera Ceratophyllus
s.s., Celeophilus, and Emmareus are associated with birds, as is the subgenus Orneacus
of Callopsylla s.l. The 1 species of Echidnophaga (gallinacea (Westwood, 1875) ) living
commonly on birds also parasitises many mammals, but its larva might be compared
with those of other strictly mammal-infesting species of the genus. Many other examples
could be examined, mainly between genera, in families such as Rhopalopsyllidae.


This study was partly supported by grants from the New Zealand Lottery Board,
to whom I extend my grateful thanks.


BACOT, A. W., AND RIDEWOOD, W. G. 1914. Observations on the larvae of fleas.
Parasitology 7(2): 157-175.
KARANDINA, R. S. 1964. [The larvae of fleas of the red-tailed Libyan Jird and other
rodents]. Trudy Armianskaia protivochumnaia stantsiia 3: 473-504 (In Russian).
SMIT, F.G.A.M. 1982. Siphonaptera, in 'Synopsis and Classification of Living Or-
ganisms', S. P. Parker (ed.) 2: 557-563.
SMIT, F.G.A.M. 1983. in 'The Ceratophyllidae: Key to the genera and host relationships'
R. Traub, M. Rothschild and J. F. Haddow. London: Rothschild and Traub: 1-36.


396 Florida Entomologist 74(3) September, 1991


Department of Zoology, Southern Illinois University
Carbondale, IL 62901


Melanolestes picipes has been separated from M. abdominalis on the basis of color
(black in picipes, red wholly or in part in abdominalis), wing form (macropterous and
brachypterous in picipes, macropterous in abdominalis), and size of ocelli (smaller in
picipes). Evaluation of these characters showed they were not diagnostic. We also
examined the male and female external genitalia and found no consistent differences
within sex between the two "species." We therefore conclude that M. abdominalis is
not a valid species but is a junior synonym of M. picipes.


Tradicionalmente, se ha separado Melanolestes picipes de M. abdominalis en base
a el color (negro en picipes, completamente rojo, o en parte rojo en abdominalis), y el
tamafio del oceli (mas pequefio en picipes). La nueva evaluacion de estos caracteres
demostro que no son suficientes para establecer un diagnostic. Se examine tambien la
genitalia externa del macho y la hembra, y no se encontraron diferencias entire las dos
specieses. Concluimos que M. abdominalis no es una especie valida, sino un sinonimo
menor de M. picipes.

Herrich-Schaeffer (1846) described two species of reduviids, Pirates (sic) picipes (p.
62) and P. abdominalis (p. 63), now placed in the genus Melanolestes StAl. The descrip-
tions (here translated), each based on a single male specimen, are as follows: M. picipes-a
black Pirates, antennae and legs black; appears very similar to Reduvius personatus,
but has the generic characters of the genus Pirates, stouter form, shorter antennae,
thicker femora, etc.; a male from North America from Mr. Sturm. M. abdominalis-a
very black insect, abdomen scarlet, anal region black; the sole of the tibia bears golden-
yellow hairs; a male from Mr. Sturm from North America.
StAl (1872, p. 107) questioned the validity of M. abdominalis as a species for he
listed it as a variety of M. picipes. He also added to the description of picipes by noting
that the hemelytra of the variety are very abbreviated.
Uhler (1876, p. 330) treated M. picipes and abdominalis as separate species. He
stated, "The evidence at present in my possession does not warrant the uniting of these
two species. Both are quite common in Maryland, sometimes occurring under the same
stone; but while I have seen the sexes united, I have never seen a male of the one caress
or unite with a female of the other. The width and proportions of the head and pronotum
and abdomen vary considerably in the specimens of both of these species, so that, in
the absence of a long series of them, they might be made to constitute a number of
species." He followed this opinion in subsequent publications (1878, p. 424; 1884, p. 281;
1886, p. 25). In 1884, he further stated that M. picipes is black with piceous legs and
antennae, while abdominalis has the sides and sometimes the whole upper surface of
the abdomen red.

McPherson et al.: Melanolestes abdominalis

Parshley (1917) in his treatment of the New England fauna listed the two forms as
M. picipes and M. picipes abdominalis without explanation. However, in 1918, he stated,
"I have in my collection examples showing all gradations from those having only the
slightest tinge of red along the connexivum to those having the abdomen entirely red;
I have also a pair taken in copulation (Framingham, Mass., C. A. Frost) in which the
male is an entirely black long-winged picipes and the female a short-winged abdominalis,
with red connexivum. It would seem therefore that the abdominalis form should be
ranked as a mere color variety and not as a species distinct from picipes, as I have done
in my New England list." Therefore, he agreed with StAl (1872). Parshley used the term
"variety" to designate subdivisions of the species (not aberrations) which he thought
should be named without structural, geographic, or genetic connotations. Subsequently
(1922), he listed abdominalis in his South Dakota study as M. picipes var. abdominalis.
Several authors followed Uhler's opinion, many of whom published prior to Parshley's
1918 paper (e.g., Provancher 1887; Champion 1899; Smith 1909; Banks 1910; Fracker
1912; Van Duzee 1916, 1917; Torre-Bueno 1923; Blatchley 1926; Brimley 1938; Wygod-
zinsky 1949; Elkins 1951; Coscaron 1983; Froeschner 1988); only Maldonado Capriles
(1990), as far as we are aware, has followed Stal and Parshley. Blatchley (1926) discussed
the opinions of StAl, Parshley, and Uhler but decided to treat the two forms as separate
species because he had not collected "a specimen with intermediate hues." In addition
to color, he separated the "species" by wing form elytraa often abbreviated in pecipes,
entire in abdominalis) and by the size of ocelli (smaller in picipes). He noted that his
abdominalis specimens (6 males, 8 females), on which he based his key, were all fully
winged; he included in his key individuals that had the abdomen in part (connexivum?)
or entirely red but included only the latter in his species description.
Drew and Schaefer (1963), Froeschner (1944), and Slater and Baranowski (1978)
treated the two forms as distinct species but indicated some problem with the status of
Readio (1927) included M. picipes and abdominalis in his monographic study of the
Reduviidae of America north of Mexico. He mentioned he had in his collection five
different kinds of individuals based on sex, color and wing form: (1) male, long-winged,
entirely black; (2) male, long-winged, reddish abdomen; (3) female, long-winged, entirely
black; (4) female, short-winged, entirely black; and (5) female, short-winged, reddish
abdomen. He stated he had no short-winged males and doubted they existed. He felt
the short-winged and long-winged females might represent the female sex of two distinct
species but that the males would have to be separated by some other character. He also
felt that color was not a reliable diagnostic character. Both suppositions were based on
the following experimental results.
To determine if the red color of the abdomen was inherited, Readio planned to mate
black and red individuals in various combinations. He found, however, that in most
cases the females collected in the field to begin the experiment had previously mated
because they produced fertile eggs without being mated in the laboratory. Therefore,
he could only observe the offspring from these fertile females. From short-winged
females (remember, all males are macropterous) of both color forms, he obtained four
long-winged males and three short-winged females; all were black. From long-winged
females of the black variety (he collected no long-winged abdominalis females), he
obtained only black long-winged individuals, two males and one female. Therefore, he
felt color was not inherited but probably resulted from laboratory rearing conditions.
Also, because he obtained only ten adult offspring, he felt that his supposition that the
long- and short-winged females might represent two different species was weak.
This paper presents the results of our study of the taxonomic status of the two
"species" of Melanolestes. We examine the validity of size of ocelli (Blatchley 1926) and
color as diagnostic characters and the significance of wing length (macroptery and
brachyptery) in the two forms. Furthermore, we examine the male and female external
genitalia of both forms.

Florida Entomologist 74(3)


We examined specimens from several collections (see Acknowledgments). We re-
corded the number of individuals exhibiting each color form and, of those, the number
that were macropterous or brachypterous. Ocelli were measured as the transverse width
of an ocellus versus interocellar distance to compensate for individual size variation.
Both male and female external genitalia were examined. Dried male specimens were
softened in warm water to facilitate extraction of the pygophore (genital capsule).
Pygophores were then soaked for 24 h in room temperature KOH (10%). The phallus
was separated from the pygophore by severing the apodemes connecting the basal plates
(BP, Fig. 2) of the phallus to the internal walls of the pygophore. Dissected pygophores
and phalluses were cleared in glycerin or clove oil and drawings prepared using a camera
lucida. The terminology used in the figures of genitalia follows Davis (1966) for males
and Scudder (1959) for females.
Pairs of means were compared (P = 0.01) using Student's t test with unequal variances,
and multiple comparisons were made using one way ANOVA with means separated
using Duncan's multiple range test (SAS Institute 1988).


Examination of 442 specimens from several states (Fig. 1) and Baja California re-
vealed that the specimens could not be divided into two or three groups based on color
(i.e., abdomen black versus abdomen in part or completely red) but, in agreement with
Parshley (1918), were more variable in color (Table 1). Even when the 'black' category
was expanded to include dark reddish black, and the 'red' category to include reddish
orange and orange, several specimens still could not be placed in one category or the


Fig. 1. Geographic distribution of Melanolestes picipes as given by Froeschner (1988)
and of specimens examined in present study.

September, 1991


McPherson et al.: Melanolestes abdominalis


Wing Form
Color of
Abdomen Sex N Brachyterous Macropterous

Black M 95 0 95
F 79 59 20
Red M 152 0 152
F 31 27 4
Red Connexivum F 41 41 0
Intermediate M 20 0 20
F 24 22 2

TOTAL 442 149 293

other but were, instead, intermediate in color. 'Intermediate' included those specimens
in which the general abdominal color was neither red nor black but somewhere in
between, and those in which the abdominal color was a mosaic (e.g., red sterna with
black spots, larger black patches, irregular black stripes, or black, transverse, interseg-
mental lines). Therefore, color cannot be used to separate the animals into two forms.
Recall that Readio (1927), from his laboratory studies, concluded that color was not
inherited but probably resulted from laboratory conditions. Blatchley (1926) stated that
the two forms could be separated by the frequent presence of brachypterous females
in picipes. However, his diagnosis of abdominalis was based only on macropterous
individuals of both sexes. As can be seen from Table 1, all males were macropterous,
thus agreeing with Readio (1927), and females were brachypterous or macropterous for
all colors. Thus, the wing form character is invalid for distinguishing the two "species."
Finally, Blatchley (1926) used ocellar size to distinguish between the two forms (i.e.,
ocelli smaller in picipes). It is obvious from a cursory examination that ocellar size does
vary (Table 2), but not in the way stated by Blatchley. Again, forcing the specimens
into the artificial categories of black and red shows that ocelli did not differ significantly
between the two forms but, if anything, tended toward larger size in picipes. Those
with a red connexivum (all females) had significantly smaller ocelli than black and red
individuals, so much so that when combined with red individuals, this combined group
also differed from black individuals (Table 2). Again, this is just the opposite of what
Blatchley said (i.e., red individuals have larger ocelli). Those intermediate in color did
not differ from red individuals or from the red plus red connexivum group but did from
black individuals and those with a red connexivum.
What do the above results concerning ocellar size indicate? First, Blatchley's state-
ments were in error (i.e., ocelli are smaller in picipes, larger in abdominalis). This
undoubtedly resulted from his small sample size of abdominalis (6 males, 8 females),
and because the sample included only macropterous individuals. Also, the two red forms
(red and red connexivum) demonstrate the unreliability of color because they differ in
ocellar size but are supposed to be two forms of the same species. Finally, keeping in
mind that the color categories in our study were somewhat arbitrary, we also found
that ocellar size was not a valid character for separating the two "species." But there
is obviously a difference in ocellar size. Is there a pattern? Excluding color as a factor
in the analyses, but including sex and wing form (macropterous and brachypterous), it
can be seen that males, which are always macropterous, have the largest ocelli; macrop-
terous females, an intermediate size; and brachypterous females, the smallest ocelli of
all (Table 2). That is, ocellar size is linked to sex and wing form.

Florida Entomologist 74(3)


September, 1991


Wing Ratio
Color Sex Form N x t SE a df Prob.

Black 174 1.160.04 A
Red 183 1.070.02 A 258.1 0.02b
Black 174 1.16 0.04 A
Red 183 1.07 .0.02 A
Red Connexi- 41 0.52+ 0.01 B 2,395 0.00c
Black 174 1.160.04 A
Red + Red Con- 224 0.970.02 B 287.4 0.00b
Black 174 1.160.04 A
Red 183 1.070.02 AB
Intermediate 44 0.92 0.06 B
Red Connexi- 41 0.52 0.01 C 3,438 0.0O
Black 174 1.16+0.04 A
Red + Red Con- 224 0.970.02 B
Intermediate 44 0.92 0.06 B 2,439 0.0O
M Macropterous 267 1.270.02 A
F Macropterous 26 1.07 0.04 B
F Brachypterous 149 0.62 0.01 C 2,439 0.00'

aMeans followed with same letter within column within each cell are not significantly different at the 0.01 probability
level using Duncan's Multiple Range Test.
bStudent's t test.
'One way ANOVA.

As noted above, in addition to Blatchley's characters, we also examined the male
and female genitalia of the black and red (including red connexivum) forms to determine
if there were any obvious differences. Female genitalia (Fig. 2a) were highly uniform
across color forms and followed the pattern outlined by Davis (1966) for peiratine re-
duviids. Male pygophores (Fig. 2b, c) and phalluses (Fig. 2d) were asymmetrical as is
typical for Peiratinae. Pygophores were notable for the large medial process (MP) that
arose from the ventral rim, leaned to the right (Fig. 2b), and curled cephalad (Fig. 2c).
The left paramere (LPM) was always larger than the right paramere (RPM). The pedicel
(PD) of the phallus (Fig. 2d) curled to the left as it descended to articulate with the
strut (STR). The caudally directed strut angled from left to right. The dorsal phallothecal
sclerite (DPS) was largely absent from the left half of the phallus. None of these structures
varied significantly between individuals of the two color forms.
From our results, therefore, we conclude that abdominalis is not a valid species and
must be treated as a synonym of picipes.


Melanolestes picipes (Herrich-Schaeffer)
Pirates picipes Herrich-Schaeffer 1846, p. 62.
Pirates abdominalis Herrich-Schaeffer 1846, p. 63.

McPherson et al.: Melanolestes abdominalis

1 GX

1 mm 1 mm

1 mm

0.5 mm

Fig. 2. Male and female genitalia of Melanolestes picipes: A. female, ventroposterior
view. B. male, pygophore, ventroposterior view. C. male, pygophore, lateral view (left
paramere removed). D. male, phallus, dorsal view. Abbreviations: BF, basal foramen;
BP, basal plate; DF, ductifer; DPS, dorsal phallothecal sclerite; 1GPO, first
gonapophysis; 1 GX, first gonocoxa; LPM, left paramere; MP, medial process; PB, basal
plate bridge; PD, pedicel; PR, proctiger; RPM, right paramere; STR, strut; T9, tergum
9; T10, tergum 10.

(For more complete taxonomic history, see Froeschner [1988] and Maldonado Capriles
Body elongate-oval. Color varying from entirely black or reddish black to head and
thorax dark brown to black and abdomen with varying amounts of red (i.e., only con-
nexivum red to terga and sterna red); tarsi brownish to black; hemelytral membrane
dark brown to black.
Compound eye with width subequal to or, most often, less than interocular space,
the latter most evident in females. Antennal segment 2 with most hairs more than
one-half width of segment in males, often shorter than one-half in females, those of
males almost perpendicular to long axis of segment giving it a bristly appearance, those
of female forming about 300 angle to subparallel to segment. Ocelli varying from large
to small, those of male subequal in width or wider than interocellar space, those of
female either wide as those of males (generally macropterous individuals) or narrower
than interocellar width (generally brachypterous individuals).

402 Florida Entomologist 74(3) September, 1991

Pronotum with posterior lobe finely rugose-granulate, humeral angles rounded. Males
macropterous, females macropterous or brachypterous; wings in macropterous form
reaching to slightly exceeding tip of abdomen, those in brachypterous form not surpassing
tergum 3.
Length, 12.0-20.0 mm.
Distribution (taken in part from Froeschner [1988]): AL, AR, AZ, CA, CO, CT, DC,
NJ, NM, NY, OH, OK, PA, RI, SC, SD, TN, TX, UT, VA (Quebec, Mexico to Brazil,


We wish to thank the following individuals for the loan of materials from their
respective institutions: R. W. Brooks, Snow Museum, University of Kansas, Lawrence;
R. C. Froeschner, National Museum of Natural History, Washington, D.C.: M. F.
O'Brien, University of Michigan Museum of Zoology, Ann Arbor; N. D. Penny, California
Academy of Science, San Francisco; A. V. Provonsha, Department of Entomology,
Purdue University, West Lafayette, IN; R. T. Schuh, American Museum of Natural
History, NY; and R. W. Sites, Department of Entomology, Texas Tech University,


BANKS, N. 1910. Catalogue of the Nearctic Hemiptera-Heteroptera. Amer. Entomol.
Soc., Philadelphia. 103 pp. + i-viii.
BLATCHLEY, W. S. 1926. Heteroptera or true bugs of eastern North America with
especial reference to the faunas of Indiana and Florida. Nature Pub. Co., In-
dianapolis. 1116 pp.
BRIMLEY, C. S. 1938. The insects of North Carolina being a list of the insects of North
Carolina and their close relatives. North Carolina Dep. Agric. Div. Entomol.
(Raleigh). 560 pp.
CHAMPION, G. C. 1899 (1897-1901). Fam. Reduviidae, in Insecta. Rhynchota. Hemip-
tera-Heteroptera. Vol. II. Biologia Centrali-Americana. London. xvi + 416 pp.,
22 plates. (1897: 1-32; 1898: 33-192; 1899: 193-304; 1900: 305-344; 1901: i-xvi
+ 345-416).
COSCARON, M.D.C. 1983. Nuevas citas de distribution geografica para la subfamilia
Peiratinae (Insecta, Heteroptera, Reduviidae). Revista de la Sociedad En-
tomologica Argentina 42: 369-382.
DAVIS, N. T. 1966. Contributions to the morphology and phylogeny of the Reduvioidea
(Hemiptera:Heteroptera). Part III. The male and female genitalia. Ann. Entomol.
Soc. Amer. 59: 911-924.
DREW, W. A. AND K. SCHAEFER. 1963. The Reduvioidea of Oklahoma (Hemiptera).
Proc. Oklahoma Acad. Sci. 43: 98-112.
ELKINS, J. C. 1951. The Reduviidae of Texas. Texas J. Sci. 3: 407-412.
FRACKER, S. B. 1912. A systematic outline of the Reduviidae of North America. Proc.
Iowa Acad. Sci. 19: 217-252.
FROESCHNER, R. C. 1944. Contributions to a synopsis of the Hemiptera of Missouri,
Pt. III. Lygaeidae, Pyrrhocoridae, Piesmidae, Tingididae, Enicocephalidae,
Phymatidae, Ploiariidae, Reduviidae, Nabidae. Amer. Midland Natur. 31: 638-
FROESCHNER, R. C. 1988. Family Reduviidae, p. 616-651. In T. J. Henry and R. C.
Froeschner (eds.). Catalog of the Heteroptera, or true bugs, of Canada and the
continental United States. E. J. Brill, New York. 958 pp.
HERRICH-SCHAEFFER, G.A.W. 1846. (1836-1853). Die Wanzenartigen Insecten. Get-
reu nach der Natur abgebildet und beschrieben. C. H. Zeh'schen Buchhandlung,
Nurnburg 8: 49-100.

McPherson et al.: Melanolestes abdominalis 403

MALDONADO CAPRILES, J. 1990. Systematic catalogue of the Reduviidae of the world.
Caribbean J. Sci. (Special Edition), Univ. Puerto Rico. 694 pp.
PARSHLEY, H. M. 1917. Fauna of New England. 14. List of the Hemiptera-Heterop-
tera. Occasional Papers Boston Soc. Natur. Hist. 7: 1-125.
PARSHLEY, H. M. 1918. Hemipterological notes. Psyche 25: 64-65.
PARSHLEY, H. M. 1922. Report on a collection of Hemiptera-Heteroptera from South
Dakota. South Dakota State Coll. Tech. Bull. 2: 1-22.
PROVANCHER, A. L. 1887 (1885-1890). Petite faune entomologique du Canada et
particulibrement de la Province de Quebec. Vol. III. CinquiBme Ordre Les H6mip-
thres. Quebec. 354 pp.
READIO, P. A. 1927. Studies on the biology of the Reduviidae of America north of
Mexico. Univ. Kansas Sci. Bull. 17: 5-291.
SAS INSTITUTE. 1988. SAS/STAT user's guide, version 6, 4th ed. Vols. 1 & 2. SAS
Institute, Cary, North Carolina.
SCUDDER, G.G.E. 1959. The female genitalia of the Heteroptera: morphology and
bearing on classification. Trans. Roy. Entomol. Soc. London 111(14): 405-467.
SLATER, J. A. AND R. M. BARANOWSKI. 1978. How to know the true bugs (Hemiptera-
Heteroptera). Wm. C. Brown Pub., Dubuque. 256 pp.
SMITH, J. B. 1909. A report of the insects of New Jersey. Annu. Rep. New Jersey
State Mus., 1909, pp. 15-888.
STAL, C. 1872 (1870-1876). Enumeratio Hemipterorum. Bidrag till en forteckning ifver
alla hittills kanda Hemiptera, jemte systematiska meddelanden. Parts 1-5. Kong-
liga Svenska Vetenskaps-Akademiens Handlingar 1870, part 1, 9(1): 1-232; 1872,
part 2, 10(4): 1-159; 1873, part 3, 11(2): 1-163; 1874, part 4, 12(1): 1-186; 1876,
part 5, 14(4): 1-162.
TORRE-BUENO, J. R. de la. 1923. Family Reduviidae, p. 677-692 in W. E. Britton.
Guide to the Insects of Connecticut. Part IV. The Hemiptera or sucking insects
of Connecticut. Connecticut State Geol. Natur. Hist. Surv. Bull. 34: 1-807.
UHLER, P. R. 1876. List of Hemiptera of the region west of the Mississippi River,
including those collected during the Hayden explorations of 1873. Bull. U.S. Geol.
Geog. Surv. Terr. 2(1): 269-361 + plates 19-21.
UHLER, P. R. 1878. Notices of the Hemiptera Heteroptera in the collection of the
late T. W. Harris, M. D. Proc. Boston Soc. Natur. Hist. 19: 365-446.
UHLER, P. R. 1884. Order VI.-Hemiptera, p. 204-296 in J. S. Kingsley (ed.). The
standard natural history. Vol. II. Crustacea and insects. S. E. Cassino & Co.,
UHLER, P. R. 1886. Check-list of the Hemiptera Heteroptera of North America.
Brooklyn Entomol. Soc., Brooklyn. 32 pp.
VAN DUZEE, E. P. 1916. Check list of the Hemiptera (excepting the Aphididae,
Aleurodidae and Coccidae) of America, north of Mexico. New York Entomol.
Soc., New York. i-xi + 111 pp.
VAN DUZEE, E. P. 1917. Catalogue of the Hemiptera of America north of Mexico
excepting the Aphididae, Coccidae and Aleurodidae. Univ. California Pub., Tech.
Bull., Entomol. 2: i-xiv + 902 pp.
WYGODZINSKY, P. W. 1949. Elenco sistematico de los Reduviiformes Americanos.
Inst. Medicine Regional, Tucuman, Pub. 473, Monografia 1: 1-102.

404 Florida Entomologist 74(3) September, 1991


Department of Zoology, University of the West Indies
Mona, Kingston 7, Jamaica


The present report is the first ever on the acaricidal activity of marine plant extracts
on Boophilus microplus. Topical application of crude ethanol extracts of five marine
algae; namely Laurencia obtusa, Padina vickerisiae, Liagorafarinosa, Liagora elongata
and Stypopodium lobatum; affected the survival of engorged adult female B. microplus
and inhibited oviposition and embryogenesis in the ticks.
The order of toxicity of the extracts (% adult mortality) on B. microplus was Lauren-
cia obtusa (40.00%) > Liagora elongata (30.00%) Liagorafarinosa (10.00%) = Padina
vickerisiae = Stypopodium lobatum. However, the order of inhibition of embryogenesis
was different from adult mortality [Laurencia obtusa (59.23%) > Liagora farinosa
(38.75%) > Stypopodium lobatum (34.40%) > Padina vickerisiae (14.00%) > Liagora
elongata (11.21%)].


Este es el primer report sobre la actividad de acaricidas obtenidos a partir de
extractos de plants marinas sobre Boophilus microplus. La aplicacion de extractos
crudos de cinco algas marinas, Laurencia obtusa, Padina vickerisiae, Liagorafarinosa,
Liagora elongata y Stypopodium lobatum, en etanol, afectan e inhiben la oviposicion y
la genesis embriologica de las garrapatas.
La escala de toxicidad de adults de B. micropulus fue: Laurencia obtusa (40.00%)
> Liagora elongata (30.00%) Liagora farinosa (10.00%) = Padina vickerisiae =
Stypopodium lobatum. Sinembargo, la escala de inhibicion de genesis embriologica fue
diferente de la mortalidad de adults: Laurencia obtusa (34.40%) > Padina vickerisiae
(14.00%) > Liagora elongata (11.21%).

Extensive investigations of terrestrial natural products over the last two decades to
replace the persistent organochlorine and organophosphorous pesticides have been done
(Grainge et al, 1986). The need for more effective acaricides for the Caribbean region
is demonstrated by the annual financial loss from the livestock industry due to tick
infestation. Rawlins & Mansingh (1987) estimated this loss to be US$62 million per year
(Commonwealth Caribbean only).
Green plants tend to produce secondary compounds which reduce or prevent grazing
by insects (Fraenkel 1959, Whittaker & Feeny 1971, Levin 1976, Harborne 1977, Rosen-
thal & Janzen 1979). It is well established that some of these secondary compounds are
toxic to insects and could be used as insecticides, antifeedants etc. (Ali et al. 1985,
Ahmed & Grainge 1985). Similarly, marine plants (algae) are known to produce secondary
metabolites which function as predator defense (Paul & Fenical 1982, 1986, Whylie &
Paul 1989, Bakus 1981, Coll et al. 1982, Gerhart 1984, La Barre et al. 1986, Pawlirk et
al. 1987). However, their acaricidal potential is unknown.
Beaver (1975) revealed that extracts from some terrestrial plants which are insecti-
cidal are usually toxic to fishes. Similarly, extracts from various species of marine algae

Williams: Acaridical Activity of Marine Algae Extracts 405

from the genus Laurencia are toxic to various fishes (Minott 1988, Minott & Pascoe
1987). These algae are rarely grazed by marine fishes. Their extracts may contain
compounds which are insecticidal. Ticks and insects are closely related physiologically
(Telfer 1965), thus toxins affecting insects may affect ticks also. The above relations
form the basis of the present investigations.


Algae known to yield extracts toxic to fishes (i.e. Laurencia spp.) as well as algae
which are not grazed by fishes and are not heavily calcified (i.e. Liagora spp.) were
selected for investigation.


Algae were collected from the sea at Discovery Bay Marine Laboratory (in the parish
of St. Ann, Jamaica), washed with distilled water, and allowed to drip-dry. Samples
weighing 20g were then chopped into small pieces with the laboratory blender and
extracted with 200ml of 95% ethanol for five days at 30 3C and 60-70% RH.
Extracts were decanted from residues and ethanol evaporated to dryness and a 10%
(w/v) concentrate in acetone prepared fro acaricidal testing.

Test Tick

Fully engorged adult female Boophilus microplus (Canesterini) weighing between
150-180mg were selected for bioassay. This species of hard tick has developed signifi-
cantly high levels of resistance to acaricides and is of economic importance, Rawlins &
Mansingh (1987).


Thirty ticks in three replicates of ten were used for each extract. Ten ul (1 ug) of
extracts were topically applied to the dorsum of ticks using a Hamilton micro-applicator.
The controls were treated with 10 ul of acetone only.
Treated and control ticks were kept at 27C and 60-70% RH in covered petri dishes
and allowed to lay eggs until dead. Ticks were classified as dead if they failed to show
appendicular responses and failed to produce eggs. Adult mortality after 96 h was
recorded. The mean weight of egg masses were determined for each treatment after 12
Eggs were placed in test tubes which were plugged with moistened cotton wool.
Cotton wool was moistened every other day or as required to maintain a 80-90% RH to
aid hatching of eggs.
After the eggs hatched three sub samples of 100-150 egg shells were collected from
each test tube and the percentage of unhatched eggs and egg shells determined. From
these data the inhibition of hatching and of reproductive potential were determined as
shown below:

1. % Inhibition of oviposition =
Mean wt of control eggs minus mean wt. of treated eggs x 100
Mean wt. of control eggs

Florida Entomologist 74(3)

September, 1991

2. % Inhibition of hatching = % of unhatched eggs in treated minus % of unhatched
eggs in control.

3. Inhibition of reproductive potential =
Inhibition of oviposition x Inhibition of hatching

Mean percentages were separated using a One Way Analysis of Variance (ANOVA)
followed by Student Newman-Keuls Multiple range test (a = 0.05). All mean percentages
were transformed to their respective Arc-Sine values before ANOVA was done (Zar 1974).


Table 1 presents results on the biological effects of extracts on B. microplus. The
data revealed that the extract obtained from Laurencia obtusa was the most toxic to
B. microplus (comparing adult mortality). The L. obtusa extract was 4.0 times more
effective in killing adult B. microplus than the extracts obtained from Liagorafarinosa,
Padina vickerisiae, and Stypopodium lobatum. However, the extract obtained from L.
obtusa was not significantly more toxic than the extract from Liagora elongata (Table 1).
These findings suggest that the active principles of the extracts may have different
modes of action on B. microplus. The ability of extracts to inhibit oviposition in ticks
could be attributed to the inhibition of nervous mechanism involved in the release of
the developed oocytes from the ovaries, as revealed by Stendel & Andrews (1973).
The overall effects of the extracts in reducing the reproductive potential of ticks are
also presented in Table 1. The order of extracts in inhibiting the reproductive potential
in B. microplus was Laurencia obtusa > Stypopodium lobatum > Liagorafarinosa >
Padina vickerisiae > Liagora elongata.
Rawlins (1977) reported the following 96 h mortality data for adult B. microplus
treated with some acaricides; Carbaryl (0.0%), lindane (12.0%), chlorodimeform (0.0%),
dioxathion (0.05%), naled (20.35%) and fenitrothion (0.0%), all tested at 1.0 ug/tick. The


96 hour
mean Mean S.E. inhibition of:'
adult Mean
mortality egg wt. oviposi- Reproduc-
Marine algae S.E. (mg) + S.E. tion Hatching tive potential

Laurencia 46.30a 36.80a 59.11a 59.23a 36.25a
obtusa 2.3 1.02 +3.43 1.5 0.8
Liagora 36.00a 57.59b 36.01b 11.21b 4.04b
elongata 0.8 2.56 0.9 0.42 +1.0
Liagora 16.00b 68.55c 23.83bc 38.75c 9.23bc
farinosa 0.91 0.98 2.5 2.0 0.6
Padina 16.00b 59.78bc 33.58c 14.01b 4.70b
vickerisiae 0.4 4.52 0.70 1.5 0.01
Stypopodium 16.00b 36.80a 59.11a 34.41c 20.34c
lobatum 1.4 3.20 0.25 0.8 1.5
Control 6.00c 90.05d -
_.3 4.36

'Means in a column with same letters are not significantly different from each other (S.N.T. P<0.05).


Williams: Acaridical Activity of Marine Algae Extracts 407

extracts from Laurencia obtusa and Liagora elongata were more active than these
acaricides, causing 40.3% and 30.0% mortality respectively.
The present paper provides data supporting the potential of marine algae extracts
as sources of effective natural product acaricides.


I wish to thank Ms. Margaret Jones of the Department of Zoology, University of
the West Indies, for the Spanish translation of the abstract.


AHMED, S., AND GRAINGE, M. 1985. Potential of the neem tree (Azadirachta indica)
for pest control and rural development. Econ. Botany 40: 202-209.
ALI, S. I., SINGH, O. P., AND MISRA, U.S. 1985. Effectiveness of plant oils against
pulse beetle Callosobrochus chinensis. Linn. Indian J. Entomol. 48: 6-9.
BAKUS, G. J. 1981. Chemical defense mechanism on the Great Barrier Reef. Science
211: 497-499.
BEAVER, B. 0. 1975. Medicinal plants in tropical West Africa. Cambridge Univ. Press.
G. T. 1982. Chemical defense in soft corals (Coelenteratea: Octocorallia) of the
Great Barrier Reef: a study of comparative toxicities. Mar. Biol. Prog. Ser. 8:
FRAENKEL, G. S. 1959. The raisson d'etre of secondary plant substance. Science 129:
GERHART, D. J. 1984. Postaglandin A2, an agent of chemical defense in the Caribbean
gorgonian Plexaura homomalla Mar. Ecol. Prog. Ser. 31: 255-263.
GRAINGE, M., AHMED, S., MITCHELL, W. C., AND HYLIN, J. W. 1986. Plant species
reportedly possessing pest control properties. An EWC/UH database resource
system institute. East-West Centre, Honolulu, HI, USA 249p.
HARBONE, J. D. 1977. Introduction to ecological biochemistry. Academic Press, New
York, pp. 104-129.
LA BURRE, S. C., COLL, J. C., AND SAMMARCO, P. W. 1986. Defensive strategies
of soft corals (Coelenterata: Octocorallia) of the Great Barrier Reef 11. The re-
lationship between toxicity and feeding deterrence. Biol. Bull. (Woods Hole,
Mass.) 171: 565-576.
LEVIN, D. A. 1976. The chemical defenses of plant to pathogens and herbivores. Ann.
Rev. Ecol. Syst. 7: 121-159.
MINOTT, A. D. 1988. Chamigranes from Jamaican Laurencia and an approach to
chamigrane synthesis. Ph.D. thesis, University of the West Indies, Mona,
Jamaica. 299 p.
MINOTT, A. D., AND PASCOE, K. 0. 1987. Chemical and biological investigation of
some Jamaica sea weeds (Laurencia obtusa and Laurencia papillosa) Proc. First
Ann. Nat. Conf. on Sci. and Tech. April 27-29. Sci. Res. Council. of Jamaica.
PAUL, V. J., AND FENICAL, W. 1982. Toxic feeding deterrents from the tropical
marine algae Caulerpa bikinensis (Chorophyta). Tetrahedron Lett. 23: 5017-
PAWLIK, J. R., BURCH, M. T., AND FENICAL, W. 1987. Patterns of chemical defense
among Caribbean gorgonian corals: A preliminary survey. J. Exp. Mar. Ecol.
108: 55-66.
RAWLINS, S. C. 1977. Toxicological and biological studies on Jamaican and other
Caribbean population of the cattle tick Boophilus microplus (Canestrini) (Acarina:
Ixodidae) Ph.D. thesis, University of the West Indies, Mona. 227p.

408 Florida Entomologist 74(3) September, 1991

RAWLINS, S. C., AND MANSINGH, A. 1987. A review of tick and screwworms affecting
livestock in the Caribbean. Insect Sci. Applic. 8: 259-267.
ROSENTHAL, G. A., AND JANZEN, D. H. 1979. Herbivores: Their interaction with
secondary plant metabolites. Academic Press New York. 718p.
STENDEL, W., AND ANDREWS, P. 1973. The development of a new compound active
against resistant ticks. Proc. 7th British Insecticide and Fungicide Conference.
pp. 282-89.
TELFER, W. H. 1965. The mechanism and control of yolk formation. Ann. Rev. En-
tomol. 10: 161-184.
WHITTAKER, R. H., AND FEENY, P. 1971. Allelochemic: Chemical interactions be-
tween species. Science 171: 757-770.
WYLIE, C. R., AND PAUL, V. J. 1989. Chemical defenses in three species of Sinularia
(Coelenterata, Alcyonaceae): Effects against generalist predators and the but-
terfly fish Chaetodon unimaculatus BlochJ. Exp. Mar. Biol. Ecol. 129: 141-160.
ZAR, J. H. 1974. Biostatistical analysis. Prentice-Hall. New York, 620p.

-.440-400- -- -- -- -


'Insect Attractants, Behavior and Basic Biology Research Laboratory,
Agricultural Research Service, U.S. Department of Agriculture
Gainesville, Florida 32604
2Pineapple Development Corporation of Costa Rica, S.A.
Apartado 4084, San Jose, Costa Rica
'Guatemala Medfly Methods Station, APHIS, USDA
American Embassy, APO Miami, Florida 34024


Papaya fruit flies, Toxotrypana curvicauda Gerstaecker, were observed on papaya
trees in Guatemala and Costa Rica to compare with reported patterns of behavior for
papaya fruit flies in south Florida. In both cases, males and females were nearly always
on fruit and not on leaves. General activity of both sexes and female oviposition were
highest in the morning in Central America, contrasting with the late afternoon activity
period in Florida. A total of 23 mating pairs was observed in Costa Rica, all on papaya
trees in late morning, compared to late afternoon to dusk in Florida. A fruit model trap
baited with the pheromone 2-methyl-6-vinylpyrazine caught significant numbers of both
male and female T. curvicauda in Costa Rica at a pheromone release rate of 1 pg/h. At
this location, counts of flies in plots on papaya trees versus traps indicated a high rate
of capture of both sexes of papaya fruit flies with the fruit-model sex pheromone trap.


Se observe el comportamiento de las moscas Toxotrypana curvicauda Gerstaecker
de la fruta en plants de papaya en Guatemala y Costa Rica y se compare dicho compor-
tamiento con el observado en moscas de papaya en Florida. En ambos casos, los machos
y las hembras fueron observados casi siempre en las frutas y no en las hojas. La actividad

Landolt et al.: Papaya Fruit Fly Behavior

de ambos sexos, y la oviposicion de las moscas en Costa Rica se observe durante la
mafiana en contrast con la actividad de las moscas de Florida la cual ocurre en las horas
de la tarde. Un total de 23 parejas en copula fue observado en Costa Rica en horas
avanzadas de la mafiana, comparado con la actividad de las moscas de Florida durante
el atardecer. Un modelo de trampa simulando una fruta, con la feromona 2-methyl-6-vin-
ylpyrazyne utilizada como atrayente, y con una velocidad de dispersion de 1 ug/h, capture
un numero significativo de hembras y machos en Costa Rica. En este lugar, la capture
del numero de moscas en plants de papaya comparado con el numero de moscas capturado
en las trampas, indico un numero alto de capture de moscas de ambos sexos en la trampa
simulante de una fruta y utilizando la feromona como atrayente.

The papaya fruit fly, Toxotrypana curvicauda (Gerstaecker), occurs throughout
much of the Caribbean, Central America, and South America (Wolcott 1933), where it
is the principal insect pest of papaya fruit. Until recently, there was little information
on the biology and behavior of this pest, and no lures or traps for monitoring or detection.
Studies have been conducted on the papaya fruit fly in South Florida on adult behavior
(Landolt & Hendrichs 1983, Landolt 1984a), on the male sex pheromone (Chuman et al.
1987), and on behavioral responses to male sex pheromone (Landolt et al. 1985, Landolt
& Heath 1988, Landolt et al. 1988). Suggested methods of cultural control for the papaya
fruit fly have been made (Landolt 1984b) and a pheromone-baited trap was developed
for monitoring and possibly control of papaya fruit flies (Landolt et al. 1988, Landolt
and Heath 1990).
In the continental U.S., commercial papaya production is restricted to the southern-
most area of the Florida peninsula. Papaya cultivation and infestations of papaya fruit
flies are much more prevalent in Central America, northern South America and the
Caribbean Islands. To determine if research findings on the papaya fruit fly in south
Florida are generally applicable to papaya fruit fly infestations in other areas, observation
and trapping studies were conducted in papaya plantings in Guatemala and Costa Rica
for comparison. This paper reports the results of those studies.


Field observations of papaya fruit fly activities were made in three small (< 1 hectare)
plantings of papaya near Retalhuleu, Guatemala, 17-23 June 1986 and 3-21 March 1987,
and in a large (> 100 hectares) commercial papaya plantation near Buenos Aires, Costa
Rica, 11-18 May 1988 and 1-2 February 1989. Bihourly visual surveys of papaya fruit
fly activity on papaya trees were made on 4 days in June 1986, and 3 days in March
1987 in 3 small plantings totalling about 0.7 hectares near Retalhuleu, Guatemala. Fly
sightings and activities were recorded from 0600 to 1800 hours, near sunrise to sunset.
The sex and location of flies sighted, as well as male calling, female oviposition, fly
interactions, and matings were noted. These observations were made in areas adjacent
to pheromone trap plots described later. A similar study was made of papaya fruit fly
behavior in a papaya plantation near Buenos Aires, Costa Rica, over 5 days in May
1988. Observations were made in a 4-row strip (5 meters) about 60 meters long, along
the eastern edge of the plantation bordering the Rio Volcan. Bihourly observations were
made from 0600 to 1600 hours with sunrise and sunset near 0600 and 1800 hours respec-
tively. The sex and location of flies sighted, as well as their activities, were recorded.
On 1-2 February 1989, male and female papaya fruit flies sighted were tallied per
observation period with no behavioral observations made. Observed patterns of ovipos-
ition and mating in Florida were made using the methods of Landolt & Hendrichs (1983).


410 Florida Entomologist 74(3) September, 1991

Trapping tests were set up in both locations to evaluate the attractiveness of the
pheromone 2-methyl-6-vinylpyrazine (2,6-MVP) and efficacy of the fruit-model trap de-
veloped in Florida (Chuman et al. 1987, Landolt et al. 1988) for the papaya fruit fly in
these areas. The trap used consisted of a 14-cm-diam. dark green sphere coated with
Tanglefoot (The Tanglefoot Co., Grand Rapids, MI) and baited with glass capillary
lures loaded with 2,6-MVP as described by Landolt & Heath (1988). The desired release
rate was obtained by using a 25 pl micropet (0.6 mm i.d.) with an air column over the
pheromone reservoir. Lures were attached to spheres near the top as described by
Landolt et al. (1988). Traps were hung from leaf petioles near, but not touching, fruit
clusters. In Guatemala, two randomized complete blocks of 5 traps baited with dispensers
providing different release rates of 2,6-MVP were originally set up on 3 March, 1987,
in 2 separate plantings. Estimated release rates of the 5 treatments were 0, 80, 160,
320, and 1060 ng/h. These two blocks were maintained and checked daily until 12 March.
Another block of 5 treatments was set up and maintained 10-12 March in a third papaya
planting, and a fourth block was maintained from 12-21 March. Traps in all blocks were
placed in papaya trees about 5 meters apart. In Costa Rica, a trapping experiment was
set up on 13 January 1989 adjacent to the observation area described comparing unbaited
traps to traps with pheromone lures releasing 1 pg/h. Five pairs of pheromone-baited
and unbaited traps were set up in a N-S line, along the eastern edge of the papaya
plantation. Traps were placed about 5 meters apart. Traps were checked and captured
flies removed on 31 January and 1-2 February, at 1600 hours. Trap catch data from the
Costa Rica test were analyzed using a paired t-test (Steel & Torrie 1960).


The papaya fruit fly observed near Retalhuleu, (and also San Jose de Puerto) in
Guatemala and near Buenos Aires, Puntarenas, Costa Rica, differed in appearance from
those collected, observed, and trapped in Dade County, Florida. All papaya fruit flies
seen in Central America during this study were brown and yellow in color, compared
to the brown and darker orange-yellow of those in Florida. Knab & Yothers (1914) noted
the papaya fruit fly resemblance in size, form, color, and behavior, to Polistes wasps,
which may be Batesian mimicry (Bates 1862). In this study, the color patterns observed
were similar to vespid social wasps collected in the respective areas. In Dade County,
Florida, the coloration is similar to that of Mischocyttarus mexicanus (Saussure),
Polistes dorsalis (Fab.), and a local race of Polistes exclamans Vierick. In Guatemala
Mischocyttarus sp. and Stelopolybia areata (Sat) were common in the study area, with
coloration similar to that of the local papaya fruit fly.
General diel activity patterns observed in both Guatemala and Costa Rica were quite
different from those observed in Florida. Near Retalhuleu, Guatemala, most fly sightings
on papaya trees were from 0800 to 1400 hours, with males preceding females somewhat
(Fig. 1). In Costa Rica near Buenos Aires, most sightings occurred from 1000 to 1200
hours (Fig. 1). During these studies 34 male and 19 female papaya fruit flies sighted at
the Guatemala site and 334 males and 160 females were sighted at the Costa Rica site.
In both locations, these observations differed from pattens exhibited by papaya fruit
flies studied in Dade County, Florida, where fly activity was concentrated principally
in the 2 h preceding sundown (Landolt & Hendrichs 1983). Temperatures recorded
during peak periods of activity were 22-250C in Dade Co., Florida (February), 33-36C
at the Guatemala study site and 28-34C at the Costa Rica study site. Shifts in diel
activity patterns are possibly due to responses to temperature differences. However,
in subsequent field experiments in Florida in June, 1987, with daytime temperatures
ranging up to 290C, papaya fruit fly activity was still concentrated near dusk (unpublished

Landolt et al.: Papaya Fruit Fly Behavior


0600 0800 1000 1200 1400 1600 1800



0600 0800 1000 1200 1400 1600


Fig. 1. Percentages of daily totals of male and female papaya fruit flies observed in
papaya groves at bihourly intervals. Solid bars are for females, cross hatched bars are
for males. Retalhuleu, Guatemala and Buenos Aires, Costa Rica.

Florida Entomologist 74(3)


0600 0800 1000 1200 1400 1600



700 900 1100 1300 1500 1700


Fig. 2. Percentages of daily totals of ovipositing females and mating pairs of papaya
fruit flies sighted in a papaya plantation at bihourly intervals. Solid bars are for ovipos-
itions, cross hatched bars are for matings. Buenos Aires, Costa Rica and Dade County,

September, 1991

Landolt et al.: Papaya Fruit Fly Behavior 413

The diel pattern of female oviposition at the Costa Rica study site was distinct with
most ovipositions observed in late morning (1000-1200 hours) (Fig. 2). In Dade Co.,
Florida most ovipositions were observed in the last 2 h of daylight, but with lesser
numbers occurring throughout the day (Landolt & Hendrichs 1983). The few ovipositions
observed in Guatemala were not recorded.
Mating pairs of papaya fruit flies were observed on papaya trees at both the Guatemala
and Costa Rica sites. At the Guatemala sites, 3 mating pairs of papaya fruit flies were
observed in June 1986. All 3 were near fruit clusters but on the trunk or on leaf petioles,
and were sighted at 1400, 1430, and 1500 hours. These sightings were not made during
systematic observations. A total of 23 mating pairs of papaya fruit flies were sighted
in 5 days of bihourly observations made at the Costa Rica site in May 1988. All were
recorded from 1000 to 1200 hours, with most at 1000 hours (Fig. 2). Again, the time of
day of mating in papaya groves differed from that observed in Florida. Of 13 matings
observed during the study of T. curvicauda activity in the field in Florida (Landolt &
Hendrichs 1983), all were from 1400 to 1700 hours.
A total of 37 papaya fruit flies were trapped in Guatemala with fruit model traps
baited with 2-methyl-6-vinylpyrazine (19 females, 18 males). Most flies (15 females, 12
males) were found in traps with the 2 highest release rates (320-1060 ng/h) (Table 1).
A total of 196 papaya fruit flies was caught in traps in Costa Rica, (140 males and 56
females) in 3 days using a 2,6-MVP release rate of about 1 pg/h. Catches of both males
and females on these pheromone-baited traps were significantly higher than on unbaited
control traps (Table 1). In trapping tests conducted in Florida, highest trap catches
were obtained with pheromone release rates of 140 to 900 ng/h.
We conclude from these studies that the pheromone-baited fruit-model trap developed
in Florida for the papaya fruit fly is efficacious in Central America for trapping papaya
fruit fly males and females. Although the trap is probably useful for monitoring general
activity patterns, additional research is necessary to develop an understanding of the
relationship between trap catches and population levels. At the Costa Rica study site
on 1-2 February 1989, numbers of papaya fruit flies sighted in bihourly observations
(total of 48 females, 81 males) were comparable to the number trapped (42 females, 77


Release Rate (ug/h)

0 .08 .16 .32 1.06

Females 0.0 0.1 0.1 0.3 0.5
Males 0.1 0.1 0.1 0.4 0.2

Release Rate (ug/h)

0 1.0

Costa Rica
Females 1.5 1.4 2.8 2.71
Males 2.9 1.8 6.8 4.3'
'Significantly greater than unbaited (0 dose) traps by paired T-test at p <0.01. For catches of females, t = 4.91,
d.f. = 13. For catches of males, t = 4.81, d.f. = 13.

Florida Entomologist 74(3)

males), suggesting the possibility of trapping out the papaya fruit fly with a suitable
trap density. The papaya fruit fly is easily spotted on papaya because it is a large,
colorful tephritid which remains principally on the large exposed papaya fruit and not
on foliage. Most papaya fruit flies in a papaya grove can be spotted and counted in a
row by row walkthrough.
Previously, we recommended (Landolt 1984b) that information on diel activity pat-
terns of the papaya fruit fly be used to maximize efficacy of control methods used, such
as pesticide applications directed at this insect. The present results show the timing of
such treatments in Guatemala and Costa Rica should be much earlier in the day (late
morning) than that recommended for south Florida (late afternoon).


Technical assistance was provided by Jorge Lopez, Alvero Gonzalez and Francesco
Marin. We thank C. O. Calkins for logistical assistance.
Mention of a proprietary product does not constitute an endorsement by the USDA.


BATES, H. W. 1862. Contributions to the insect fauna of Amazon valley. Trans. Linn.
Soc. (London). 23: 495-566.
CHUMAN, T., P. J. LANDOLT, R. R. HEATH, AND J. H. TUMLINSON. 1987. Isolation,
identification, and synthesis of a male-produced sex peromone of the papaya fruit
fly, Toxotrypana curvicauda (Gerstaecker) (Diptera: Tephritidae). J. Chem. Ecol.
13: 1979-1992.
KNAB, AND W. W. YOTHERS. 1914. Papaya fruit fly. J. Agric. Res. 2: 447-455.
LANDOLT, P. J. 1984a. Reproductive maturation and pre-mating period of the papaya
fruit fly, Toxotrypana curvicauda (Diptera: Tephritidae). Florida Entomol. 67:
LANDOLT, P. J. 1984b. Behavior of the papaya fruit fly, Toxotrypana curvicauda
Gerstaecker (Diptera: Tephritidae) in relation to its host plant Carica papaya L.
Folia Entomol. Mexicana 61: 215-224.
LANDOLT, P. J., AND R. R. HEATH. 1988. Effects of age, mating, and time of day
on behavioral responses of female papaya fruit fly, Toxotrypana curvicauda
Gerstaecker (Diptera: Tephritidae) to synthetic sex pheromone. Environ. En-
tomol. 17: 47-51.
LANDOLT, P. J., R. R. HEATH, AND J. R. KING. 1985. Behavioral responses of female
papaya fruit flies, Toxotrypana curvicauda (Diptera: Tephritidae), to male-pro-
duced sex pheromone. Ann. Entomol. Soc. Am. 78: 751-755.
LANDOLT, P. J., AND R. R. HEATH. 1990. Effects of pheromone release rate and
time of day on catches of male and female papaya fruit flies (Diptera: Tephritidae)
on fruit model traps baited with pheromone. J. Econ. Entomol. 83: 2040-2043.
1988. Sex pheromone-based trapping system for papaya fruit fly (Diptera: Tep-
hritidae). J. Econ. Entomol. 81: 1163-1169.
LANDOLT, P. J., AND J. HENDRICHS. 1983. Reproductive behavior of the papaya fruit
fly, Toxotrypana curvicauda (Gerstaecker) (Diptera: Tephritidae). Ann. Entomol.
Soc. Am. 76: 413-417.
STEEL, R.G.D. AND T. H. TORRIE. 1960. Principles and Procedures of Statistics.
McGraw-Hill. Book Co., NY. 481 pp.
WOLCOTT, G. N. 1933. An Economic Entomology of the West Indies. Clay and Sons
Ltd., Bungay, Suffolk.

September, 1991

Price and Kring: D. morator in Caladium


University of Florida, IFAS
Gulf Coast Research and Education Center
5007 60th Street East
Bradenton, Florida 34203


Adult Dyscinetus morator (F.) fed on developing caladium (Caladium x hortulanum
Birdsey) leaf buds and petioles in field and greenhouse observations. Adults were found
in caladium fields but immatures were absent. Periods of adult flight occurred primarily
in mid-January through mid-April, before the crop was planted, and in June and July
following planting. There were no significant differences in the amount of leaf tissue
consumed between that of the known wild host, water hyacinth (Eichhornia crassipes
(Martius), and that of caladium, poinsettia (Euphorbia pulcherrima L.), lettuce (Lactuca
sativa L.), garden pea (Pisum sativum L.), yellow squash (Cucurbita pepo L.) or tomato
(Lycopersicon esculentum L.) indicating that these crops potentially could be damaged.
Carbaryl, chlorpyrifos, diazinon and oxamyl were the most effective of 10 insecticides
applied as soil surface sprays for control of D. morator, but none provided greater than
72.5% reduction after 24 h. Alternative insecticides or methods of D. morator manage-
ment may be needed to protect caladium crops.


Se observe en el campo y en la casa de mallas adults de Dyscinetus morator (F.)
alimentandose de yemas y peciolos de caladium (Caladium x hortulanum Birdsey).
Cuando se encontraron adults, los estados inmaduros estaban ausentes. Los periods
de vuelo de adults ocurieron desde mediados de Enero hasta mediados de Abril, antes
de la siembra del cultivo, y en Junio y Julio, despues de la siembra. No hubo diferencias
significativas en la cantidad de tejido foliar consumido en el hospedero silvestre, jacinto
acuatico (Eichornia crassipes Martius), y en el consume en caladium, poinsetia (Euphor-
bia pulcherrima L.), lechuga (Lactuca sativa L.), alverja (Pisum sativum L.), calabacitas
amarillas (Cucurbita pepo L.) o tomate (Lycopersicon esculentum L.), lo cual indica el
potential de dafto en estos cultivos. De 10 insecticides aplicados en forma de aspercion
a el suelo, carbaryl, chlorpyriphos, diazinon y oxayml fueron los mas efectivos para el
control de D. morator, pero ninguno brindo un control de mas del 72.5% 24 horas despues
de la aplicacion. Se necesitan otros insecticides o metodos alternos para un manejo de
protection eficaz de los cultivos de caladium.

Dyscinetus morator (F.), sometimes called the "rice beetle," is distributed throughout
the eastern U.S. (Woodruff 1970). Even though this was the most abundant scarab
caught in blacklight traps in southern Florida (Foster et al. 1986), few aspects of its life
history are understood. Adults and larvae have been found in compost and near pigpens
(Phillips & Fox 1924). This insect feeds on cultivated crops such as rice (Oryza sativa
L.) (Phillips & Fox 1924), corn (Zea mays L.) (Anonymous 1980), pangola grass (Digitaria



Florida Entomologist 74(3)

September, 1991

decumbens Stent) (Anonymous 1956), cranberry (Vaccinium macrocarpon Aiton) (Scam-
mell 1917), radish (Raphanus sativus L.), lettuce (Lactuca sativa L.) and carrot (Daucus
carota L.) (Foster et al. 1986) in addition to a wild host, water hyacinth (Eichhornia
crassipes (Martius) (Buckingham & Bennett 1989). The senior author has collected full
grown larvae that were damaging roots of ornamental juniper (Juniperus sp.) potted
in a pine bark medium. Adults may be abundant in Florida sugarcane (Saccharum
officinarum L.) fields but larvae are not known to damage that crop (Gordon & Anderson
Caladium (Caladium x hortulanum Birdsey) growers report that D. morator adults
invade their fields (Woodruff 1970) and reduce yields. Caladium tubers are produced
for landscape planting or for pot production. In 1986, 64 million caladium tubers ($8.4
million wholesale), representing most of the world production, were grown in Florida
(Waters et al. 1987). Little is known about caladium losses to D. morator or the means
to alleviate losses. This study was performed to determine periods of D. morator flight
activity as it relates to caladium production, to determine D. morator damage to caladium
and additional plant species, and to evaluate insecticides as a means of managing this
pest in caladium.


Damage to Caladium

In the spring and summer 1985, several Highlands Co. caladium fields were naturally
infested by adult D. morator. One field was inspected periodically soon after tuber
pieces were planted in May and until tubers matured in October. At each observation,
50 plants were dug to expose leaves, petioles, tubers and roots. Holes in plant parts
and missing tissues, indicating feeding damage, were noted. The plants and the soil
surrounding them were inspected for all lifestages of the beetle.
A greenhouse experiment also was performed to evaluate D. morator damage to
sprouting tubers. Single adult beetles were placed into ten 15 cm diameter pots containing
organic muck soil, taken from a caladium field, and a sprouting jumbo 'Candidum'
caladium tuber. Ten similarly prepared pots were left without beetles and all pots were
covered with screens to prevent beetles from leaving or entering. Tubers were inspected
for damage after 10 d.

Seasonal Flight Activity

A 15 watt AC blacklight trap was erected in an open field at the Gulf Coast Research
and Education Center, Bradenton and was operated continually for 3 yr (1 January 1986
31 December 1988) to determine the period of adult movement in relation to caladium
phenology. D. morator adults were removed and counted at 1 d to 2 wk intervals,
depending on numbers of beetles caught, and were summed over each 2 wk period.

Food Plant Acceptance

Leaves of 16 ornamental, food and weed plants, including the known wild host, water
hyacinth, were tested for acceptance as food by adult D. morator. Experimental units
consisted of moistened filter paper placed in bottoms of 15 cm diameter petri dishes
along with a 791 mm2 leaf disk and one adult beetle trapped the previous night by
blacklight trap. Remaining leaf area was measured and the percent consumed was
calculated after 24 h. The experiment was replicated four times.

Price and Kring: D. morator in Caladium 417

Insecticide Effects

No insecticides currently are registered for D. morator control on caladium. Thus
labels were searched to identify insecticides with possible toxicity to the pest and reg-
istrations allowing use on caladium. Ten active ingredients were found that were regis-
tered for control of one or more coleopterans and also were labelled for use on field-grown
caladium, flowers or ornamentals. One commercial product from each of the 10 active
ingredients was tested as a soil surface spray. Two additional experiments were per-
formed in a similar manner using five of the insecticides found earlier to be most effective.
All experiments were performed as follows: Plastic 473 ml delicatessen containers
were provided with 400 ml of moist muck soil taken from a caladium field. Each insecticide
was sprayed onto the soil surface of four containers at the highest labelled concentration
for caladium at approximately 947 liters of preparation per ha. Sprayed surfaces were
permitted to dry for 2 h, after which 10 adult beetles, taken the preceding night at a
blacklight trap, were placed into each cup. Lids with their centers removed to retain
the beetles but to permit ventilation were placed onto the containers. Containers were
placed into a ventilated insect rearing room maintained at 27 20C and 12 h light and
12 h dark. After 24 h, living and dead insects were recorded.


Percentage data were transformed to the arcsine to stabilize error variance and were
analyzed using an analysis of variance. Least significant difference (LSD) among means
was calculated or means were separated by Duncan's new multiple range test. The
general linear model procedure was used in these analyses (SAS Institute 1985, 113-137),
and data are reported in the original scale.


Damage to Caladium

Adults were found periodically in caladium fields but eggs, larvae and pupae were
not, indicating that beetles entered fields as adults but reproduced elsewhere. Early
season damage consisted of adults consuming leaf buds and petioles. This can result in
reduced crop canopy and increased light penetration to facilitate weed growth. Current
techniques of caladium production emphasize rapid canopy development to prevent ex-
cessive weed growth and do not provide space for mechanical or chemical weed control
in absence of an adequate canopy.
In July and August, after tubers enlarged, adults chewed holes (ca. 0.5-1.0 cm
diameter by ca. 0.5 cm deep) into tubers. Damaged tubers often rotted, further reducing
the crop canopy.
Adults confined to sprouting tubers in the greenhouse consumed emerging leaf tissue.
Tubers in pots with beetles for 1 wk produced 1.3 leaves within 10 d of the beetles'
introduction, but without beetles, produced a significantly (P= 0.05) greater 4.8 leaves
during the same period.

Seasonal Flight Activity

During 1986 and 1987, moderate numbers of beetles (up to 800/trap/2 wk) were
caught during the preplanting period, mid-January through mid-April (Fig. 1). Few
beetles were caught in any year from mid-April through May, when caladium crops are

Florida Entomologist 74(3)

..... 1988


September, 1991



Fig. 1. Numbers of adult D. morator beetles captured by a blacklight trap in Braden-
ton, Florida during 2 wk periods 1986-1988.






W 1800

Price and Kring: D. morator in Caladium


Percent leaf area
consumed SEM


Eichhornia crassipes (Martius)
Water hyacinth
Solanum nigrum L.
Hydrocotyle umbellata L.
Water pennywort
Trifolium repens L.
White clover

Ornamental Plants
Caladium x hortulanum Birdsey
Caladium 'Frieda Hempel'
'Red Frill'
Euphorbia pulcherrima L.
Poinsettia (green leaf)
(red bract)
Gladiolus x hortulanus L. H. Bailey
Gladiolus (petal)

Food Plants
Lactuca sativa L.
Brassica oleraceae L.
Capsicum annuum L.
Bell Pepper
Pisum sativum L.
Garden pea
Cucurbita pepo L.
Yellow squash
Lycopersicon esculentum L.
Musa acuminata Colla
Musa xparadisiaca L.
Saccharum officinarum L.

50.6 28.5

4.5 2.2

0.0 0.0

0.0 0.0

33.5 11.1
45.0 9.7

22.7 8.1
30.7 1.3

2.5 1.3

74.2 + 14.7

8.0 0.7

10.8 5.4

28.4 19.3

31.6 1.7

35.0 11.5

12.5 8.0

0.0 0.0

0.6 0.6

(LSD (P = 0.05) = 30.1)


420 Florida Entomologist 74(3) September, 1991

planted. The highest numbers of beetles (up to 4,600/trap/2 wk) were caught each year
during June and July, when the young crop is particularly susceptible to defoliation and
weed competition. Insecticides could be useful at this time to protect the developing
crop. Few beetles were caught after July in any year.

Food Plant Acceptance

No leaf tissue was consumed in significantly greater quantities than from the known
wild host, water hyacinth (Table 1). Significantly less leaf tissue was consumed from
nightshade (Solanum nigrum L.), water pennywort (Dyrocotyle umbellata L.), white
clover (Trifolium repens L.), gladiolus (Gladiolus x hortulanus L. H. Bailey) petal,
cabbage (Brassica oleraceae L.), banana (Musa acuminata Colla), plantain (Musa x
paradisiaca L.), or sugarcane than from water hyacinth. There was no significant differ-
ence in the amount of leaf tissue consumed between water hyacinth and caladium,
poinsettia (Euphorbia pulcherrima L.), iceberg lettuce, garden pea (Pisum sativum
L.), yellow squash (Cucurbita pepo L.) or tomato (Lycopersicon esculentum L.), indicat-
ing that these crops potentially could be damaged by D. morator.

Insecticide Effects

Effects of soil-applied, residual sprays to control D. morator are presented in Table
2. In the first experiment, designed to identify possibly useful insecticides, the most
effective products were carbaryl, chlorpyrifos, diazinon, and oxamyl. However, no insec-
ticide killed more than 72.5% of the adult D. morator. Insecticides producing mortality
not significantly different from the untreated check included azinphos-methyl, bendiocarb,
endosulfan, fenvalerate, malathion, and pyrethrum with piperonyl butoxide. In the
second experiment, the most effective insecticides were carbaryl, chlorpyrifos and
oxamyl, with no insecticide killing more than 65%. Similar results occurred in the third
experiment except that none of the insecticides was more effective than diazinon.


Percent Mortality
Experiment No.

Insecticide Amt AI/liter 1 2 3

Untreated check 16.7 ef 7.5 c 2.5 b
Pyrethrum and piperonyl butoxide 0.94 ml 12.4 ef
Fenvalerate 0.82ml 12.5 f
Malathion 4.80 g 25.0 def
Endosulfan 1.19g 25.0 def
Azinphos-methyl 2.49 ml 28.3 def
Bendiocarb 1.48g 37.5 cde
Carbaryl 2.40g 45.7 bed 62.5 a 62.5 a
Diazinon 1.25ml 55.0abc 30.0b 50.0a
Chlorpyrifos 1.19g 72.5a 37.5ab 47.5a
Oxamyl 4.97ml 72.5ab 65.0a 66.7a

'Values within a column followed by the same letter are not significantly different (P = 0.05), Duncan's (1955)
multiple range test).

Price and Kring: D. morator in Caladium 421

Oxamyl resulted in the highest level of mortality achieved in each of the three
experiments; but even oxamyl provided no greater than 72.5% mortality in 24 h. This
low level of performance possibly could lead to considerable losses from D. morator.
These data indicate that more effective insecticides need to be identified or alternative
means of D. morator management should be sought for the June and July production


The authors are grateful for the technical support provided by Emily Vasquez and
Preston Young. This manuscript was approved for publication as Florida Agricultural
Experiment Station Journal Series No. R-01113.


ANONYMOUS. 1956. U.S. Dept. Agric. Cooperative Econ. Insect Rep. 3: 725.
ANONYMOUS. 1980. U.S. Dept. Agric. Cooperative Econ. Insect Rep. 5:66.
BUCKINGHAM, G. R., AND C. BENNETT. 1989. Dyscinetus morator (Fab.) (Coleop-
tera:Scarabaeidae) adults attack water hyacinth, Eichhornia crassipes (Pon-
tederiaceae). Coleopterists Bull. 43(1): 27-33.
FOSTER, R. E., J. P. SMITH, R. H. CHERRY, AND D. G. HALL. 1986. Dyscinetus
morator (Coleoptera:Scarabaeidae) as a pest of carrots and radishes in Florida.
Florida Entomol. 69(2): 431-2.
GORDON, R. D., AND D. M. ANDERSON. 1981. The species of Scarabaeidae (Coleop-
tera) associated with sugarcane in South Florida. Florida Entomol. 64(1): 119-38.
PHILLIPS, W. J., AND H. Fox. 1924. The rough-headed corn stalk-beetle. U.S. Dept.
Agric. Bull. 1267: 1-3.
SAS INSTITUTE, INC. 1985. SAS Institute, Cary, N.C. 956 pp.
SCAMMELL, H. B. 1917. Cranberry insect problems and suggestions for solving them.
U.S. Dept. Agric. Farmers' Bull. 860: 1-42.
WATERS, W. E., G. J. WILFRET, AND B. K. HARBAUGH. 1987. The flower industry:
Description and changes, future potential, and research and educational needs.
Univ. Florida Gulf Coast Res. and Education Cent. Bradenton Res. Rep.
BRA1987-24. 6pp.
WOODRUFF, R. E. 1970. The "rice beetle," Dyscinetus morator (Fab.) (Coleoptera:
Scarabaeidae. Florida Dept. Agric. and Consumer Services Division of Plant
Industries Entomol. Circ. No. 103. 2pp.

422 Florida Entomologist 74(3) September, 1991


Institute de Investigaciones
Universidad del Valle de Guatemala, Apartado 82
Guatemala, GUATEMALA


Four new montane species of Petrejoides are described, 2 from Mexico, 1 from
Guatemala and 1 from Panama. Larvae are described for 2 of the species. A revised
key is given including all species of the genus.


Se described cuatro nuevas species montanas de Petrejoides. Dos de Mexico, una
de Guatemala y una de Panama. Se described larvas de 2 de las species. Se present
una clave modificada, la cual incluye todas las species del genero.

Most of the 14 described species of Petrejoides are found inhabiting rotting logs in
mesophytic or cloud forests above 1000 m altitude (the only exceptions being P. subrec-
ticornis (Kuwert) in wet lowland forests, and P. orizabae Kuwert, whose range extends
down to 600 m) (Castillo and Reyes-Castillo 1984, Reyes-Castillo and Schuster 1983,
Schuster 1988, 1989). With the possible exception of P. subrecticornis, each species is
relatively restricted in its geographical range (see maps in Castillo & Reyes-Castillo
1984). Mesoamerica (Chiapas through Panama) is relatively unexplored with respect to
Passalidae, especially in the montane areas. Only 5 species of Petrejoides are known
from this region compared with 9 to the north of the Isthumus of Tehuantepec. Here I
describe 1 new species to the north of the isthmus, 3 new mesoamerican species: 1 each
from Chiapas, Guatemala and Panama, (Fig. 1) and present a revised key to all described
Diagnostic characters of the genus include: razor-thin anterior border of clypeus; a
strong fronto-clypeal suture (sometimes erased in the middle); frons twice or less the
length of clypeus (measured to base of median frontal structure); usually a high dorsal
ridge on tibia II; and body length <33 mm. I find Coniger Zang to be quite similar,
differing in size (>30 mm) and by having a frons length 4 times that of the clypeus.
Pseudoarrox Reyes-Castillo is also similar; however, it lacks a fronto-clypeal suture.
For explanation of morphological terminology see Reyes-Castillo (1970).

Petrejoides panamae Schuster NEW SPECIES
Figures 2d,e,f
DESCRIPTION. Head: anterior border of labrum concave, anterior angles rounded.
Clypeus inclined, short (anterior-posterior), elongate-rectangular or pentagonal shaped;
anterior border linear, with or without indentations, anterior angles rounded and directed
downwards; smooth and shiny throughout; fronto-clypeal suture curved, absent in middle
1/6th. External tubercles small, rounded, almost absent.
Frontal area long, without frontal ridges or inner tubercles. Frontal fossae glabrous.
Median frontal structure of "striatopunctatus" type, center horn elongate with pointed

Schuster: New Passalidae

- --- z

Fig. 1. Distribution of 4 new species of Petrejoides: open circle-P. michoacanae,
dark circle-P. chiapasae, star-P. pokomchii, square-P. panamae.

Florida Entomologist 74(3)

September, 1991

(I- -


.!' ---

i- V!

' < J


Fig. 2. New species of Petrejoides. a-c. P. pokomchii a. head, b. aedeagus, dorsal
view, c. aedeagus, ventral view; d-f. P. panamae d. head, e. aedeagus, dorsal view, f.
aedeagus, ventral view; g. P. chiapasae-head; h. P. michoacanae-head.

INDEX WORDS: Petrejoides Passalidae Mesoamerica montane

apex reaching clypeus or beyond, without median longitudinal groove, without lateral
ridges. Occipital groove well marked, terminating at supraorbital ridge.
Anterior 1/2 of supraorbital ridge bituberculate, anterior tubercule more pronounced;
posterior 1/2 bifurcate. Anterior cephalic angle rounded or slightly protruding. Canthus
narrow or slightly swollen distally, not reaching lateral eye margin. Eyes small, dorsal
width of an eye 1/10 head width (though 1/10 is the same measure as P. guatemalae,
the eyes do not appear to be as reduced due to the small canthus).


Schuster: New Passalidae

Ligula between insertions of labial palps wide to narrow, flat. Lateral lobes of mentum
with anterior external border rounded, whole surface pubescent, lateral border straight.
Medial basal mentum bare, anterior border convex to slightly biconvex. Hypostomal
process with slight lateral depression, separated from mentum by distance less than
process width. Infraocular ridge indistinct, pubescent.
Superior and inferior apical mandibular teeth approximately same length. Dorsal
tooth occupies 1/2 length of mandible. Internal tooth of left mandible bifid.
Thorax: lateral fossa of pronotum with 2-4 punctations, 2-6 punctations in lateral
pronotum posterior to fossa. Pronotum with marginal groove narrow, with punctations;
anterior angles rounded. Prosternum rhomboidal with post-apendix truncated. Lateral
margins of mesosternum shiny, marginal grooves well marked, pubescent and punctate.
Center of mesosternal shield without punctations.
Metasternum with 25-35 punctations delimiting each latero-posterior side of disc,
anterior angles pubescent, lateral fossa wide and very pubescent.
Anterior elytral profile convex, elytral striations marked uniformly with small, light
punctations, somewhat heavier in lateral striations; junction of striations 1 and 10 with
extra punctations giving appearance of 2 rows in places.
Wings: as in P. recticornis (Burmeister), not reduced (see Fig. 5 in Castillo and
Reyes-Castillo (1984)).
Legs: femur I with antero-ventral groove distinctly marked for 4/5 of anterior border;
posterior 1/2 of ventral face pubescent; dorsal ridge extends total length of tibia II, with
2 rows of setae same length as that of lateral border.
Abdomen: marginal groove of last sternite complete or almost complete. Form and
coloring of aedeagus given in Figs. 2 and 3.
Dimensions (mm): total length, mandibles to tip of elytra 24.5-26, x = 25.5, males
24.5-25, female 26; elytral length 13.8-15.0, i = 14.5; pronotal length 5.7-6.0, i = 5.9;
pronotal width 7.8-8.0; i = 7.9; humeral width 7.6-7.8, i = 7.7; head width 5.4-5.6, x
= 5.5; aedeagal length 2.4.
MATERIAL EXAMINED. Four whole specimens, including 2 general males, 1 general
female and 1 black adult of unknown sex, plus pieces of at least 3 other individuals, 5
third instar larvae partially decomposed (head widths: 3.8, 4.2, 4.2, 4.2, 4.5 mm).
TYPE MATERIAL. Holotype: general male-PANAMA, Chiriqui Dept., Respingo 7-
XII-1985. Altitude 2450 m. J. C. Schuster collector. Wet Quercus Alnus forest. In
oak log 90 cm dia. Log field number PAN-2e. Will be deposited in the Florida State
Collection of Arthropods in Gainesville (FSCA).
Paratypes: 3, all from same log as holotype. Types will be divided between the
collections of the FSCA, Pedro Reyes-Castillo, Instituto de Ecologia, Mexico; and the
author's personal collection.
ETYMOLOGY. This species is named after the country where discovered.
LARVA. The larva has the same basic setal pattern of other species of Petrejoides
(Schuster & Reyes-Castillo 1981, Schuster 1988). It differs from other Petrejoides,
except P. guatemalae, in having 14 anal ring setae instead of 12. It is similar to P.
olmecae and P. recticornis in having 5 or more internal coxal setae. P. panamae is a
relatively large species, slightly smaller than P. reyesi (Schuster 1988). Instar III head
width 3.8-4.5mm.
DISTRIBUTION AND ECOLOGY. Known only from Respingo, Chiriqui Dept., Panama
at 2450 m altitude in Quercus-Alnus wet forest (Fig. 1). Also known from this forest
are Publius sp. nov. and Passalus (Pertinax) sp. The Publius (= Publius sp. A of
Schuster & Reyes-Castillo 1981) is a large high altitude species known from Costa Rica
and Panama. The Passalus may be a new species.
AFFINITIES. P. panamae seems to be in the "orizabae" group; however, it lacks,
or has very weak, infraocular ridges. It is similar to the "guatemalae-reyesi-salvadorae"

Florida Entomologist 74(3)

unit (Schuster 1989) of northern Central America in size and in lacking inner tubercles.
It is distinct from all other Petrejoides in lacking lateral ridges on the median frontal

Petrejoides pokomchii Schuster NEW SPECIES
Figure 2a,b,c

DESCRIPTION. Head: anterior border of labrum concave, anterior angles rounded.
Clypeus inclined, long (anterior-posterior), almost rectangular shaped; anterior border
linear with median indentation, anterior angles rounded and directed downward; post-
erior 1/2-2/3 rugose. External tubercles distinct, rounded.
Frontal area short, without inner tubercles. Frontal ridges absent or lightly present
as short lines parallel to lateral ridges of median frontal structure. Frontal fossae glabr-
ous. Median frontal structure of "falsus" type, center horn short with apex not free,
with median longitudinal groove in posterior 1/3, lateral ridges at 90 degrees or slightly
less, straight and prominent, terminal tubercles rounded and set slightly forward, pro-
nounced. Occipital groove well marked except extreme posterior portion, terminating
in frontal fossae.
Anterior 1/2 of supraorbital ridge bituberculate, anterior tubercles more pronounced;
posterior 1/2 bifurcate. Anterior cephalic angle rounded or slightly protruding. Canthus
swollen distally, apex rounded, protruding slightly beyond lateral eye margin. Eyes
small, dorsal width of an eye less than 1/12 head width, ratio eye width to length = 0.7.
Ligula between insertions of labial palps wide, flat, pubescent or bare anteriorly.
Lateral lobes of mentum with anterolateral border rounded, whole surface pubescent-
punctate, lateral border straight. Medial basal mentum with abundant setae and punc-
tations, anterior border convex. Hypostomal process narrow, without lateral depression.
Infraocular ridge absent or slightly present, area very punctate-pubescent. Superior
and inferior apical mandibular teeth approximately same length. Dorsal mandibular
tooth occupies >1/2 mandible length. Internal tooth of left mandible bifid.
Thorax: lateral fossa of pronotum with at most 1-2 punctations, fossa may be absent;
at most 1-2 other punctations on pronotum outside narrow marginal groove. Anterior
angles of pronotum rounded. Prosternum rhomboidal with posterior apex truncated.
Lateral margin of mesosternum opaque, marginal grooves present only slightly.
Center of mesosternal shield with at most 1 or 2 punctations.
Metasternum totally glabrous, with 2-6 punctations delimiting each latero-posterior
side of disc; marginal fossa very narrow entire length, with a few short hairs.
Anterior elytral profile convex; elytral striations marked uniformly with small, light
punctations, slightly heavier in lateral striations; junction of striations 1 and 10 without
extra punctations. Rounded elytral form suggests reduced wings.
Legs: femur I with antero-ventral groove distinctly marked for 4/5 of anterior border;
posterior 1/2 of ventral face pubescent; dorsal ridge extends most of length of tibia II,
with 2 rows of setae same length as that of lateral border.
Abdomen: marginal groove occupies median 3/5 of last sternite. Form and patterning
of aedeagus given in Figure 2b,c.
Dimensions (mm): total length, mandible to elytral tip 26-29, x = 27; elytral length
13.9-16.1, i = 14.8; pronotal length 6.2-7.0, i = 6.5; pronotal width 8.2-9.0, i = 8.5;
humeral width 7.7-8.6, i = 8.3; head width 5.4-6.0, i = 5.6; edeagal length 2.3-2.6, i
= 2.4 (n = 4 for all measurements, except total body length and aedeagal length n = 3).
MATERIAL EXAMINED. Three whole specimens and 1 specimen found in pieces, and
1 instar II larva (head width 3.1 mm).
TYPE MATERIAL. Holotype: male-GUATEMALA, El Progreso Dept., Cerro Pinal6n
above Los Albores. 1-VII-1989. Altitude 2710 m. J. C. Schuster collector. In log in oak
cloud forest. Field number WGe-1. It will be deposited in the FSCA.


September, 1991

Schuster: New Passalidae

Paratypes: 2 males with same data as holotype. Field numbers WGe-2,3. Another
beetle with same data as holotype except at 2700 m in oak cloud forest with more pine
interspersed than in area of holotype. A Chondrocephalus debilis (Bates) also in this
log, both beetles dead and in pieces. Field number WGj-1.
ETYMOLOGY. This species is named for the Pokomchi speaking Mayans which inhabit
the region of the Sierra de las Minas where this species was found.
LARVA. Has same basic setal patterns of other species of Petrejoides (Schuster &
Reyes-Castillo 1981, Schuster 1988). It is similar to the larvae of P. reyesi and P.
guatemalae in having 2 internal coxal setae and differs from all other instar II Petrijoides
known in lacking raster setae.
DISTRIBUTION AND ECOLOGY (Fig. 1). Known only from Cerro Pinal6n in the Sierra
de las Minas of Guatemala at approximately 2700 m altitude in virgin oak cloud forest.
This forest extends to the top of Pinal6n and down to approximately 2450 m where more
tree species appear and the passalid fauna changes as well. In the genus Ogyges above
this limit I found only 0. furcillatus Schuster & Reyes-Castillo; below it I found only
0. laevior (Kaup). Other passalids in the forest with P. pokomchii were Chondrocephalus
debilis, C. granulifrons (Bates) and a species of Vindex. P. pokomchii was the rarest
species, 0. furcillatus and C. debilis the commonest.
AFFINITIES. P. pokomchii is a member of the "orizabae" group (Castillo & Reyes-
Castillo 1984). The partially rugose clypeus, and body size indicate a close relationship
to the "guatemalae-reyesi-salvadorae" group, the other component high altitude Pet-
rejoides species of northern Central America. All of these species are geographically
quite isolated from P. pokomchii, the closest being P. guatemalae in the Cuchumatan
Mountains and the Maria Tecin ridge of Guatemala. I suspect other new species of this
group will be found on other, unexplored, high mountains in northern Mesoamerica.
This species also suggests a close relationship of Petrejoides to Chondrocephalus in
its similarity to C. granulifrons in the distribution of rugosity on the clypeus and the
form of the median frontal structure.

Petrejoides chiapasae Schuster NEW SPECIES
Figure 2g
DESCRIPTION. Head: anterior border of labrum concave, anterior angles rounded.
Clypeus inclined, short (anterior-posterior), rectangular shaped; anterior border linear
without median indentation, anterior angles slightly rounded and directed downward;
shiny throughout. External tubercles indistinct, rounded.
Frontal area long, frontal ridges and inner tubercles distinct but only slightly ele-
vated. Frontal fossae pubescent. Median front structure of "falsus" type (Reyes-Castillo
1970), center horn short with apex free and median longitudinal groove in posterior 1/3.
Lateral ridges indistinct, marked by grooves running slightly posteriorly to terminal
tubercles which are rounded and only slightly pronounced. Occipital groove well marked,
terminating in frontal fossae.
Anterior 1/2 of supraorbital ridge bituberculate, anterior tubercles more pronounced
but smaller; posterior 1/2 bifurcate. Anterior cephalic angles rounded or slightly protrud-
ing. Canthus swollen distally, apex rounded, not protruding beyond eye margin. Eyes
normal size, approximately 1/10 head width, ratio eye width to length = 0.6.
Ligula between insertions of labial palps narrow, pubescent. Lateral lobes of mentum
with anterior external borders rounded, whole surface pubescent, lateral border straight.
Medial basal mentum bare, posterior 1/2 opaque, anterior border slightly convex. Hypos-
tomal process with slight lateral depression, separate from mentum by distance less
than process width. Infraocular ridge indistinct, punctate and pubescent.
Superior and inferior apical manidbular teeth approximately same length. Dorsal
mandibular tooth occupies approximately 1/2 mandible length. Internal tooth of left
mandible bifid.


428 Florida Entomologist 74(3) September, 1991

Thorax: lateral fossa and surrounding area of pronotum with >30 punctations. Pro-
notum with marginal groove narrow with indistinct punctations; anterior angles rounded.
Prosternum rhomboidal with posterior apex truncated.
Lateral margin of mesosternum opaque, marginal grooves shallow, pubescent slightly
punctate. Center of mesosternal shield with longitudinal depression, lightly punctate.
Metasternum with >40 punctations delimiting each latero-posterior side of disc,
anterior angles pubescent, lateral fossae wide and very pubescent.
Anterior elytral profile slightly convex. Elytral striations marked uniformly with
small, round, light punctations, somewhat heavier in lateral striations; junction of stri-
ations 1 and 10 with extra punctations giving appearance of 2 rows in places.
Legs: femur I with antero-ventral groove distinct for only 1/4 of anterior border;
posterior 1/4 of vertical face pubescent; dorsal ridge extends total length of tibia II,
with 2 rows of setae same length as that of lateral border.
Abdomen: marginal groove of last sternite complete.
Dimensions (mm): total length, mandibles to tip of elytra 25.5, elytral length 14.5,
pronotal length 5.8, pronotal width 7.8, humeral width 7.7, head width 5.8.
MATERIAL EXAMINED. One female specimen.
TYPE MATERIAL. Holotype: female-MEXICO, Chiapas, 2 miles north of Ejido Leiva
Velasquez (approximately 29 miles northeast of Las Margaritas). Altitude 2060 m. J.
C. Schuster collector. Wet pine forest with hardwood understory; grades into cloud
forest with Podocarpus, Quercus and tree ferns. Field number TS-1. Hardwood log. It
will be deposited in the FSCA.
ETYMOLOGY. P. chiapasae is named for the state of Chiapas where it was found.
DISTRIBUTION AND ECOLOGY. The holotype was the only specimen of Petrejoides
in the tunnel; however, at end of the tunnel was a specimen of Chondrocephalus sp. In
over 20 years of collecting passalids, this is only the second time I have found 2 species
in the same or connecting tunnel systems. Also, apart, in the same log were an Ogyges
laevior and pieces of Odontotaenius sp. In the same forest I found Spurius dichotomus
Zang and Oileus sargi (Kaup).
I suspect this species is endemic to this small, isolated cloud forest. To the north,
the forest ends at a broad, deep river valley with the lower Lancand6n ridges beyond.
To the southwest are many mountain ridges and valleys, mostly with pine or relatively
dry pine-oak forest. Other passalid species associated with this specimen are usually
restricted to cloud forest or mesophytic habitats above 1000 m. Other isolated montane
areas of Chiapas have also shown endemism in Passalidae (Reyes-Castillo, da Fonseca
& Castillo, 1987; Reyes-Castillo & Castillo 1985). Many of these isolated forests in
Chiapas are relatively unexplored, but are disappearing fast to ax and saw.
AFFINITIES. This species belongs to the "orizabae" group of Castillo & Reyes-Castillo
(1984). It seems most related to P. orizabae Kuwert and P. silvaticus Castillo & Reyes-
Castillo, found in the southern and northern Sierra Madre Oriental, respectively. It
differs from them in having the internal left mandible tooth bifurcate instead of trifurcate
and by possessing many punctations delineating the metasternal disc. It is similar in
size to P. silvaticus.

Petrejoides michoacanae Schuster NEW SPECIES
Figure 2h

DESCRIPTION. Head: anterior border of labrum highly concave, more so than any other
species of Petrejoides; concavity not as wide as P. salvadorae and without its bare
depression posteriorly; anterior angles rounded. Clypeus inclined, short (anterior-
psoterior), rectangular; anterior border linear with a strong median indentation, anterior
angles sharp and directed downward; smooth and brilliant. External tubercles distinct
and rounded.

Schuster: New Passalidae

Frontal area long, without inner tubercles. Frontal ridges poorly marked. Frontal
fossae with numerous long setae. Median frontal structure of "falsus" type; center horn
short, reaching less than 1/2 the distance of clypeus, apex free, without median longitud-
inal groove posteriorly; lateral ridge at right angles to median horn, rounded, terminal
tubercles absent. Occipital groove well marked, terminating in supraorbital ridges.
Anterior 1/2 of supraorbital ridge bituberculate; posterior 1/2 not bifurcate. Anterior
cephalic angle rounded. Canthus not swollen distally, apex rounded, not reaching lateral
eye margin. Dorsal width of an eye 1/9 head width.
Ligula between insertions of labial palps somewhat narrow, flat, pubescent. Lateral
lobes of mentum with anterior, external border rounded, whole surface punctate and
pubescent, lateral border straight. Medial basal mentum smooth, glabrous throughout,
anterior border slightly convex. Hypostomal process narrow, without lateral depression.
Infraocular ridge almost absent, punctate and pubescent.
Mandibles with 3 apical teeth. Dorsal mandibular tooth occupies < 1/2 mandible length.
Internal tooth of left mandible trifid.
Thorax: lateral fossa of pronotum with approximately 10 very light punctations and
1 or 2 setae, rest of pronotum smooth. Anterior angles of pronotum rounded. Prosternum
rhomboidal with posterior apex truncate.
Mesosternum mostly brilliant throughout, with many punctations, glabrous. Lateral
depressions obsolete.
Metasternum pubescent in anterior corners, without punctations delimiting latero-
posterior sides of disc; marginal fossae narrow, pubescent.
Anterior elytral profile convex; elytral striations marked with small, light puncta-
tions, slightly heavier in lateral striations; junction of striations 1 and 10 with few extra
Legs: femur I with antero-ventral groove distinctly marked for 4/5 of anterior border;
posterior 1/4 of ventral face pubescent; dorsal ridge extends most of length of tibia II,
though height not very pronounced, with 2 rows of setae with hairs longer than those
of lateral border row.
Abdomen: marginal groove complete around last sternite.
Dimensions (mm): total length, mandible to elytral tip 30, elytral length 17.8, pronotal
length 7.1, pronotal width 9.1, humeral width 6.3, head width 6.3.
TYPE MATERIAL. Holotype: male-MEXICO, Michoacan, 16 km N.W. Coalcoman.
15-16 VIII 1982. T. W. Taylor, P. M. Sullivan Collectors. It will be deposited in the FSCA.
ETYMOLOGY. The name refers to the Mexican state in which the holotype was
DISTRIBUTION. This species is known only from this presumably high altitude site
in Michoacan.
AFFINITIES. This is the largest species of the "recticornis" group.

Key to the Species of Petrejoides
(modified from Castillo & Reyes-Castillo 1984)

1 Dorsal ridge of tibia II long, same height (usually high) for most of tibial
length ...................................................... ................................. 2
1' Dorsal ridge of tibia II short, high for <1/2 tibial length .......................... 13
2(1) Frons without inner tubercles ......................................... ............. 3
2' Frons with inner tubercles .................................................................. 8
3(2) Clypeus entirely or partly rugose and opaque, long ................................ 4
3' Clypeus entirely glossy, short, rectangular ........................................... 7
4(3) Central horn of medial frontal structure [MFS] elongate, apex free, reach-
ing, or almost reaching clypeus ....................................... .............. 5


Florida Entomologist 74(3)

4' Central horn of MFS short, apex not free, not reaching clypeus. Body
length 26-29 mm. GUATEMALA, Sierra de las Minas ..... P. pokomchii n. sp.
5(4) Labrum with a glabrous depression or concavity behind its mid-anterior
border, clypeus totally rugose. EL SALVADOR, Trifinio area .............
.............................. ..... P. salvadorae Schuster
5' Labrum without a distinct depression behind anterior border, clypeus par-
tially rugose ....................................................... ......................... 6
6(5') Clypeus trapezoidal rugose and opaque throughout except for narrow,
glossy anterior margin; fronto-clypeal suture indistinct medially. Lateral
ridges of median frontal structure at right angles to longitudinal body axis.
Femur I with anterior-ventral groove indistinct or absent. Body length
24.5-32 mm. GUATEMALA ......... P. guatemalae Schuster & Reyes-Castillo
6' Clypeus sub-trapezoidal, rugose and opaque only in posterior-medial area;
frontal-clypeal suture curved and distinct throughout. Lateral ridges of
median frontal structure curve slightly forward. Femur I with anterior-
ventral groove distinct. Body length 27.5-30 mm. HONDURAS .........
............................. ............................... ......... P. reyesi Schuster
7(3') Frontal-clypeal suture incomplete in middle, curved. Anterior border of
labrum slightly concave. Body length 24.5-26 mm. PANAMA, Chiriqui
............................................................. ................. P panam ae n. sp.
7' Frontal-clypeal suture complete, straight. Anterior border of labrum highly
concave. Body length 30 mm. MEXICO, MichoacAn ........ P. michoacanae n. sp.
8(2d) Infraocular ridge glabrous and smooth, internal tooth of left mandible
bidentate .................................................... ............................... 9
8' Infraocular ridge pubescent and punctate, or absent; internal tooth of left
mandible bi- or tridentate ............................................... ... .............. 10
9(8) Mesosternum with lateral opaque areas, anterior angle of pronotum not
projecting. Body length 22.5-26.5 mm. MEXICO: Sierra Madre del Sur,
Oaxaca .................................................................. P. jalapensis (Bates)
9' Mesosternum with central opaque areas, anterior angle of pronotum pro-
jecting. Body length 21.9-24.9 mm. COSTA RICA (also cited [erroneously?]
from Guatemala and El Salvador) .................... P. subrecticornis (Kuwert)
10(8') Horn of MFS short, not reaching clypeus ............................................ 11
10' Horn of MFS long and narrow, extending past clypeus. Body length 19.0-
20.7 mm. COSTA RICA, Sierra de Talamanca ................ P. tenuis Kuwert
11(10) Body length >22 mm, femur I with antero-ventral groove weak and incom-
plete ...................................................... ............ ... .................. 12
11' Body length 18.4-21.8 mm, femur I with antero-ventral groove absent.
MEXICO, southern Sierra Madre Oriental ................... P. orizabae Kuwert
12(11) Ligula with protuberance between insertion of labial palps, metasternum
without punctations delimiting posterior-lateral corners of disc. Body
length 23-27 mm. MEXICO, northern Sierra Madre Oriental, Nuevo Le6n..
............................................. P. sylvaticus Castillo & Reyes-Castillo
12' Ligula flat between insertion of labial palps, metasternum with >40 punc-
tations delimiting each posterior-lateral corner of disc. Body length 25.5
mm. MEXICO, Chiapas ............................................ P. chiapasae n. sp.
13(1') Horn of MFS short, not reaching clypeus .................................... 14
13' Horn of MFS long, reaching clypeus. Body length 23-28mm. MEXICO,
southern Sierra Madre Oriental, Veracruz ................ P. laticornis (Truqui)
14(13) Frons without central longitudinal ridge ............................................... 15
14' Frons with central longitudinal ridge. Body length 21.7-25.5 mm. MEXICO,
southern Sierra Madre Oriental, Oaxaca ..... P. mazatecus Castillo & Reyes Castillo

September, 1991


Schuster: New Passalidae

15(14) Horn of MFS with dorsal groove, dorsal anterior profile of elytra straight
................................................... .......... ................. .... ...... 16
15' Horn of MFS without dorsal groove, dorsal anterior profile of elytra V-
shaped ............................................................................................. 17
16(15) Femur I without anterior-ventral groove. Union of striae 1-10 with a row
of fine punctations. Dorso-lateral pronotum punctate. Body length 16.6-19.0
mm. MEXICO, Sierra Madre Oriental...P. nebulosus Castillo & Reyes Castillo
16' Femur I with anterior-ventral groove. Union of striae 1-10 with various
rows of punctations. Dorso-lateral pronotum smooth, without punctations.
Body length 24.1-26.7 mm. MEXICO, Sierra Madre del Sur, Guerrero
......................................... ...... ..... P. olmecae Castillo & Reyes Castillo
17(15')Infraocular ridge present. Clypeus inclined. Left internal mandibular tooth
bidentate. Body length 18.0-21.4 mm. MEXICO, Oaxaca ..................
............................................................ ..... P. recticornis (Burmeister)
17' Infraocular ridge absent. Clypeus vertical. Left internal mandibular tooth
tridentate. Body length 19.0-21.8 mm. MEXICO, Sierra Madre del Sur,
Guerrero .................................................................. P. imbellis (Casey)


At the time of the recent monograph of Petrejoides (Castillo & Reyes-Castillo 1984),
only 1/4 of the described species were from Mesoamerica. The presence of only 1 highland
species from northern Mesoamerica (the Central American Nucleus) was postulated to
be due to the diversification of another passalid, Chondrocephalus Kuwert, which oc-
cupies similar habitats. Chondrocephalus is known only from Nuclear Central America
(Chiapas, Guatemala, Honduras and El Salvador) above 1200 m elevation in wet forests
and cloud forests. Since then, however, the discovery of additional Petrejoides species
in Honduras (Schuster 1988), and El Salvador (Schuster 1989), as well as those described
here from Chiapas and Guatemala, raise the number of species from Nuclear Central
America to 5. Of these, all except the Honduran species are found in areas where at
least 1 species of Chondrocephalus is present. With the additional Panamanian species,
the total Mesoamerican assemblage is 8, almost 1/2 of the species in the genus. Also,
other possible new species from Chiapas and Colombia, in the collection at the Instituto
de Ecologia, Xalapa, Veracruz, are yet to be described. I believe further new species
may be found, especially in high mountains of Honduras, and perhaps eastern Panama.


Thanks to C. MacVean for reading the manuscript and the Universidad del Valle de
Guatemala for support. Thanks to M. Rojas for his interest and help in collecting P.
panamae and logistics and to C. Castillo and P. Reyes-Castillo for their opinions. Special
thanks to Defensores de la Naturaleza, Magali Rey-Rosa and Francisco Asturias for
arranging the trip to the Sierra de las Minas to collect P. pokomchii, and don Carlos
Mendez Montenegro and family for his expert field guidance and their warm hospitality.
Thanks to Ronaldo Perez and Luis Rodriguez for aid in collecting P. chiapasae, and
Camelia Castillo and Pedro Reyes-Castillo for confirming its identification as a new
species. Thanks to Pat Sullivan for providing the type specimen of P. michoacanae.


CASTILLO, C. AND P. REYES-CASTILLO. 1984. BiosistemAtica del g6nero Petrejoides
Kuwert (Coleoptera, Lamellicornia, Passalidae). Acta Zool. Mexicana 4: 1-84.

Florida Entomologist 74(3)

REYES-CASTILLO, P. 1970. Coleoptera: Passalidae: morfologia y division en grandes
grupos; generos americanos. Folia Entomol. Mexicana 20-22: 1-240.
REYES-CASTILLO, P., AND C. CASTILLO. 1985. Nuevas species de Coleoptera Pas-
salidae de la zona de transici6n Mexicana. An. Inst. Biol. Univ. Nal. Aut6n.
Mexico 56(1): 141-154.
un nuevo gdnero mesoamericano de Passalidae (Coleoptera: Lamellicornia). Folia
Entomol. Mexicana 73: 47-67.
REYES-CASTILLO, P. AND J. SCHUSTER. 1983. Notes on some Mesoamerican Pas-
salidae (Coleoptera: Petrejoides and Pseudacanthus. Coleopt. Bull. 37(1): 49-54.
SCHUSTER, J. C. 1988. Petrejoides reyesi sp. nov. (Passalidae) from Honduras. Coleopt.
Bull. 42(4): 305-309.
SCHUSTER, J. C. 1989. Petrejoides salvadorae sp. nov. (Coleoptera: Passalidae) from
El Salvador. Florida Entomol. 72(4): 693-696.
SCHUSTER, J. C. AND P. REYES-CASTILLO. 1981. New World genera of Passalidae
(Coleoptera): a revision of larvae. An. Esc. nac. Cienc. Biol., Mexico 25: 79-116.


Ministry of Agriculture, Animal Husbandry and Fisheries
P. O. Box 160
Paramaribo, Suriname
Fax 597-70301


The identification of a species of Bactrocera (formerly Dacus) in Suriname in 1986
resulted in a survey to obtain information on the host plants of this fruit fly and its
geographical distribution. During four years, 77 plant species were collected to obtain
fruit flies infesting those fruits and Jackson traps were set to collect adults. Only one
Bactrocera species and several Anastrepha species were reared from fruits. The Caram-
bola fruit fly was found in 15 plant species and flies were trapped from the eastern
border up to 560 30' longitude west in the Coronie district; to the south up to the
Brokopondo Lake and southwest isolated along the western border with Guyana.


A raiz de la identification de una especie de Bactrocera (conocida antes como Dacus)
en Surinam en 1986, se inicio un muestreo para obtener information sobre las plants
hospederas y la distribution geografica de la mosca. Duurante 4 aiios se colectaron 77
species de plants y se utilizaron trampas Jackson con el fin de atrapar adults. Un-
icamente un adulto de Bactrocera y varias species de Anastrepha emergieron de las
frutas colectadas. La mosca de la carambola se encontro en 15 species de plants y las
moscas fueron atrapadas desde el limited oriental, hasta la latitud 56' 30' oeste en el
distrito Coronie, y desde el sur hasta el lago Brokopondo y en el sureste a lo largo del
limited oeste con Guyana.

September, 1991


van Sauers-Muller: Bactrocera in Suriname 433

The Carambola fruit fly is a species of the genus Bactocera that has been introduced
into Suriname. It is part of a complex of some 44 species, known as the dorsalis complex
(R.A.I. Drew, pers. comm.). This species is endemic to Indonesia, Malaysia and Southern
Thailand. The Carambola fruit fly has not been formally described, although Drew &
Hancock (in prep.) will soon provide a scientific name. The Carambola fruit fly in
Suriname was originally misidentified as Dacus dorsalis (Hendel); this and other Bactro-
cera species were until recently included within the genus Dacus (Drew 1989). It has
also been referred to as Dacus sp. No true Dacus (sensu Drew 1989) occur in the New
Until 1986 there were no published records of Dacini (Dacus or Bactrocera spp.
occurring in Central or South America. Recent infestations of the Oriental fruit fly,
Bactrocera dorsalis (Hendel) in California and the Mariana Islands (USA) have been
eradicated (Stibick 1989).
The larvae of the fruit flies in the dorsalis complex have been recorded in at least
150-236 different hosts (Weems 1964, Stibick 1989), as well in fruits as in vegetables;
at that time all were considered to be the Oriental fruit fly. No information on specific
host plants for the different species in the complex is yet available.


In 1975 a few Carambola fruit flies were reared from curacao (or java) apple (Syzygium
samarangense, syn. Eugenia javanica). These flies were pinned and placed in the insect
collection of the Agricultural Experiment Station in Paramaribo where they remained
In December 1981, more Carambola fruit flies were reared from curacao apples,
bought at the Central Market in Paramaribo. A sample of adults was sent to the Insect
Identification Center in Beltsville, Maryland, and identified as Dacus sp. probably D.
dorsalis. They were brought to the attention of the Surinamese Ministry of Agriculture,
because no Dacini are native and if a species has been introduced, additional collections
should be made. However the importance of a species of Dacini possibly existing in
South America was not recognized at that time and therefore no additional fruit collec-
tions were made.
In March 1986, some Carambola fruit fly adults were reared from guava (Psidium
guajava) and sapodilla fruits (Manilkara zapota). These fruit flies showed a close re-
semblance to the depicted species of Dacus dorsalis in the Entomology Circular (Weems
1964). Specimens were sent by the author to the Insect Identification Center in Beltsville,
together with some specimens collected in 1975. These fruit flies were identified on July
31, 1986, by Dr. A. L. Norrbom, USDA-ARS, as Dacus dorsalis (Hendel). Drew (1989)
revised the Dacinae of the Australasian Region and transferred many of the species,
including dorsalis (Hendel), to the genus Bactrocera. It has been determined by R.A.I.
Drew (Entomology Branch, Queensland Department of Primary Industries, Australia),
that Bactrocera dorsalis actually is a complex of species, occurring in South East Asia,
of which several species are as yet undescribed. Drew and Hancock have examined
Surinamese specimens and determined that the flies belong to one of these undescribed
species (Drew & Hancock, personal communication). It is proposed by these authors
that the Surinamese Bactrocera species be called the Carambola fruit fly.
This species is endemic to Indonesia, Malaysia and Southern Thailand. The origin
of the Carambola fruit fly in Suriname is not known.
The human population of Suriname is very heterogenous; an important part (50%)
originates from Asian countries such as India, Indonesia and China. Because of these
historical ties there still is much traffic between these nations and Suriname. The intro-
duction of the Carambola fruit fly by trade or tourists from Indonesia to Suriname is
the most probable hypothesis.

Florida Entomologist 74(3)

Situation and Climate

Suriname is situated on the north coast of South America, between 2 and 6 north
latitude and from 54 to 580 west longitude. The total area is 163,820 sq. km.
Suriname has a tropical climate with daily temperatures ranging from 270 to 33C.
The mean annual rainfall ranges from 2200 to 2400 mm in the South East. The country
is still covered with native forest (90.6%), the majority of which is mesophytic forest.
Only about 3% of the land area is cultivated (including abandoned plantations).


After the identification of the Carambola fruit fly in 1986, a survey was started by
the Experimental Station of the Ministry of Agriculture of Suriname to gain information
on its geographical distribution and host plants. Fruits were collected from March 1986
to June 1990 and McPhail and Jackson traps baited with methyl-eugenol were installed
from October 1987.

Fruit collections

From March 1986 to June 1990, 1303 samples were taken of 77 plant species. The
samples contained one to 30 or more fruits, depending on fruit tree species and fruit
availability. The fruits were placed on sterilized soil or on sawdust in plastic containers,
covered with paper to prevent secondary infections. Pupae were collected every 2-3
days and placed in small jars containing sterilized soil/sawdust. These were held until
adults emerged at a temperature of 270-32C and a humidity of 80-95%. The total fruit
collections and fruit flies reared from each sample are presented in Table 1 and 2. The
most important hosts of the Carambola fruit fly in Suriname are Curacao apple (Syzygium
samarangense) and the sweet variety of carambola (Averrhoa carambola); other hosts
are star apple (Chrysophyllum cainito), sapodilla (Manilkara zapota), mango (Mangifera
indica), West-Indian cherry (Malpighia punicifolia) and guava (Psidium guajava).
Minor hosts are cashew (Anacardium occidentale), malay apple (Syzygium malaccensis),
Indian jujube (Zizyphus jujuba), surinam cherry (Eugenia uniflora), tropical almond
(Terminalia catappa), orange (Citrus sinensis), grapefruit (C. paradisi) and mandarin
(C. reticulata). The only species of Dacini found was the Carambola fruit fly. Many
Anastrepha spp. were reared, but no Ceratitis capitata were found.
All host fruits are cultivated; no wild fruits were infested. Six of the host fruits
originate from Central America, South-America or the Caribbean.


Trapping was started in October 1987 with 10 plastic McPhail traps, baited with 4
ml methyl-eugenol in the top of the trap and a water/soap mixture in the basal part of
the trap. All traps were placed in major or minor host trees (based on fruit collections)
in Saramacca and Wanica (from Paramaribo up to 80 km west), an area where the
Carambola fruit fly occurs. In May 1988, 23 plastic McPhail traps were installed in
Coronie, Nickerie, Para and Brokopondo, an area where the Carambola fruit fly was
unknown. The plastic McPhail traps were replaced by Jackson traps from June to April
1989. The majority of Jackson traps were installed after August 1989, when transport
was made available for the project. A total of 484 traps were installed in the coastal
area and in the interior by June 1990. The traps were serviced every two weeks up to
once a month, sometimes longer.

September, 1991

van Sauers-Muller: Bactrocera in Suriname



0 o






0 Co 00 tu 3~
M t M cCD f

~0<0- C000


00 0'11 0<1 Q 0<10


0 a
W' btjo.

F: g

E .F?
t3Ssr E

Zsi~ ,


00 00000 000000 000000-


0000000000 v0000000000

0 ~crl~r00 m~
M4 -4 C t014InN 0o-Io M

-1 ~ ~ -- -4 1-4NM10 oN L

ZFzzz---ZZI-Z-Z--- z z

M0 -
C~ C4

0 u U



ce~ 0

1-4 0
.~~~ @






Florida Entomologist 74(3)








C l '-4 N -0- 4 C- l


CI 0000000



Si- Cl lG


1 a

C d
^ Eg grS

s Il

U s; IImG .s G
riF3 C. ICI

~ S




September, 1991

x 1- O q

loi Ocoooooo00D

coo t0000000000

o0o C

LO co M


Z ---


I .

0 a

Cl CO r-4

- Cl cO r-4

" z z z


I ll
:V i


{(cd ,

41 ** ^


S I- O2 -M C- M
-l1-4 2 "

C XC t- UM '-4C CM
CO '-I C
1-4 C-4

C:IN CO 4 -CWZM Cl r-4
Tf --4 C3


van Sauers-Muller: Bactrocera in Surinar

000 C 0


000 C 0



to 00 cc Ni 0 0
00-4CC hC 0
~-4 --4-
i- l i:- Cl O


Z f-4- -'-

0 .B
I- Q)
E cb



cM -4 M 0

C 000

Cl CO 10

+ r

L0 m C 1-4

0t'-0 T-l4--C

t-( -4-1 -






(N C

C1 t-l

II 4 o, I i-l
+1 ,

2 s c ( S oc

m ~ .4 ^

ap ~ Ss ^
. i s

s s J s; a ,
E2 ? ? ?- S s
,s-s,-s_ .S -:
*. *^ ,+ ?- a 0



cd .0e

ne 437







C8 3

I e

i b
I~ E
a i -


So -g ^



El 'r,
yl d

438 Florida Entomologist 74(3) September, 1991

The traps were installed in apple or carambola trees. If these were not available,
minor host trees were used. No trap rotation was executed.
High densities of Carambola fruit flies (up to 800/month/Jackson trap) have been
encountered in populated areas in the districts Paramaribo, Wanica, Saramacca, Com-
mewijne, Para and Coronie; lower numbers (about 1/month/Jackson trap) were found
in the districts Sipaliwini and Brokopondo.


The results of fruit sampling and trapping are shown in Fig. 1.The Carambola fruit
fly has been recovered in Albina; about 150 km east of Paramaribo, near the border
with French-Guyana, by fruit sampling. To the west it has been found by trapping,
about 142 km west of Paramaribo. To the southwest flies have been found by fruit
sampling and trapping about 90 km from the capital and along this same road, in
Apura at the Corantyn River, 7 flies were recovered by trapping. In the south, about
90 km from Paramaribo, 16 flies were recovered by trapping. No recent information is
however available, due to reasons of safety.


The Carambola fruit fly originates from Indonesia, Thailand and Malaysia (pers.
com. D. L. Hancock, Entomology Branch, Queensland Department of Primary Indus-
tries, Australia). One of its major host plants in these countries is the carambola, as in
Suriname. The other major host plant in Suriname, the Curacao apple, also originates
from southeast Asia. No information is available on the species of Bactrocera infesting
this fruit tree in Asia. The true B. dorsalis, which, according to Dr. Drew, occurs in
Hawaii, has as major host plants avocado, banana, coffee, guava, mango, mountain
apple, papaya, rose apple and Surinam cherry (Bess & Haramoto 1961). No fruit fly
infestation is known to occur in Suriname from rose apple, bananas, avocado, papaya
and coffee, however only a few samples were taken during this research. Tropical almond
and Surinam cherry are minor hosts; only guava and mango have a higher infestation,
but are certainly not major hosts. A. carambola and S. samarangense both support
only a light infestation ofAnastrepha obliqua. These results agree with research executed
in Brazil (Malavasi et al. 1980), where only a very light infestation of carambola by
Anastrepha spp. is found.


The difference in host preference between the Carambola fruit fly in Suriname and
the B. dorsalis in Hawaii support the hypothesis that the many widespread fruit fly
populations that had been recognized as B. dorsalis are not one species as was supposed
until a few years ago, but a complex of species.
As the Carambola fruit fly is now introduced on the mainland of South America, the
feasibility of eradication must be reexamined. Special attention should be paid to wild
fruits, because if the Carambola fruit fly breeds in these fruits, eradication would not
seem feasible. In that case biological control might be a more promising approach. The
egg-larval parasite Biosteres arisanus (Sonan) (formerly Opius oophilus), which was
introduced in Hawaii in 1947 from Malaysia (Clausen et al. 1965) or the parasite Diachas-
mimorpha longicaudata (formerly Opius longicaudatus) could prove successful.

van Sauers-Muller: Bactrocera in Suriname 439



I '1 ^ 'I

I I., -- ,
z i' '

00 a*
I (

I5 0
,'7^, ', ,M -
I~ ~ ~ ~ ~ ~ ~ r J 'LJ*'"\' r

v^ v~^^^ ^^cr rt ^
o o E .
g ^.^|i|1
S ; ^ ^ "* o o !

Florida Entomologist 74(3)


This research was supported by the Ministry of Agriculture, Animal Husbandry and
Fisheries of Suriname, the Government of Brazil, the Food and Agricultural Organiza-
tion, the United States Department of Agriculture and the Inter-American Institute
for Cooperation on Agriculture. I am grateful to R.A.I. Drew and D. L. Hancock for
the identification of the Carambola fruit flies and to C. O. Calkins and A. L. Norrbom
for comments on the manuscript. The author thanks M. Sanches, C. Bram, N. Ramkhela-
wan and E. Donoe from the Ministry of Agriculture of Suriname for assisting in the
field work.


BESS, H. A., AND F. H. HARAMOTO. 1961. Contributions to the biology and ecology
of the Oriental fruit fly, Dacus dorsalis Hendel (Diptera: Tephritidae) in Hawaii.
Hawaii Agricultural Experiment Station, Technical Bulletin no. 44, June, 30 pp.
CLAUSEN, C. P., D. W. CLANCY, AND Q. C. CHOCK. 1965. Biological control of the
Oriental fruit fly (Dacus dorsalis Hendel) and other fruit flies in Hawaii. USDA,
Technical Bulletin no. 1322, march, 102 pp.
DREW, R.A.I. 1989. The taxonomy and distribution of tropical and subtropical Dacinae
(Diptera: Tephritidae), pp 9-14 in: Robinson, A. S. & Hooper, G. (eds). Fruitflies,
their biology, natural enemies and control. Elsevier Science, Amsterdam. Vol.
3A, chapter 1.2.
MALAVASI, A., J. S. MORGANTE, AND R. A. ZUCCHI. 1980. Biologia de Moscas-das
Frutas (Dipt.: Tephritidae). II Indices de infestacao em diferentes hospedeiros
e localidades. Rev. Brasil. Biol. Rio de Janeiro, 40(1): 17-24.
STIBICK, J.N.L. 1989. Action Plan Oriental Fruit Fly. USDA, APHIS, Cooperating
State Departments of Agriculture. 55 pp.
WEEMS, JR., H. V. 1964. Oriental fruit fly (Dacus dorsalis Hendel) (Diptera: Tep-
hritidae). Entomology Circular no. 21. Florida Dept. of Agriculture. 2 pp.


U.S. Horticultural Research Laboratory
2120 Camden Road
Orlando, FL 32803 USA

Florida Department of Agriculture and Consumer Services
Division of Plant Industry
Gainesville, FL 32602 USA


Pocket-like domatia on the underside of the leaves of wild grape, Vitis munsoniana
Simpson, were occupied by fungivorous (47.0%), predatory (7.8%), and rarely herbivor-

'Current address: Dept. of Botany & Zoology, Monash University, Clayton, Melbourne, Victoria3168, Australia.

September, 1991


Walter & Denmark: Wild Grape Vitis munsoniana

ous (0.8%) arthropods, primarily mites. At two sites in Orlando, FL, fungivorous taxa
increased from low levels during the spring dry season to high levels (40-96% of domatia
occupied by fungivores) after the onset of the rainy season in late May. These results
support the hypothesis that leaf domatia are used primarily by mites that are potentially
beneficial to the plant, and suggest that there may be a mutualistic relationship between
the plant and fungivorous mites.


Las domatia en forma de bolsillo en el enves de hojas de uva, Vitis munsoniana
Simpson, se encontraron ocupadas con fungivoros (47.0%), predadores (7.8%), y muy
pocos herbivoros (0.8%) artropodos, principalmente acaros. En dos localidades en Or-
lando, Florida, la taxa de fungivoros aumento de bajos niveles durante la primavera y
estacion seca, a altos niveles (40-96% de las domatia ocupadas con fungivoros) al comenzar
la estacion lluviosa a finales de Mayo. Estos resultados soportan la hipotesis de que las
domatia en las hojas son usadas principalmente por acaros que son potencialmente
beneficiosos a la plant, y sugiere que puede existir una relacion de beneficio mutuo
entire la plant y los acaros fungivoros.

Leaf domatia are pocket- or pit-like invaginations or tufts of hairs on the underside
of the leaves of many plants. Domatia are often inhabited by mites (Acari), and it has
been proposed that domatia are produced to encourage beneficial mites to remain on
leaves in mutualistic association with the plant (O'Dowd & Wilson 1989, Pemberton &
Turner 1989). Wild grape (Vitis munsoniana Simpson) is a common vine of mesic sites
in central Florida. The underside of the leaf is smooth and hairless, but at vein junctures
a roof-like extension of the veins and associated trichomes form pocket-like domatia.
The purpose of this study was: 1) to test the hypothesis that wild grape domatia
housed beneficial arthropods; 2) determine what species of Phytoseiidae and other pre-
datory arthropods were present on wild grape; and 3) determine if domatium use varied
over a period of time.
Samples consisted of 10 leaves of Vitis munsoniana Simpson chosen haphazardly
between 1 and 2 m above the ground from a patch of grapes between 9:00 and 16:00
hours. Unusually small leaves and leaves with extensive chewing-insect damage were
avoided. Leaves were picked, placed in a plastic bag in an ice chest, and transported
to the laboratory where they were frozen at -5C. Five sites were sampled: an area of
scrubby vegetation on the margin of Lake Ivanhoe in Orlando, FL (Ivanhoe), an open
stand of trees along a creek above Lake Adair in Orlando (Adair), a flatwoods forest at
Tosohatchie Reserve, 56 km east of Orlando (Tosohatchie), an area of scubby vegetation
on the margin of Wildcat Lake (Wildcat), and an oak-pine forest near Clearwater Lake
(Clearwater), both in the Ocala National Forest. The Adair site was sampled six times
between mid-April and early July 1990, and the Ivanhoe site was sampled on an approx-
imately weekly basis during the same period. Leaf surfaces were scanned under a
stereomicroscope, and all phytoseiid mites were collected. Then, all domatia on each leaf
were dissected under a stereomicroscope. Predatory arthropods and representatives of
morphospecies of other arthropod taxa were mounted in Hoyer's medium on microscope
slides for identification. The number of domatia occupied by live arthropods (including
eggs), arthropod "exuvia" (cast skins, empty egg shells, or dead animals) only, and
'empty (no icatin f past r current use by arthropods) was recorded. Leaf length
was measured from the petiole to leaf tip. Voucher specimens were deposited at the
Florida State Collection of Arthropods, Division of Plant Industry, Gainesville. Trophic
roles were estimated from gut contents, observations of trophic interactions, and the

442 Florida Entomologist 74(3) September, 1991

The average wild grape leaf contained 19.97 + 0.29 domatia. At the Ivanhoe site,
there was no significant change in number of domatia per leaf over time, but at the
Adair site there was a slight, but significant positive (slope = 0.063) linear relationship
(P < .003, r2 = .1448) with date.
Between 13 April and 8 July 1990 4394 domatia on 220 leaves of wild grape were
dissected. Pooled over all sites and dates, 53% of these domatia were occupied by live
arthropods; an additional 14% contained arthropod exuvia indicating previous occupation.
Many of the taxa collected from wild grape leaves are known to feed omnivorously on
foods as varied as pollen and spider mites, but most taxa could be classified by their
primary food. Predatory arthropods were found in 7.8%, fungivores in 47.0%, and her-
bivores in only 0.8% of the domatia (Table 1). Two or more species were sometimes
found in the same domatium. Eggs of thrips, phytoseiid, stigmaeid, tydeid, and win-
terschmidtiid mites were identified in domatia. Unidentified eggs were observed in
0.43% of domatia. Insects were found in only 1.98% of the domatia; the remainder were
occupied by mites.
Preliminary observations at the Tosohatchie site in late March indicated that
phytoseiid and tydeid mites were already present in domatia. Phytoseiid mites were
the most abundant predatory taxa, occupying on average one domatium per leaf (1.01
- 0.09), and being about four times as abundant in leaf domatia as on the leaf surface
(0.27 0.04). Nearly all egg (96.6%) and larval (87.5%) Phytoseiidae were collected in
domatia. Phytoseiid and cheyletid (Oudemansicheyla denmark (Yunker)) mites, Hap-
lothrips macroocellatus (Watson), and a cecidiomyiid larva were all observed feeding
on tydeid mites. There was no clear pattern of change in the number of domatia use by
predators at either the Ivanhoe (Fig. 1) or Adair sites (Fig. 2).
Tydeid mites were observed feeding on leaf spot fungi. These mites became increas-
ingly abundant with time, often with 30 or more eggs and molting nymphs packing a
domatium. Orthotydeus sp. nr. lindquisti (Marshall) was the most abundant species.
Although densities often exceeded 200 per leaf, there was no indication of leaf damage,
and body colors were pale or amber with no indication of chlorophyll. Acarid, win-
terschmidtiid, and oribatid mite guts all contained extensive fungal material. There
were significant positive linear relationships between number of domatia occupied by
fungivores and date at both the Ivanhoe and Adair sites P < .0001 for both sites, r2 =
.4891, .5225, respectively). Herbivores were never abundant in domatia.
The hypothesis that leaf domatia are used preferentially by beneficial mites is sup-
ported by this study. Predatory taxa (1.55 + 0.10 domatia per leaf) were ten times as
common as herbivores (0.16 0.03 domatia per leaf). Although leaf domatia comprise
only a small fraction of leaf area, phytoseiid mites were four times as likely to be collected
in domatia as on the leaf surface. Eggs of these mites are preferentially laid in domatia
since about 95% of all eggs and larvae (the first, generally non-feeding instar) were
collected there. Nearly one-half of all domatia examined contained fungivores (9.38
0.48 per leaf). In the early summer samples, domatia were often completely filled by
tydeid and winterschmidtiid mites. Winterschmidtiid mites ingest fungal spores and
mycelium which can be observed as gut boluses. Although high densities of these mites
may be associated with leaf damage (Dosse & Schneider 1957, Hughes 1962), there is
no indication that these mites, or the similarly feeding acarid and oribatid mites are
Tydeid mites ingest only fluids, so diet cannot be estimated from observations of
gut contents. Schruft (1972) reported that Tydeus goetzi (Schruft) and Pronematus
staerki (Schruft) were predators on eriophyid mites on grape leaves. Homeopronematus
anconai (Baker) occurs on grape and other crops in California where it appears to feed
primarily on pollen, honeydew, fungi, and eriophyoid mites, but it is a poor predator of
spider mites (Knop & Hoy 1983, Hessein & Perring 1986, 1988). McCoy et al. (1969)

_ ___

Walter & Denmark: Wild Grape Vitis munsoniana

Cecidomyiidae (0.30%)
Genus sp.
Thysanoptera (0.95%)
Haplothrips macroocellatus (Watson)
Tydeidae (44.37%)
Afrotydeus "munsteri" sensu Baker 1970
Tydeus Kochi Group
Tydeus spp.
Tarsonemidae (0.05%)
Tarsonemus waitei Banks
Tarsonemus sp. nr. yoshidai Ito
Winterschmidtiidae (2.25%)
Czenspinskia transversostriata (Oudemans)
Acaridae (0.15%)
Tyrophagus putrescentiae (Schrank)
Genus sp. undetermined
Oribatidae (0.05%)
Scapheremaeus sp.
Dometorina sp.
Phytoseiidae (5.06%)
Euseius hibisci (Chant)
Euseius mesembrinus (Dean)
Galendromus annectans (De Leon)
Galendromus halveolus (Chant)
Paraseiulella elliptica (De Leon)
Proprioseius anthurus Denmark & Muma
Proprioseius meridionalis Chant
Typhlodromalus limonicus (Garman & McGregor)
Typhlodromalus peregrinus (Muma)
Stigmaeidae (1.10%)
Agistemus denotatus Gonzalez
Agistemus divisus Gonzalez
Agistemus floridianus Gonzalez
Agistemus terminalis (Quayle)
Eryngiopus sp.
Bdellidae (0.10%)
Spinibdella depressa (Ewing)
Cheyletidae (0.50%)
Oudemansicheyla denmarki (Yunker)
Hemicheyletia wellsi (Baker)
Tetranychidae (0.09%)
Eotetranychus sp.
Homoptera (0.32%)
Diaspididae, Coccidae, Pseudococcidae
Thysanoptera (0.41%)
Scirtothrips nr. citri Moulton


Florida Entomologist 74(3)

25 I I I 1


o 20 --- PREDATORS
c /)

< 15 -

100 110 120 130 140 150 160 170 180 190 200

Fig. 1

found that Parapronenmatus acaciae Baker would not prey on phytophagous mites on
citrus, but would develop on two common leaf-inhabiting fungi. We observed tydeid
mites feeding on leaf spot fungi, and did not observe any eriophyoid mites that would
serve as alternate prey. However, it is likely that the tydeid mites on wild grape are
also facultatively predaceous.
In Central Florida, the winter dry season is followed by increasing rainfall in late
spring, and monsoon-like conditions in summer that facilitate the growth of phylloplane
fungi. At the Adair site, fungivores occupied only a quarter of the domatia during the
dry season in mid-April, but occupied > 70% after the rains returned in late-May. At
the Ivanhoe site, only 7% of domatia were occupied by fungivores in mid-Aprl, but
40-96% contained fungivores after late-May. A mutualistic benefit to the plant from the
presence of mites remains to be experimentally demonstrated, however, the results of
this study suggest that mites may help to remove fungi from the phylloplane.


We would like to express our appreciation to Herb Barrett, U.S. Horticultural
Research Laboratory, Orlando, FL for his help in determining the identity of the wild
grape, and to John Kethley, Field Museum of Natural History, Chicago, IL and Evert
grape, and to John Kethley, Field Museum of Natural History, Chicago, IL and Evert

September, 1991


Walter & Denmark: Wild Grape Vitis munsoniana




15 -

< /
10 -

W -

0 A----- - --- -

100 110 120 130 140 150 160 170 180 190 200


Fig. 2

Lindquist, Biosystematics Research Centre, Ottawa, Ontario, Canada for their help in
determining the identity of some of the mite taxa.
Contribution No. 740, Bureau of Entomology. Division of Plant Industry, Fla. Depart-
ment of Agriculture and Consumer Serivces, Gainesville, Florida 32602.


DOSSE, G., AND I. SCHNEIDER. 1957. Biologie und Lebenweise von Czenspinskia
lordi Nesbitt. Z. agnew. Zool. 44: 403-418.
KNOP, N. F., AND M. A. HOY. 1983. Biology of a tydeid mite, Homeopronematus
anconai (n. comb.) (Acari: Tydeidae), important in San Joaquin valley vineyards.
Hilgardia 51: 1-30.
HESSEIN, N., AND T. M. PERRING. 1986. Feedinghabits of the Tydeidae with evidence
of Homeopronematus anconai (Acari: Tydeidae) predation on Aculops lycopersici
(Acari: Eriophyidae). Internat. J. Acarol. 12: 215-221.
HESSEIN, N., AND T. M. PERRING. 1988. The importance of alternative foods for
the mite Homeopronematus anconai (Acari: Tydeidae). Ann. Entomol. Soc.
Amer. 81: 488-492.
HUGHES, A. M. 1962. The genus Calvolia Oudemans, 1911 (Acari: Sarcoptiformes).
Acarologia 4: 48-63.

Florida Entomologist 74(3)

McCoY, C. W., A. G. SELHIME, AND R. F. KNAVEL. 1969. The feeding behavior
and biology of Parapronematus acaciae (Acarina: Tydeidae). Florida Entomol.
52: 13-19.
O'DOWD, D. J., AND M. F. WILSON. 1989. Leaf domatias and mites on Australasian
plants: ecological and evolutionary implications. Biol. J. Linn. Soc. 37: 191-236.
PEMBERTON, R. W., AND C. E. TURNER. 1989. Occurrence of predatory and fungivor-
ous mites in leaf domatia. American J. Bot. 76: 105-112.
SCHRUFT, V. G. 1972. Das vorkommen von milben aus der Familie Tydeidae (Acari)
an reben. Z. ang. Ent. 71: 124-133.

--L --- L-- -


114 Long Hall, Department of Entomology
Clemson University
Clemson, SC 29634-0365


Improved techniques were developed for the rearing and maintenance of Mantispa
viridis Walker. These have proven highly successful and have eliminated problems
encountered by other researchers working on mantispine developmental studies, and
should allow subsequent studies to be conducted under standardized conditions. These
techniques also have been used in maintaining colonies of beaded lacewings (Neuroptera:
Berothidae), various spider egg parasitoids and predators, and are adaptable for other
arthropods that require individual confinement.


Se desarrollaron tecnicas mejoradas para la crianza y mantenimiento de Mantispa
viridis Walker. Estas tecnicas tuvieron much exito y se eliminaron problems encon-
trados por otros investigadores; de la misma forma estas tecnicas permitieron que se
realizaran studios subsiguientes en condiciones normales. Estas tecnicas se utilizaron
tambien para mantener colonies de los insects de encaje con borlas (Neuroptera: Be-
rothidae), y para various parasites de huevos de arafias. Estas tecnicas pueden ser adap-
tadas para la crianza de otros artropodos que requieran confinamiento individual.

Most previous developmental data reported for members of the Mantispinae
(Neuroptera: Mantispidae) are difficult to compare intra- and interspecifically due to
studies having been conducted under uncontrolled conditions of temperature, photo-
period, relative humidity (RH) and larval rearing techniques. The exception is a stand-
ardized rearing technique developed by Redborg & MacLeod (1983, 1985). They con-
structed rearing cells by drilling wells into a hardened mixture composed of powdered,
activated carbon and plaster of Paris. Spider eggs were added to the wells, a first instar
mantispid was placed in each, and the top was sealed with a glass cover. Egg incubation

September, 1991

Brushwein and Culin: Mantispa viridis 447

and larval development were monitored under controlled conditions of 25C, 80% RH,
and 16L:8D photoperiod. Their standardized rearing procedure allowed the calculation
of mean developmental data for each stage of three species, Climaciella brunnea (Say)
(Redborg & MacLeod 1983), Mantispa sayi Banks (as Mantispa uhleri Banks), and
Mantispa viridis Walker (Redborg & MacLeod 1985).
This report presents the modification of laboratory techniques used to rear and
maintain colonies of M. viridis. These rearing procedures, while based on the work of
the preceding authors, eliminate several problems they encountered and provide a sav-
ings of both time and space.


Rearing Materials

Larval Rearing Containers. Five dram plastic pill-bottles with "child-proof" caps
(Inventive Packaging Corp., Denver, CO) were used to rear individual M. viridis larvae.
The top 6 mm of these have a flange 1 mm wider than the bottom portion. Pill-bottle
size was reduced by shaving 3.4 cm off the bottom using a flat wood boring bit (2.54
cm diameter) set in a drill press. This produced a flanged rearing container 2.5 cm high
by 2.5 cm top outer diameter (Fig. 1). Reducing the size of the containers served two
purposes. First, it decreased the interior volume from approximately 19.2 ml to 9.3 ml,
thereby reducing the amount of substrate needed. Second, it made the container consid-
erably less prone to being knocked over.

a c

Fig. 1. Rearing container for M. viridis larvae, cross section. a, top-molded, cover
slip seating well (15.8 mm diameter by 0.8 mm deep); b, rearing cell (8.3 mm diameter
by 7.1 mm deep); c, cover slip access well (3.2 mm diameter by 1.6 mm deep); d, acrylic
support; e, activated carbon/plaster of Paris substrate.
Fig.1. earig cntaier.or.................... setion....opmoldd,..ve
slip seating.w.l......8m di etrb 0.........................3 mm diaete
by 7.1 mm de................... a*ce...........2m d aetrb 1.6.m de p;.,acyi
suppot; e actiatedcarbo ..later.o.Pars.subtrat.......

448 Florida Entomologist 74(3) September, 1991

To construct a rearing cell that would effectively confine first instar M. viridis larvae,
a built-in seal for the top of the rearing cell was incorporated. Using a number 9 cork
borer, 15.8 mm diameter circles of 0.8 mm thick polypropylene were cut out and appressed
to the centers of 3.8 cm squares of masking tape. The top of a cut-down container was
then attached to the masking tape so that the polypropylene circle was centered within
the opening. Each container was then filled with a wet mixture of 1 part powdered,
activated carbon to 9 parts plaster of Paris (Redborg & MacLeod 1985). A 5:4 (wt:v)
ratio of substrate to distilled water produced an optimum consistency for mixing and
pouring. To eliminate air pockets trapped in the wet substrate, each container was
tapped several times on a flat, hard surface. An hour later, substrate-filled containers
were transferred to an environmental chamber (350C) and held for an additional 48 h to
complete drying. When the substrate was completely dry, the masking tape with the
polypropylene circle still attached, was peeled away leaving a smooth outer surface
bordering a 15.8 mm diameter by 0.8 mm deep molded well. This well prevented a cover
slip from sliding off the rearing cell as it would be seated below the top surface. Using
an 8.3 mm diameter drill bit set into a drill press, a 7.1 mm deep hole that would serve
as the rearing cell was drilled in the center of the molded well. A second hole, 3.2 mm
diameter by 1.6 mm deep, was drilled at the junction of the top surface and molded
well. This provided access to facilitate removal of the 12 mm diameter circular glass
cover slip used to seal the rearing cell. Compressed air was used to remove pulverized
substrate material produced by the drilling.
Finished rearing containers were supported inside a closed, rectangular plastic box
(18.7 x 13.5 x 8.4 cm) (Stock No. T69C, Tri-State Molded Plastics, Inc., (TSMP), Dixon,
KY) by means of a 3.2 mm thick acrylic sheet in which 20 evenly-spaced 23.8 mm holes
had been drilled (Fig. 2). Each rearing container fit snugly into the plastic support
reducing the possibility of jarring the cover slip from the rearing cell. The 6 mm lip of
the rearing container that projected above the surface of the acrylic support facilitated
manipulation of individual containers. Relative humidity (RH) was maintained at 80%
within the plastic box with a saturated water solution of KBr (Solomon 1951).
To test the effectiveness of this design, twenty newly closed (<12 h old) first instar
M. viridis from a single clutch of eggs were placed individually in twenty cells. Cells
were examined after 24 h and the presence or absence of larvae recorded. The test was
replicated six times using larvae produced by six different wild-caught females.
Spider Eggs. Eggs of numerous spider species belonging to the families Theridiidae,
Lycosidae, Salticidae, and Araneidae were used to rear M. viridis. However, nonaggluti-
nate eggs of theridiids such as the house spider, Achaearanea tepidariorum (C. L.
Koch), and the black widow, Latrodectus mactans (F.), were easiest to employ. Within
72 h of deposition, eggs of these species separate from each other and could be poured
from an opened egg sac into a rearing cell or storage container. Eggs which remained
attached to the silken lining of egg sacs could be freed by carefully dislodging them with
forceps or a moistened camel's-hair brush.
Field-collected egg sacs and those obtained from laboratory colonies of mated female
spiders were opened and the eggs collected in 50 mm x 9 mm plastic petri dishes having
tight-fitting lids (Falcon Co., Oxnard, CA). Bottoms and sides of petri dishes were lined
with 55 mm diameter filter paper disks to prevent eggs from "jumping out" of, or
adhering to, the sides of the dish due to static electricity present on the plastic surface.
Colonies of A. tepidariorum and L. mactans were maintained in an environmental
chamber at 27.8 + 2C, 80 5% RH with a 14L:10D photoperiod. Spiders were fed
either Drosophila melanogaster Meigen or Musca domestic L. adults depending on
the size of the spider.
Eggs which were not immediately used to rear M. viridis were stored for later use
at -20C (Redborg and MacLeod 1985). Prolonged storage (> 3 months) of small eggs,


I _

Brushwein and Culin: Mantispa viridis


Fig. 2. Maintenance container for M. viridis larvae, top removed, a, plastic box; b,
acrylic support; c, rearing containers; d, plastic Petri dish containing a saturated water
solution of KBr.

such as those of A. tepidariorum, resulted in complete desiccation of the eggs. Larger
eggs, such as those of L. mactans, could be stored for longer than 6 months with no
appreciable desiccation.
Freezer-stored eggs were held at room temperature for at least 2 h before being
used to rear M. viridis. This allowed the eggs to thaw completely and any water to
evaporate that initially condensed on the chorions. Damaged and desiccated eggs were
discarded to reduce the possibility of fungus developing on the eggs within a sealed
rearing cell.
Adult Maintenance Containers. Seven dram, plastic vials fitted with snap-cap lids
(Cat. No. 133157-1, American Scientific Products, McGraw Park, IL) were used to house
individual adults. Bottoms of these vials had been cut away and replaced with 25 mesh/cm
Saran (Style 5038400, Chicopee, Gainesville, GA) screening to ensure adequate venti-
lation. Two-thirds of the interior wall was lined with filter paper to provide a surface
where pharate adults could undergo the final ecdysis. Vials were supported inside a
plastic box (Stock No. T69C, TSMP, Inc.) as described previously (Fig. 3).
Adults were allowed to mate in 506 ml capacity, round (10.5 cm diameter by 7 cm
deep), plastic containers with screw-top lids (Stock No. T40C, TSMP, Inc.). Two 5 cm
diameter holes were cut in opposite sides of the container and a 7.6 cm diameter hole
was cut in the lid. These openings were covered with Saran screening (25 mesh/cm)
to provide ventilation.


Florida Entomologist 74(3)

Fig. 3. Maintenance container for M. viridis adults, top removed, a, plastic box; b,
acrylic support; c, bottom screened, 7 dram vials; d, plastic Petri dish containing a
saturated water solution of KBr.

Rearing and Maintenance

Larval Rearings. Larval hearings were initiated by introducing either fresh or
freezer-stored eggs into a rearing cell. In general, enough eggs were added to fill the
cell one-half to two-thirds full. Using a camel's-hair brush moistened with distilled water,
a first instar M. viridis was transferred into a cell and the top was sealed with a glass
cover slip. Sealed rearing containers were placed in the plastic box and transferred to
an environmental chamber maintained at 25C with a 16L:8D photoperiod. Five days
after the third instar had spun a cocoon, the cover slip was carefully removed from the
rearing cell and the cocoon was transferred to an adult maintenance container.
Adult Maintenance and Reproduction. Adult male and female M. viridis were main-
tained individually in 7 dram, ventilated vials as described previously (Fig. 3) and were
fed one housefly per day. Interior sides of vials containing females were lined with 8.9
mm x 3.7 mm sheets of clear acetate (Ful-Vu Report Covers, Cooks' Inc., Darby, PA).
The interior surface of the cap was lined with a 2.5 cm diameter acetate circle. The
clear liner allowed direct observation of female oviposition and feeding behavior, and
provided an easily removed and replaced oviposition substrate. This facilitated making
egg counts either directly with the aid of a stereomicroscope or indirectly by photograph-
ing the clutch of eggs and counting the number of eggs from a black and white print
(Redborg & MacLeod 1985). Females rarely oviposited on the screened bottoms of the
vials and usually this occurred with females >90 d old. In such cases, screens were
easily removed and replaced.


September, 1991

Brushwein and Culin: Mantispa viridis 451

For mating, a single pair of 4 to 10 day old, recently fed adults was placed into a
mating container. All pairings were initiated during the photophase cycle under condi-
tions of 27.8 + 20C, 60 + 10% RH, and 15L:9D photoperiod.
Egg Incubation. Acetate vial liners with attached eggs were transferred to 7 dram,
snap-cap vials which had the bottom surfaces replaced with fine-mesh polyester cloth.
Before closing the vial, the top was sealed with Parafilm to prevent larvae from crawling
over the lip of the vial and becoming lodged between the lip and cap. Alternatively, a
1 cm band of Fluon applied to the top interior sides of the vial effectively prevented
the highly mobile first instars from escaping. Egg clutches were incubated in a plastic
box containing a saturated water solution of KBr and maintained under the same environ-
mental conditions as adults.


These techniques for rearing and maintaining laboratory colonies of M. viridis offer
several advantages over those developed and employed by Redborg & MacLeod (1985).
Of primary importance in rearing mantispids is the initial establishment of first instar
larvae. The small size of unengorged M. viridis first instars (approximately 1.3 mm long
by 0.1 mm thick) and their mobility presents the problem of containing larvae within a
defined area while allowing direct observation of larval development.
Redborg & MacLeod (1985) addressed this problem by covering the top of the larval
rearing cell with a 1 cm square glass cover which had been cut from a standard microscope
slide. To completely seal the rearing cell, the substrate surrounding the open cell was
moistened with distilled water. They noted that sealing the rearing cell with water could
cause condensation within the cell, promoting the growth of a mold lethal to mantispid
The design presented here eliminates this potential problem by employing a seal for
the rearing cell that does not require the addition of water. Because the cover slip rests
on the molded flat surface surrounding the rearing cell, contact between the glass and
edge of the substrate is virtually complete (Figs. 1 and 2), providing an effective barrier
against larval escape. None of the 120 larvae used to test the efficiency of this design
escaped. Further evidence for the effectiveness of this seal comes from the number of
successful mantispid hearings recorded over a 4 year period. Less than 1% (n=237)
failed because the first instar escaped. The cover slip also functions as a surface on
which engorged third instar larvae could attach an irregular scaffolding of silk lines
which serve as a framework for the silken cocoon in which pupation occurs.
As an additional benefit, size of the rearing cell can be varied without decreasing
the effectiveness of the seal. Cells have been used which range in size from 4.8 mm
deep by 3.2 mm top diameter to 11.1 mm deep by 10.3 mm top diameter to rear M.
viridis. In no instance did varying the size of the rearing cell affect the integrity of the
Maintaining adult M. viridis in 7 dram, ventilated vials supported within a plastic
box offers considerable space savings within an environmental chamber. Because each
box is a self-contained unit, they can be stacked, transported to another location, or
easily manipulated. These features would be of considerable value to any researcher
who has limited facilities (i.e., a single environmental chamber) or who must share space
with other individuals.
To determine the fecundity and fertility ofM. sayi, Redborg & MacLeod (1985) used
filter paper to line the interior top and sides of glass jars in which individual females
were kept. They counted the number of eggs per clutch from a black and white photograph
taken after eclosion and after the filter paper had been stained with India ink. In contrast,
we found that clear acetate provides a suitable oviposition surface and offers several

452 Florida Entomologist 74(3) September, 1991

advantages over filter paper linings. The behavior of M. viridis females can be observed
both during oviposition and feeding. As mentioned previously, fecundity and fertility
can be calculated either by directly counting the number of eggs per clutch using a
stereomicroscope or from a photograph of the clutch. In both instances the acetate does
not have to be stained or colored prior to counting or photographing eggs, but merely
placed against a dark background.
While developed specifically to rear and maintain colonies of M. viridis, these tech-
niques are not limited to this species. They have proven equally suitable for incubating
eggs, rearing larvae, and maintaining adults of two other mantispid species, Mantispa
interrupt Say and Mantispa pulchella (Banks). With few or no modifications this
rearing technique should prove equally adaptable to other mantispid species.
With minor changes, adult rearing containers have been used to rear and maintain
colonies of three species of beaded lacewings (Neuroptera: Berothidae) (Brushwein 1987,
unpublished data), numerous predators and parasitoids of spider eggs (Diptera and
Hymenoptera) (Brushwin unpublished data), and six species of external spider
parasitoids (Hymenoptera: Ichneumonidae: Polysphinctini) (Brushwein unpublished
data). This demonstrates the utility and adaptability of these rearing techniques. They
should be suitable for many other insect species.


We thank Kevin Hoffman, Gloria McCutcheon, Tom Skelton and three anonymous
reviewers for their comments and suggestions on earlier drafts of this manuscript. We
thank Frances Scarborough for making editorial comments on the Spanish summary.
J. R. Brushwein's current address is 517 Lake Ave., Lehigh Acres, FL 33936. This is
Technical Contribution No. 3098 of the South Carolina Agricultural Experiment Station,
Clemson University. Address reprint requests to J. D. Culin.


BRUSHWEIN, J. R. 1987. Bionomics of Lomamyia hamata (Neuroptera: Berothidae).
Ann. Entomol. Soc. Amer. 80: 671-679.
REDBORG, K. E., AND E. G. MACLEOD. 1983. Climaciella brunnea (Neuroptera:
Mantispidae): A mantispid that obligately boards spiders. J. Nat. Hist. 17: 63-73.
REDBORG, K. E., AND E. G. MACLEOD. 1985. The developmental ecology of Mantispa
uhleri (Neuroptera: Mantispidae). Ill. Biol. Monogr., No. 53. Univ. Illinois Press,
Urbana, IL.
SOLOMON, R. H. 1951. Control of humidity with potassium hydroxide, sulfuric acid,
or other solutions. Bull. Entomol. Res. 42: 543-554.

Galliart & Shaw: Mating Success in Katydid Males 453


Department of Zoology
Iowa State University
Ames, IA 50011


Amblycorypha parvipennis StAl males are unique chorusers; adjacent males alternate
4-5 s phrases frequently overlapping the end of a partner's phrases, and where phrases
overlap, phonatomes (phrase subunits) are synchronized. We explored the effect of
weight and a number of sound parameters, including frequency, intensity, temporal
parameters and phrase phase relationships during chorusing, on male mating success
using laboratory tests in which single females were exposed to pairs of chorusing males.
Females mated with males that were heavier, louder, and overlapped their partner's
phrases less when in the presence of a sexually-receptive (sound-producing) female.
Male spermatophore weight was correlated with male body weight; this suggests that
females may choose heavier males to obtain larger spermatophores (10-20% of male's
wet body weight) upon which females feed. Sound level rather than sound frequency
may be important in mating success because competing males produce sounds in close
proximity to competitors and females and this proximity negates the environmental
degradation of sound level over greater distances. The reduction in the rate at which
eventual successfully mating males overlap unsuccessful males in the presence of a
female suggests that males compete to reduce the rate at which they overlap the end
of a competitor's phrases. This ability, as well as weight and sound level, appears to be
utilized by females or competitors in determining the "superior" male.


Los machos de Amblycorypha parvipennis Stal conforman un coro unico; los machos
produce 4-5 frases alternas las cuales sobreponen en el final de la frase emitida por
otro macho, y cuando estas frases estan sobrepuestas, los tonos foneticos (subunidades
de frase) se sincronizan. Se investigo, en experiments de laboratorio, utilizando hembras
expuestas a un coro de machos, el efecto del peso y various parametros del sonido (los
cuales incluyen, frequencia, intensidad, parametros temporales y fases de frases y sus
relaciones durante el coro), y su efecto en el exito del apareamiento. Las hembras se
aparearon con machos que eran mas pesados, mas sonoros, y los que sobreponian menos
las frases en las de sus companfieros, cuando estaban en la presencia de una hembra
receptive (productora de sonido). Se relaciono el peso del espermatoforo del macho con
el peso del cuerpo; esto sugiere que las hembras escojen machos mas pesados con el fin
de obtener espermatoforos mas grandes (10-20% del peso fresco del macho) de los cuales
ellas se alimentan. El nivel del sonido fue mas important que la frequencia en el exito
del apareamiento, porque los machos que estan compitiendo produce sonidos junto a
otros competidores, y junto a las hembras, y esta proximidad reduce la degradacion
ambiental del sonido en largas distancias. La reduction de la proporcion en que los
machos se aparearon con exito sobrepasa a los machos sin exito de apareamiento en
presencia de una hembra, sugiere, que los machos compiten para reducir la velocidad a
la cual ellos sobreponen frases a el final de las frases de los competidores. Esta abilidad,
asi como el peso y el volume del sonido, paracen ser utilizados por las hembras o por
los competidores con el fin de determinar el macho "superior".

Florida Entomologist 74(3)

The males of many species of singing Orthoptera aggregate spatially (Campbell &
Clarke 1971, Cade 1976, Campbell & Shipp 1979, Shaw et al. 1981, 1982) and perform
communal sexual displays, i.e., exhibit temporal aggregations of song which Walker
(1983) termed "sprees." Alexander (1960, 1967, 1975) called such spreeing aggregations
"choruses" and considered them analogous to male anuran choruses which attract con-
specific males and females to breeding waters (Wells 1977). There are a variety of chorus
types classified on the basis of the temporal relationships of the song components of
adjacent males (Greenfield & Shaw 1983).
The songs of Orthopteran males are undoubtedly involved in determining mating
success whether by intrasexual competition or intersexual mate choice. Males may
compete for singing sites and chorusing is involved in spacing dynamics (Shaw 1968,
Greenfield & Shaw 1983, Latimer & Schatral 1986, Meixner & Shaw 1986). There is
experimental evidence that male spacing within choruses is affected by males hearing
the sounds of neighbors (Campbell & Shipp 1979, Thiele & Bailey 1980, Bailey & Thiele
1983, Latimer & Schatral 1986, Latimer & Sippel 1987).
Orthopteran songs function to attract sexually receptive females (Alexander 1960,
1967, Dumortier 1963, Otte 1977, Forrest 1980, 1983) and attract or repel competing
polygamous males (Morris 1972, Cade 1979, Forrest 1980, 1983). Studies of male acoustic
cues involved in mating success have shown that females mate with males that produce
louder songs (Cade 1979, Forrest 1980, 1983, Gwynne 1982, Bailey 1985), lower song
frequencies (Bailey 1985, Latimer & Sippell 1987), and that initiate song and sing
more during collective calling bouts (Busnel 1967).
Only Busnel's (1967) studies implicated some aspect of chorusing in mating success.
He studied three species of katydids of the genus Ephippiger in which adjacent males
tend to alternate production of phrases. In paired interactions, the male which initiated
more calling bouts and sang more phrases during calling bouts attracted more females.
Male weight is another apparently important factor in male mating success of singing
Orthoptera (Gwynne 1982, 1983, Forrest 1983, Simmons 1986a,b). This may be related
to heavier males producing larger external spermatophores upon which females feed
(Gwynne 1982, 1983).
Our study of the acoustic and reproductive behavior of a phaneropterine katydid,
Amblycorypha parvipennis Stil (Shaw et al. 1990) suggested that this would be an
excellent species to investigate regarding the role of song and chorusing in mating
success. In paired interactions, males alternate 4-5 s phrases, which frequently overlap
the end of the partner's phrases, and, where phrases overlap, phonatomes (sounds
produced by a single cycle of wing movement: Walker & Dew 1972) are synchronized
(Fig. 1). During paired interactions, males differ in the number of times the beginnings
of their phrases overlap the phrase endings of the other katydid (Fig. 1) (Shaw et al.
1990). Unlike most katydids, phaneropterine females produce short "ticking" sounds in
response to male sounds, and one or more males move to the ticking female (Shaw et
al., 1990). When near the female, A. parvipennis males frequently encounter one another
and may kick or push with front or hind legs. crawl over, or mount one another. Males
court by ceasing singing, raising their wings, and backing toward a female or another
male. These interactions can last for hours and may result in some males abandoning
courtship and leaving the area (Shaw et al., 1990).
This study was designed to observe mating under controlled laboratory conditions
where acoustic parameters could be readily measured and physical interactions easily
observed. This involved a series of trials where individual females were given a choice
of two chorusing males. The following questions were asked. 1) Do acoustic differences
between males, including phrase phase relationships during acoustic interaction, reflect
differences in the males' abilities to achieve mating success? 2) Do females mate with
heavier males and, if so, is male size correlated with spermatophore size? 3) if more


September, 1991

Galliart & Shaw: Mating Success in Katydid Males 455

A -a- -- PHRASE-

1...... I


2 I, t i g | ltlll. lt9 I P N IIUlII e IH WIgIg iIg,,gg~gg|ij 1111 I|IIi III gtlglggig|wl alglrg iit
_______________--- --b ---_________
6 8 16 24

Fig. 1. Oscillographs of two paired acoustic interactions (choruses) by A. parvipennis
males. A-several phrases recorded from each of a pair of chorusing males. The initial
part of the phrase of each male overlaps the latter part of the phrase of the other male.
In this selection, both katydids overlap the other katydid three times, i.e., the overlap
number of each katydid is three. The mean length of overlap, i.e., the mean length of
time that each katydid overlaps the other katydid is the sum of the period that each
phrase is overlapped divided by three. B-several phrases recorded from another pair
of chorusing males. In this selection, katydid 2 overlaps katydid 1 but 1 does not overlap
2. Therefore, the overlap number and mean overlap time for 1 is zero. a-indicates the
time that the phrase of 1 overlaps the phrase of 2. b-indicates the time that the phrase
of 2 overlaps the phrase of 1. t t designates the initial phonatomes of katydid 2 that are
out of phrase with the terminal phonatomes of katydid 1.

than one parameter is related to mate choice, are these parameters correlated with one


Subjects and Housing

Specimens of A. parvipennis were collected from a prairie west of Ames High School,
Ames, Iowa. Using flashlights, individuals were collected at night from 1 July to 6
August 1986 and 22 June to 21 July 1987. Singing males were easily collected as needed
throughout the testing period. Females were much more difficult to collect. Most females
were collected by searching food plants, especially horsemint (Mentha longifolia) and
wild grape (Vitis), at night. A few nymphs were collected by sweeping food plants with
an insect net during the day. In order to ensure that females were receptive (assumed
virgin), they had to be captured as nymphs or adults prior to the second week of male
singing. Only 13% of the sexually receptive females maintained in the laboratory were
captured after the first week of male singing. The approximate age range for females
used in trials was 1 to 4 weeks. Female nymphs were reared in the laboratory in
environmental chambers (Percival Refrigeration and Manufacturing Co.) (14L:10D, 24-
Males were individually marked using colored nail polish and isolated in 10x10x17
cm wire screen cages and placed around the laboratory which was maintained on the
same light-dark and temperature regimes as the environmental chambers. Females were
housed together in 34x33x31 cm wire screen and wood cages inside an environmental

Florida Entomologist 74(3)

chamber. This isolated them from male sounds until they were used in a mate choice
trial. After they were used in a trial, females were housed collectively in a separate
cage. All insects were fed leaves of horsemint or wild grape and chicken starter feed.
Water was provided in cotton-capped vials.

Two-Choice Discrimination Tests

Two-choice discrimination tests enabled individual females to listen to and interact
with two males. Receptive females were chosen by placing a cage of females in the
laboratory of singing males and selecting a ticking female. We ran 24 trials using a
different female and different pair of males in each trial. The number of trials was limited
by the difficulty in finding receptive females. Females were weighed before and after
every trial; males were weighed before and after the first 10 trials but only following
the last 14 trials. Handling males prior to a trial greatly reduced the probability that
they would sing again that day.
All trials were performed in an acoustic isolation chamber (4.6x5.3x2.4 m; Industrial
Acoustics Co., Inc.) at room temperature (24-25C). Caged males were placed 3.4 m
apart (within the range of the most common nearest neighbor distances in the field
[Shaw et al. 1981] ) at the ends of three tables placed end-to-end. After males began to
sing, sound level was measured approximately 5 cm above each male's stridulatory
apparatus. This close-range measurement was taken because sound level at the intermale
midpoint could vary several dB (re: 0.002 dynes/cm2) depending upon the position of the
male in the cage. Maximum sound levels were measured at frequency bands centered
at 8 and 16 kHz using a Bruel & Kjaer (B & K) type 2203, precision sound level meter
in conjunction with a B & K, type 1613, octave filter set. Following sound level measure-
ments, a 30 s recording (B & K, type 4133, microphone; B & K, type 2615, microphone
preamplifier; B & K, type 2801, power supply; Tektronix, type 122, preamplifier; Tek-
tronix, type 160A, power supply; Precision Data, type PI-6204, instrumentation tape
recorder) was made for frequency analysis. Frequency spectra of song samples were
determined using a B & K, type 2033, frequency analyzer. The upper frequency limit
of the recording equipment was 40 kHz. Although the upper frequency limit of the B
& K frequency analyzer is 20 kHz, by recording the songs at 37.5 in/s on the Precision
Data recorder and playing the songs at 15 in/s on a Nagra III tape recorder, frequencies
up to 40 kHz could be analyzed.
The frequency spectra of A. parvipennis male sounds are analyzed elsewhere (Shaw
et al., 1990). Although sound frequencies extend beyond 100 kHz, most sound energy
is below 50 kHz with maximum energy occurring at a mean frequency of 10.5 kHz.
However, male sounds may show one to three lower energy peaks between 10 and 40
kHz, the latter being the upper frequency range we used in the analysis of male sounds.
In this study, we asked whether pairs of males differed in the lower two peak frequencies
(X = 10585 + 1643 and 19330 1458 Hz, Shaw et al. 1990).
Following the recordings for frequency analysis, 10-min recordings were made of
the males chorusing, without and then with a sexually receptive (ticking) female (within
a circular screen cage, 10 cm in diameter) placed between and equidisatant (1.7 m) from
each male. The singing males were recorded using two uni-directional dynamic micro-
phones (GC Electronics, #30-2374), each placed approximately 6 cm from the cage of a
singing male, and a Sony TC-6300 2-channel tape recorder. The temporal parameters
of each male's song and the phase relationships of pairs of acoustically interacting males
were determined using a Commodore 128 computer in conjunction with a computer
interface and software designed for this experiment.
After completing the 10-min recordings, both males and female were released by
opening their cages. In order to obtain spermatophores, copulations were terminated
by the observer as soon as the male's spermatophore was visible. Fifteen of the 24 males

September, 1991


Galliart & Shaw: Mating Success in Katydid Males 457

continued to emit the whole spermatophore after separation from the female. When this
happened the spermatophore was removed and weighed. Males and females were
weighed following the end of the trial. When a male emitted a spermatophore, the
spermatophore weight was included in determining each male's weight following initia-
tion of copulation. In order to increase the sample sizes for determining the relationship
between male body weight and spermatophore weight, spermatophores were collected
from 26 other males placed with females already utilized in two-choice discrimination

Data Analysis

Male mating success was examined in relation to male weight, sound level (8 kHz
and 16 kHz), sound frequency, temporal sound parameters, phase relation of the sound
phrases of the two males during chorusing, and number of physical interactions. Temporal
parameters included number of phrases, mean phrase length, mean phrase interval,
mean phrase period (phrase length + phrase interval) and total sound produced (number
of phrases x mean phrase length). Phrase phase relations were determined by recording
the number of phrase overlaps (the number of times the beginnings of a katydid's phrases
overlapped the ends of the phrases of the other katydid [Fig. 1] ), mean phrase overlap
time (the mean length of time that the phrase beginnings of both katydids overlapped
the phrase endings of their chorusing partner's phrase endings; this does not include
phrases that were not overlapped [Fig. 1] ) and total overlap time (number of overlaps
x mean overlap time). Physical interactions measured included number of male mountings
and mounts by other male, individual male mountings by the female, males kicking or
being kicked by the other male, male crawling on or being crawled on by the other male,
separating other male and female by backing into (courting) or crawling over them,
males jumping away, and males walking away. Temporal sound parameters were
examined for two conditions (treatments): chorusing with and without a ticking female
in the acoustic chamber.
Weight, sound level and number of physical interaction differences between males
that achieved copulation (successful males = SSs) and those that did not (unsuccessful
males = Uds) were analyzed by paired-comparison Student's t-tests (SS US). A
split-plot analysis, with male mating success (Sd or Ud) as the whole plot treatment
and female absence or presence as the split-plot treatment, was used to analyze the
temporal sound parameters and phrase phase relationships. Since temporal sound
parameters and phrase phase relationships were measured for each male in the absence
and presence of a female, the split-plot factor in this analysis was regarded as a repeated
measure. This ANOVA enabled us to examine male mating success regardless of the
presence of the female, the effect of the presence of the female on both males as well
as the effect of the female on S d and U separately. Based on the results of previous
studies (see Introduction), we hypothesized that females would mate with heavier males,
louder males, males with lower song frequencies, and males that produced more sound
energy either in more or longer phrases and/or total sound produced. Pearson's correla-
tion coefficients were determined for the parameters implicated in mate choice. The
peak frequencies of winner and loser sounds were compared using chi-square tests.


Females mated more often with males that produced louder sounds before the trials
and that were heavier following the trials (Table 1). Of special interest, there was a
significant interaction between male mating success and female presence for the number
of times a male's phrases were overlapped (F = 8.98. df = 1, 44; p < 0.0045). Eventual

Florida Entomologist 74(3)


Difference Means
Parameter S-U S.E.a S US N t pC
Weight (g) 0.04 0.02 0.73 0.69 24 1.77 0.05
SI-8kHb (dB) 1.73 0.97 71.73 70.00 24 1.79 0.04
SI-16kHz (dB) 1.63 0.83 73.27 71.65 24 1.97 0.03
aS.E. = standard error
bSI-8kH,--sound intensity at 8 kHz filter band; SI-16kHz-sound intensity at 16 kHz filter band
0p-probability that the mean (S-U) difference equals zero using one tailed, paired-comparison Student's t-test

Sds overlapped Uds more when a female was absent but the situation was reversed
when a ticking female was present (Fig. 2). Paired-comparison t-tests indicated that
differences between males approached a p-level of < 0.05 under both conditions (female
absent: t = -1.94, p = 0.070, two-tailed test; female present: t = 2.02, p = 0.055,
two-tailed test). With a female absent and omitting ties, Sd overlapped Ucs more in
14 of 21 trials; in contrast, Sds overlapped Uds less in 14 of 21 trials with a female
present. The presence of a female resulted in a reduction in total number of phrases
overlapped in 15 of 21 trials and this was accompanied by Sds overlapping U s less in












Fig. 2. Comparison of the mean number of times each of two chorusing males over-
lapped his chorusing partner when a sexually receptive ("ticking") female was absent
and present. S Male-male that eventually mated with female; U Male-male that did
not mate with female. N = 24.


September, 1991

Galliart & Shaw: Mating Success in Katydid Males

18 of 22 trials after the female was added. In contrast, Sds were overlapped more in
13 of 23 trials after the female was added.
The repeated-measures ANOVA also indicated that the presence of a female increased
the length at which the phrase of each katydid overlapped the phrases of the other
katydid (female absent [x S.E.]: 1.27 0.71 s; female present: 1.83 0.74 s; F =
21.35; p = 0.0001).
Pearson's correlation coefficients were run using the three parametres most likely
involved in determining mating success, i.e., male weight, sound level and phrase overlap
number in the presence of a female. From the six paired comparisons, there were four
correlations with a p < 0.05 and one with p < 0.10 (Table 2).
Of all the physical interactions recorded, only one was significantly higher for Sds;
females mounted eventual Sds more than U(s (Sd: x S.E. = 8.63 1.26; Ud:
5.43 0.94; t = 3.02; p = 0.006, two-tailed test).
Unlike Latimer and Schatral's (1986) and Latimer and Sippel's (1987) findings for
Tettigonia cantans, we found no relationship between frequency of male sounds and
whether or not they achieved copulation. Peak frequencies were quite variable among
males (Shaw et al. 1990) and Sd s did not produce lower or higher peak frequencies than
Uds (first peak frequency: chi-square = 0.17, p > 0.50, df = 1, N = 24; higher peak
frequency; chi-square = 0.05; p < 0.80, df = 1, N = 18).
A. parvipennis males produce large spermatophores (x S.E. = 95.20 4.83 mg,
N = 41) which comprise an average of 12% (range = 10-20%) of the males' body weights
(ix S.E. = 768.30 9.00 mg, N = 41). Following spermatophore extrusion, males
do not sing for 1-5 days. Male weight is correlated with spermatophore size (r = 0.454,
P < 0.005) (Fig. 3).


As has been determined for some other species of sound-producing Orthoptera (see
Introduction), A. parvipennis males successful in achieving copulation were heavier and
produced louder sounds. Spermatophore weight was correlated with body weight and
proteins in spermatophores are known to enhance the fitness of female Orthoptera
(Gwynne 1988). Females took hours to choose between competing males usually mounting
each male a number of times and the number of mountings was correlated with mate
choice. This would be an opportune time to determine male size.
In two-choice discrimination tests utilizing artificially-generated sounds, Latimer &
Sippel (1987) found that differences in sound frequency were more important than sound


8kHz 16kHz OVN-fd

Weight 0.34a 0.35 -0.20
0.019b 0.016 0.172
SI-8kHz 0.77 -0.31
0.0001 0.035
SI-16kHz 0.073

"Pearson's correlation coefficients
bp values
cSI--as in Table 1
dOVN-f-number of times Sd overlaps end of phrases of Ud in presence of a ticking female

460 Florida Entomologist 74(3) September, 1991


*0) -

o .10 * *

* *. *
2 0.06 *
c 0.02

0.7 0.8 0.9
Male Weight (g)

Fig. 3. The relationship between spermatophore weight and male weight.

level in attracting females of the katydid Tettigonia cantans. Latimer & Sippel argue
that, because of variation in sound level caused by differences in vegetation density,
male body orientation, etc., frequency differences would be more effective in communicat-
ing male status. The failure to find any correlation between sound frequency and mating
success for A. parvipennis may be related to the differences in mating systems between
T. cantans and A. parvipennis. T. cantans females move toward a particular singing
male; the mating success of A. parvipennis males may be determined after they have
moved to a female.
An especially interesting finding of this study was that females preferentially mated
with males who reduced the number of times they overlapped the phrase beginnings of
their competitors in the presence of a ticking female. If eventual S d s overlapped eventual
U s more in the absence of a female, the situation was usually reversed when a ticking
female was perceived. A closer examination of the data indicated the total number of
phrases overlapped for both katydids decreased in the presence of a female and that
the major reason for this was a reduction in the number of times the SC overlapped
the Ud in 18 of 22 trials. This evidence suggests that Sds adjusted their phrase rates
in order to avoid overlapping the ends of the eventual U s' phrases and thereby,
probably because of limitations on Uds' ability to adjust phrase rate, forced the Uds
to overlap Sds phrases more. In fact, the mean length of phrase overlap actually in-
creased in the presence of the female. These data suggest that the ability of a male to
adjust its phrase rate so that he can initiate his phrase following the end of the other
male's phrase, communicates, either to competing males or to the female, that he is the
superior male. Greenfield (in press) suggests that the nature of chorusing is the evolutio-
nary result of males adjusting their phrase rate so that the loudest portion of their song
is free of jamming, i.e., not overlapped by the song of another katydid. Generalizing
from Greenfield's hypothesis, it is the portion of the phrase carrying the most information
relating to male-male competition or female choice that males should attempt to free
from jamming by the sounds of other males.

Galliart & Shaw: Mating Success in Katydid Males

The most rapidly produced and loudest phonatomes of A. parvipennis occur during
the middle of the phrase and this is the portion of the phrase usually free of overlap
during acoustic interaction of pairs of males. During the first few phonatomes of the
phrase, phonatome rate and sound level increased; during the last few phonatomes,
phonatome rate decreased (Fig. 1). Unlapped middle portions of phrases may be impor-
tant in male-male communication (e.g., in maintaining uniform spacing between adjacent
males; Shaw et al., 1981) and in male-female communication, (e.g., in attracting con-
specific females to the vicinity of one or more males prior to the female beginning to
tick). However, in the field, chorusing may involve more than one male and females
may perceive considerably more overlap between songs of neighboring males. If
phonatome rate enables females to identify conspecific males and release phonotaxis,
then selection would favor synchrony. This rhythm preserving model of synchrony
function has been proposed for sound-producing insects (Walker 1969, Alexander 1975,
Otte 1977, Greenfield & Shaw 1983) and light-flashing fireflies (Lloyd 1973a,b, Otte &
Smiley 1977, Otte 1980, Buck 1988).
Because, like firefly females, phaneropterine females characteristically respond to
male signals at species-identifying intervals (Spooner 1968, Heller and von Helverson
1986) and because A. parvipennis is unique in that females produce short sounds (=
ticks) that may fall between the latter phonatomes of a male's phrase (Shaw et al., 1990),
males may synchronize phonatomes in order to detect a female, to recognize her as a
conspecific (female detection and delay recognition models of Otte & Smiley [1977] ),
and to better perceive the location of the female. If synchrony enables males to perceive
and identify the species and location of conspecific, sexually receptive females, then why
the apparent adjustment of a male's phrase rate to avoid overlapping the end of another
male's phrases in the presence of a ticking female? This also may enhance the perception
of female ticks. Unlike fireflies, synchronizing male Orthoptera may be unable to detect
the signals of conspecifics while simultaneously producing signals (see discussion by
Greenfield, in press). Males may attempt to avoid overlapping an adjacent male to better
detect the location of a ticking female. Although males typically synchronize phonatomes,
the rate of the last few phonatomes slows rapidly and the phonatomes tend to be out
of phase with the other male's phonatomes (see Fig. 1). This could result in overlapping
(jamming) of male phonatomes and the ticks of a nearby female. Unjammed female ticks
may be especially important to males who, as in this experiment, have not yet moved
to the location of the ticking female. It also is possible that there is information in the
rapidly changing initial portion of a male's phrase that relates to his ability to compete
for females. If this is true, then using Greenfield's (in press) logic, males should compete
to avoid overlaps.
Although the correlation coefficients are not very large, most of the correlations
between parameters implicated in mate choice are significant at p 0.05 or close to it
(Table 2). This indicates that competitor males and/or females can use one or a combina-
tion of these cues in determining a "superior" male. Male size or weight could be deter-
mined during male-male or male-female physical interactions. Males continue to chorus
as they move toward the female and when in proximity to the female or other male.
Future experiments will examine what is happening to sound level, temporal parameters
and phrase phase relationships when males are in proximity.
As with many studies of sexual selection, it is impossible to state whether mating
is the result of male competition or female choice, and it is very possible that both
processes are involved (Halliday 1978, Burk 1983). In 10 of the 24 trials, one male did
not encounter the female, i.e., he either did not leave the opened cage or left the vicinity
of the female after encountering the other male acoustically and physically. These data
suggest mating success resulting from male competition. As indicated above, future
experiments will examine what is happening to male acoustic parameters after release
from the cages.

Florida Entomologist 74(3)

The long period of time that females spent interacting with one or both males suggests
that females are in a process of actively making a choice. Females spend up to 10 min
on the back of some males while continually pulling their abdomens away from the male
and, not infrequently, they separated and resumed ticking in response to the singing
of the competitor male. All of this is difficult to comprehend if mating success is strictly
the result of male competition.
This type of observational/correlative study raises many questions and provides few
answers. Are all of these or any of these factors crucial to mating success? Are these
the important factors or are they simply correlated with the yet undiscovered crucial
factors? We are in the process of designing experiments to explore the role of these
factors, separately and in various combinations, in mate mating success and female mate


We wish to thank Dr. Michael Greenfield, Department of Biology, UCLA, Dr. Darryl
Gwynne, Department of Zoology, Erindale Campus, University of Toronto, and anony-
mous reviewers for critiquing drafts of this manuscript. Drs. David Cox and Paul Hinz,
Department of Statistics, Iowa State University served as consultants for the statistical
analysis. A special thanks to Mike Reilly, a student in the Department of Electrical
Engineering, Iowa State University, for designing the computer program and interface
used in our data analysis. This work was partially funded by a grant from the Iowa
Science Foundation.


ALEXANDER, R. D. 1960. Sound communication in Orthoptera and Cicadidae, pp.
38-72 in W. E. Lanyon and W. N. Tavolga, (eds.), Animal Sounds and Communi-
cation. AIBS, Washington, D.C.
ALEXANDER, R. D. 1967. Acoustical communication in arthropods. Ann. Rev. En-
tomol. 12: 495-526.
ALEXANDER, R. D. 1975. Natural selection and specialized chorusing behavior in
acoustical insects, pp. 35-77 in D. Pimentel (ed.), Insects, Science and Society.
Academic Press, N.Y.
BAILEY, W. J. 1985. Acoustic cues for female choice in bushcrickets (Tettigoniidae),
in K. Kalmring and N. Elsner (eds.), Acoustic and vibrational communication in
insects. Paul Parey, Berlin.
BAILEY, W. J., AND D. R. THIELE. 1983. Male spacing behavior in the Tettigoniidae:
an experimental approach, pp. 163-184 in Orthopteran mating systems: sexual
selection in a diverse group of insects. Westview press, Boulder, CO.
BUCK, J. 1988. Synchronous rhythmic flashing of fireflies. II. Quart. Rev. Biol. 63:
BURK, T. 1983. Male aggression and female choice in a field cricket (Teleogryllus
oceanicus); the importance of courtship song, pp. 97-119 in D. T. Gwynne and
G. K. Morris (eds.), Orthopteran mating systems: sexual selection in a diverse
group of insects. Westview Press, Boulder, CO.
BUSNEL, M. C. 1967. Rivalit6 acoustique et hi6rarchie chez 1' Ephippiger (Insecte,
Orthoptere, Tettigoniidae). Z. vergl. Physiol. 54: 232-245.
CADE, W. 1976. Male reproductive competition and sexual selection in the field cricket,
Gryllus integer. Ph.D. Thesis, U. Texas, Austin, TX.
CADE, W. 1979. The evolution of alternative male reproductive strategies, pp. 343-380
in M. S. Blum and N. A. Blum (eds.), Sexual selection and reproductive compe-
tition in insects. Academic Press, N.Y.


September, 1991

Galliart & Shaw: Mating Success in Katydid Males 463

CAMPBELL, D. J. AND D. J. CLARKE. 1971. Nearest neighbor tests of significance
for nonrandomness in the spatial distribution of singing crickets, Telegryllus
commodus (Walker). Anim. Behav. 19: 750-756.
CAMPBELL, D. J., AND E. SHIPP. 1979. Regulation of spatial pattern in populations
of the field cricket Telegryllus commodus (Walker). Z. Tierpsychol. 51: 260-268.
DUMORTIER, B. 1963. Ethological and physiological study of sound emissions in Ar-
thropda, pp. 583-654 in R. G. Busnel (ed.), Acoustic behaviour of animals. Elsevier
Publishing Co., Amsterdam.
FORREST, T. G. 1980. Phonotaxis in mole crickets: its reproductive significance. Fla.
Entomol. 63: 45-53.
FORREST, T. G. 1983. Calling songs and mate choice in mole crickets, pp. 185-204 in
D. T. Gwynne and G. K. Morris (eds.), Orthopteran mating systems: sexual
competition in a diverse group of insects. Westview Press, Boulder, CO.
GREENFIELD, M. D., in press. Evolution of acoustic communication in the genus
Neoconocephalus: discontinuous songs, synchrony, and interspecific interactions,
in W. J. Bailey and D. C. Rentz (eds.), Biology of the Tettigoniidae. E. J. Brill,
Publ., Leiden.
GREENFIELD, M. D., AND K. C. SHAW. 1983. Adaptive significance of chorusing
with special reference to the Orthoptera, pp. 1-27 in D. T. Gwynne and G. K.
Morris (eds.), Orthopteran mating systems: sexual competition in a diverse group
of insects. Westview Press, Boulder, CO.
GWYNNE, D. T. 1982. Mate selection by female katydids (Orthoptera: Tettigoniidae,
Conocephalus nigropleurum). Anim. Behav. 30: 734-738.
GWYNNE, D. T. 1983. Male nutritional investment and the evolution of sexual differ-
ences in Tettigoniidae and other Orthoptera, pp. 337-366 in D. T. Gwynne and
G. K. Morris (eds.), Orthopteran mating systems: sexual competition in a diverse
group of insects. Westview Press, Boulder, CO.
GWYNNE, D. T. 1988. Courtship feeding and the fitness of female katydids (Orthoptera:
Tettigoniidae). Evol. 42: 545-555.
HALLIDAY, T. R. 1978. Sexual selection and mate choice, pp. 18-213 in J. R. Krebs
and N. B. Davies, (eds.), Behavioral ecology: an evolutionary approach. Blackwell
Sci. Publ., Oxford.
HELLER, K. G. AND D. VON HELVERSEN. 1986. Acoustic communication in phanerop-
terid bushcrickets: species-specific delay of female stridulatory response and
matching male sensory time window. Behav. Ecol. Sociobiol. 18: 189-198.
LATIMER, W., AND A. SCHATRAL. 1986. Informational cues used in male competition
by Tettigonia cantans. Anim. Behav. 34: 162-168.
LATIMER, W., AND M. SIPPEL. 1987. Acoustic cues for female choice and male com-
petition in Tettigonia cantans. Anim. Behav. 35: 887-900.
LLOYD, J. E. 1973a. Model for the mating protocol of synchronously flashing firelifes.
Nature 245: 268-270.
LLOYD, J. E. 1973b. Fireflies of Melanesia: bioluminescence, mating behavior, and
synchronous flashing (Coleoptera: Lampyridae). Environ. Entomol. 2: 991-1008.
MEIXNER, A. J., AND K. C. SHAW. 1986. Acoustic and associated behavior of the
coneheaded katydid, Neoconocephalus nebrascensis (Orthoptera: Tettiogniidae).
Ann. Entomol. Soc. Amer. 79: 554-565.
MORRIS, G. K. 1972. Phonotaxis of male meadow grasshoppers (Orthoptera: Tet-
tigoniidae). J. N. Y. Entomol. Soc. 80: 5-6.
OTTE, D. 1977. Communication in Orthoptera, pp.334-361 in T. A. Sebeok (ed.), How
animals communicate. Indiana U Press, Bloomington.
OTrE, D. 1980. On theories of flash synchronization in fireflies. Amer. Natural. 116:

464 Florida Entomologist 74(3) September, 1991

OTTE, D., AND J. SMILEY. 1977. Synchrony in Texas fireflies with a consideration of
male interaction models. Biol. Behav. 2: 143-158.
SHAW, K. C. 1968. An analysis of the phonoresponse of males of the true katydid,
Pterophylla camellifolia (Fabricus) (Orthoptera: Tettigoniidae). Behaviour 31:
SHAW, K. C., R. C. NORTH, AND A. J. MEIXNER. 1981. Movement and spacing of
singing Amblycorypha parvipennis males. Ann. Entomol. Soc. Amer. 74: 436-
SHAW, K. C., R. J. BITZER, AND R. C. NORTH. 1982. Spacing and movement of
Neoconocephalus ensiger males (Conocooephalinae: Tettigoniidae). J. Kansas En-
tomol. Soc. 55: 581-592.
SHAW, K. C., P. L. GALLIART, AND B. SMITH. 1990. The acoustic behavior of
Amblycorypha parvipennis (Orthoptera: Tettigoniidae). Ann. Entomol. Soc.
Amer. 83: 617-625.
SIMMONS, L. W. 1986a. Inter-male competition and mating success in the field cricket,
Gryllus bimaculatus (deGeer). Anim. Behav. 34: 567-569.
SIMMONS, L. W. 1986b. Female choice in the field cricket, Gryllus bimaculatus (de-
Geer). Anim. Behav. 34: 1463-1470.
SPOONER, J. D. 1968. Pair-forming acoustic systems of phaneropterine katydids (Or-
thoptera: Tettigoniidae). Anim. Behav. 16: 197-212.
THIELE, D. R., AND W. J. BAILEY. 1980. The function of sound in male spacing
behaviour in bushcrickets (Tettigoniidae: Orthoptera). Aust. J. Ecol. 5: 275-286.
WALKER, T. J. 1969. Acoustic synchrony: two mechanisms in the snowy tree cricket.
Science 166: 891-894.
WALKER, T. J., AND D. DEW. 1972. Wing movements of calling katydids: fiddling
finesse. Science 178: 174-176.
WELLS, K. D. 1977. The social behavior of anuran amphibians. Anim. Behav. 25:

Scientific Note



'Universidad Sim6n Bolivar
Departamento de Biologia de Organismos
Apartado 89000, Sartenejas, Miranda 1080-A
2Fundaci6n Terramar
Apartado 89000, Sartenejas, Miranda 1080-A

Paper wasps usually construct nests beneath horizontal substrates (Hermann et al.
1985, Krispyn & Hermann 1977) and this is particularly true for wasps of the genus
Mischocyttarus. Rarely do wasps construct nests on vertical substrates of a short-lived
nature such as a plant leaf. Bernon (1969) reports a nest of Polistes fuscatus pallipes
Lepeletier in the inside of a tubular leaf of Sarracenia purpurea L. (Sarraceniaceae) as
unique after reviewing all the associated insects of this plant.
During an expedition to the summit of Auyan Tepui (1,800 m.), Estado Bolivar,
Venezuela, a nest of Mischocyttarus cf. commixtus Richards was found attached to a
leaf within the tubular rosette of a Brocchinia hectioides (Bromeliaceae). The nest was
constructed at 6 cm. below the apex of the vertical leaves and contained 3 eggs, 2 larvae
of different instars, 2 pupae, 1 empty cell, and 1 cell under construction. It also had 3
adults, one of which escaped. After collection, one pupae emerged.
The petiole leaned 300 from the leaf before expanding in the vertically hanging nest
cells. This situation is analogous to observations made by Downing & Jeanne (1988) in
Polistes fuscatus (F.).
The occurrence of a paper wasp nest inside the rosette of the above mentioned
Brocchinia species is also a rare finding since such a construction site does not correspond
to the cues that seem to regulate nest construction behavior in these wasps as stated
by Downing & Jeanne (1988).
Two other nests, apparently also of M. cf. commixtus, were observed constructed
on a stone substrate within small concave depressions of the rock surface. The petioles
also made an angle before the nest started hanging vertically, but we could not measure
them through our binoculars.
The collected nest and wasps have been deposited in the Coleccion Entomologica
"Dr. Francisco FernAndez YWpez" of the Instituto de Zoologia Agricola, Maracay,
We wish to thank Mr. Karlheinz Baumann and Mr. Volker Arzt, who called our
attention to the wasp nests inside the Brocchinia plant.


BERNON, G. L. 1969. Paper wasp nest in pitcher plant, Sarracenia purpurea L.
Entom. News. 80(6): 148.
DOWNING, H. A., AND R. L. JEANNE. 1988. Nest construction by the paper wasp,
Polistes: a test of stigmergy theory. Anim. Behav. 36: 1729-1739.
HERMANN, H. R., J. M. GONZALEZ, AND B. HERMANN. 1985. Mischocyttarus
mexicanus cubicola (Hymenoptera), distribution and nesting plants. Florida En-
tomol. 68(4): 609-615.
KRISPYN, J. W., AND H. R. HERMANN. 1977. The social wasps of Georgia: Hornets,
Yellowjackets, and Polistine paper wasps. U.S.D.A. Agric. Res. Bull. pp. 207-239.

466 Florida Entomologist 74(3) September, 1991


Centro de Desarrollo de Productos Bioticos (CEPROBI-IPN)
Yautepec, Morelos, Mexico

The papaya fruit fly, Toxotrypana curvicauda, has been considered to date a strictly
monophagous species infesting wild and cultivated papaya (Carica papaya L.). Castre-
jon-Ayala (1987) reported infestations in another Caricaceae, Jacaratia mexicana A.D.C.
(=Pileus mexicana) in the State of Morelos, Mexico. Baker et al. (1944) reported that
T. curvicauda also infested a plant locally known as "talayote" or "talayotillo" but never
provided any precise botanical identification.
Here we report T. curvicauda infestations on Gonolobus sorodius A. Gray
(Asclepiadaceae), locally known as "pancolote." We collected 32 larvae of T. curvicauda
infesting 6 fruit of G. sorodius during October, 1990 in Yautepec (Barranca Honda, San
Isidro), Morelos, Mexico. Larvae, pupae and adults emerging from the larvae were sent
to Vicente Hernandez (Instituto de Ecologia, A.C., Xalapa, Veracruz, Mexico) for
taxonomic identification. Voucher specimens have been placed in the permanent insect
collections of the Instituto de Ecologia, A.C. and the Centro de Desarrollo de Productos
Bioticos del Instituto Politecnico Nacional. Taxonomic identification of G. sorodius was
carried out by the Dept. of Botany at the Universidad de Morelos in Cuernavaca, Mexico.
Our findings reported here confirm the report by Castrejon (1987), who also recovered
T. curvicauda (not confirmed at the time) from G. sorodius fruit.
Based on our findings, T. curvicauda should not be considered henceforth a
monophagous insect.
We thank Martin Aluja for reviewing and translating this note and Vicente Hernan-
dez-Ortiz for identifying the T. curvicauda specimens reported here.


BAKER, A. C., STONE, W. E., PLUMMER, C. C. AND MCPHAIL, M. 1944. A review
of studies on the Mexican fruitfly and related Mexican species. USDA Misc. Publ.
CASTREJON-AYALA, F. 1987. Aspectos de la biologia y hibitos de Toxotrypana cur-
vicauda Gerst. (Diptera: Tephritidae) en condiciones de laboratorio y su distribu-
ci6n en una plantaci6n de Carica papaya L. en Yautepec, Mor. BS Thesis, Instituto
Politecnico Nacional, Mexico D.F., Mexico. 88 p.

Scientific Note



Departamento de Biologia, Instituto de Biociencias
Universidade de S. Paulo, C. Postal 11.461
05508 S. Paulo, S. Paulo, Brazil
Gainesville, Florida 32604

McPhail traps have been used throughout the new world for many decades to detect
and monitor for fruit flies of genus Anastrepha. Since no sexual lures or host attractant
have been developed for Anastrepha spp., a food lure containing protein hydrolysate,
torula yeast and carbohydrate has been used for bait in McPhail traps (Lopez & Hernan-
dez Bacerril 1967, Lopez et al. 1971, Malavasi & Morgante 1981). Many modifications
in the original design of McPhail have been proposed in recent years but none have
substantially increased effectiveness. The traditional glass McPhail has two problems.
First, it is expensive when compared with polyethylene made traps, and second, it is
fragile, heavy and difficult to handle. Its lifetime is short compared with other models.
Under extreme conditions, for example when the temperature reaches over 400C inside
the trap, spontaneous breakage is common.
In regions where large Anastrepha spp. monitoring and detection programs have
been conducted for certification of fruit fly-free areas, the large number of traps required
is a major expense. In an attempt to replace the original glass McPhail trap with a more
economical plastic model of identical shape and similar size, we tested the efficiency of
both models in trapping the South American fruit fly, Anastrepha fraterculus
The experiments were conducted in a non-commercial grove, where previous studies
had determined large populations of A. fraterculus were found (Malavasi & Morgante
1980, Amaral 1987). The experiment was conducted in Itaquera, 30 Km east Sao Paulo
city where there are more than 15 species of Anastrepha host trees. Traps were sus-
pended in guava (Psidium guajava), Surinam cherry (Eugenia uniflora) and loquat
(Eryobotrya japonica) during 8 consecutive weeks in March-May 1989.
Two traps models were tested. The traditional McPhail glass trap was made in
Mexico, and the plastic model made in Brazil. Both were similar in shape, the glass trap
with 19 cm diameter and the plastic trap with 17 cm diameter. Both traps were the
same height (14 cm) and had a similar entrance hole (glass 4.5 cm, plastic 4.0 cm
diameter). All traps were baited with 250 ml of 3% protein hydrolysate solution with
borax added to prevent decomposition. Two traps (one of each model) were placed in
opposite quadrants of each tree. Traps were hung at 1.7 to 2.0 m high, in the peripheral
canopy with a minimum of 2.0 m between each trap. Trap positions were alternated,
the bait was changed and the adults were collected and counted weekly.
Ten sets of the two traps were hung up in 5 loquat, 3 guava, 1 grumixama and 1
Surinam cherry trees. Statistical comparisons of means were made with t tests, and by
analysis of variance (ANOVA) (SAS Institute 1987).
A total of 7,148 A. fraterculus flies were caught in the grove. No statistical difference
between glass and plastic traps was observed using t test analysis (F = 1.18; df = 15;
p > 0.7450) (Table 1). There were no significant differences in the sex ratios of the trap
catches between trap models. More females were captured than males. Females may
require more frequent feeding than males and so may be captured in greater numbers
in food-baited traps (Davis et al. 1984).

Florida Entomologist 74(3)

September, 1991


Total catch Mean + SE flies captured/trap/weeki
type S 9 total S Y Total

glass 1335 1941 3296 16.9 12.6a 24.3 + 16.6a 41.2 28.9a
plastic 1687 2165 3852 21.115.3a 27.117.1a 48.231.8a
'Means in the same column followed by the same letter are not significantly different at the 5% level by paired t test.

There were sharp variations in weekly captures related with host availability, climatic
conditions and generation time. The range was 102 to 845 A. fraterculus for glass traps
and 98 to 915 for plastic traps per week. There was no significant difference between
mean number of flies captured in each trap model throughout all weeks, according to an
analysis of variance (F = 2.83; df = 3; p > 0.0305). The importance of weekly variation
in trap catch affecting the results of trap efficiency tests was pointed out by Mason &
Baranowski (1989).
Nakagawa et al. (1975) designed a trap made from a plastic tube with 8 lateral holes
as a substitute for the McPhail trap in Hawaii in monitoring Oriental and melon fruit
fly. However, that trap, as well as several other trap designs, was significantly less
effective in capturing Caribbean fruit fly, Anastrepha suspense in Florida (Witherell
1982). In Brazil, small scale experiments for A. fraterculus have been conducted using
different designs and kinds of material (Amaral unpublished data). However, when
compared with standard McPhail trap, most are less effective in capturingA.fraterculus.
Our results have demonstrated that the plastic trap is as effective as glass in capturing
wild A. fraterculus. When cost is an important factor as in large detection programs,
the use of plastic McPhail traps will be more economical. Studies on additional fruit fly
species should be done to determine the effectiveness of plastic traps in other situations.
The technical assistance of Lourivaldo dos S. Pereira is greatly appreciated. We
thank John Sivinski (IABBBRL, ARS, USDA, Gainesville, Florida) for reviewing an
earlier draft of this manuscript; Joao Morgante (Dep. Biologia, USP) for his suggestions
and Jose Claudio Giusti for permitting the use of his orchard. This study was supported
by grants from FAPESP to MDB and from CNPq to PMA and AM.


AMARAL, P. M. 1987. Desenvolvimento de armadilhas para o monitoramento e control
de tefritideos (Diptera). MSc Thesis, University of Sao Paulo, S. Paulo, Brazil,
125 pp.
DAVIS, J. C., H. R. AGEE, AND D. L. CHAMBERS. 1984. Trap features that promote
capture of the Caribbean fruit fly. J. Agric. Entomol. 1: 236-248.
LOPEZ-D, F., L. F. STEINER, AND F. R. HOLBROOK. 1971. A new yeast hydrolysate-
borax bait for trapping the Caribbean fruit fly. J. Econ. Entomol. 64: 1541-1543.
MALAVASI, A., AND J. S. MORGANTE. 1981. Adult and larval population fluctuation
of Anastrepha fraterculus and its relationship to host availability. Environ. En-
tomol. 10: 275-278.
MASON, L. J., AND R. M. BARANOWSKI. 1989. Response of Caribbean fruit fly (Dipt-
era: Tephritidae) to modified McPhail and Jackson traps: effects of trapping du-
ration and population density. J. Econ. Entomol. 82: 139-142.
NAKAGAWA, S., D. SUDA, AND E. J. HARRIS. 1975. Gallon plastic tub a substitute
for the McPhail trap. J. Econ. Entomol. 68: 405-406.
SAS INSTITUTE. 1987. SAS user's guide. Statistics. SAS Institute, Cary, N.C.
WITHERELL, P. C. 1982. Efficacy of two types of survey traps for Caribbean fruit
fly, Anastrepha suspense (Loew). Florida Entomol. 65: 580-581.


Scientific Note



Institute of Food and Agricultural Sciences
Tropical Research and Education Center
University of Florida, Homestead, Florida 33031

Sweetpotato weevil, Cylas formicarius (Fabricius), is the most limiting factor of
sweet potato, Ipomoea batatas (L.) Lam., production worldwide (Jansson & Raman
1991). Feeding and oviposition damage in roots and vines can cause severe cosmetic and
economic damage. In response to tissue damage, sweet potato roots produce foul tasting
terpenoids (Uritani et al. 1975) rendering them unpalatable for human consumption.
Thus, commercial growers can tolerate very little sweetpotato weevil damage.
Historically, weevil management has relied heavily on cultural and chemical control
(Sutherland 1986, Chalfant et al. 1990). However, chemical control provides little protec-
tion once an egg is laid because immatures develop within roots and vines (Chalfant et
al. 1990). Cultural controls, such as sanitation, crop rotation, planting away from weevil-
infested fields, removal of alternate hosts, hilling plants, etc., can help to reduce damage
(Talekar 1991); however, they require considerable labor and have not been universally
adopted. Recent studies indicate that entomopathogenic nematodes and sex pheromones
may provide new means to control the sweetpotato weevil (Jansson 1991, Jansson et al.
Pheromones have typically been used three ways: (1) monitoring insect populations
using pheromone-baited traps; (2) mass-trapping, where large numbers of traps are
used to reduce the insect population; and (3) mating disruption, in which the pheromone
is used at high dosages to permeate the atmosphere so as to disrupt communication
between the sexes, thus reducing mating (Kydonieus & Beroza 1982, Campion 1984).
Mating disruption has been examined in many insect systems (McLaughlin et al. 1972,
Shorey et al. 1972, Landolt et al. 1982, Sower 1982, Schwalbe & Mastro 1988, for reviews
see Kydonieus & Beroza 1982, Campion 1984), and in general, some success has been
achieved. The objective of our study was to examine the potential of sweetpotato weevil
synthetic sex pheromone as a mating disruptant by examining male confusion (i.e.
communication disruption) in a large commercial field plot.


The pheromone used in the field for evaluation of communication disruption was
obtained from the USDA, ARS, Insect Attractants, Behavior, and Basic Biology Re-
search Laboratory in Gainesville, Florida. The synthetic pheromone, (Z)-3-dodecen-l-ol
(E)-2-butenoate, (>99.9% pure), was applied to methylene chloride-extracted rubber
septa (Thomas Scientific, Swedesboro, NJ) as described by Heath et al. (1986).

Field Studies

Communication disruption potential was evaluated by comparing trap counts in two
widely separated sections (0.5 ha each) of a commercial white-fleshed 'Picadito' sweet
potato field (planted 1 June 1989). Sections were separated by an 1.3 ha non-pheromone
treated buffer zone also planted in sweet potatoes. Experimental sections were east-west
of the buffer zone due to the prevailing winds coming out of the east. The eastern section
contained pheromone traps (5 traps per section 25 m apart running north-south)

470 Florida Entomologist 74(3) September, 1991

baited with an extremely low synthetic pheromone dosage (1 ng per trap). These traps
were to simulate calling feral females. This section was compared with the western
section that also contained pheromone traps baited with 1 ng placed identically to those
in the eastern section. In addition to the low dose traps, twelve wooden stakes baited
with a high pheromone dosage (100 ig per stake) to disorient male C. formicarius were
placed (25 m apart and 10 m east and west of the low dose trapline running north-south)
in the western section. Pheromone loaded rubber septa were pinned to the stakes at a
level just above the height of the canopy. A plastic petri-dish was glued in an inverted
position at the top of each stake to protect the rubber septa from rain. Septa on the
stakes and in the traps were replaced every 3 weeks. Trap counts were recorded twice
per week. Total counts were summed each week. The experiment was repeated, in a
widely separated, distinct area of the same commercial field, south of the first replication.
All plots were treated with monthly chemical insecticide applications parathionn (2.25
kg a.i./ha), endosulfan (0.96 kg a.i./ha), and methamidophos (0.56 kg a.i./ha). Data were
analyzed by analysis of variance and t-test analysis of log transformed weevil captured
totals, using SAS statistical analysis package (SAS Institute 1985).
Pheromone trap counts were significantly lower on most dates in the pheromone-
treated plots than the non-pheromone treated plots (P<0.05) in both trials. For example,
trial 2 trap counts in the pheromone-treated section averaged 12 males/trap on week 1,
whereas in the non-treated sections counts averaged over 400 males/trap. In trial 1, the
pheromone-treated section had significantly lower trap counts on 13 of 21 weeks (Fig.
1); while in trial 2, the pheromone-treated section had significantly lower trap counts
11 of 21 weeks (Fig. 1). Trap counts dropped to zero and were not significantly different
after 21 October when a cold front dropped temperatures considerably. These data
suggest that pheromone communication was probably disrupted in pheromone-treated
sections. However, due to the location of the plots in a large commercial field, there
was no isolation from immigration of mated females. Therefore no data were obtained
relevant to weevil control.
It is not known if the lower trap counts in the pheromone-treated sections due to
male disorientation would translate to a reduction in plant damage ratings without
insecticide treatment. Information critical to the assessment of mating disruption studies
includes distances over which insects can disperse for mating; if there is a dispersal
flight after mating; whether virgin females search for males and/or visa versa (Campion
et al. 1989); or, the breadth of the host range (Rothschild 1981).
Insects with a restricted host range may be more limited in their movement than
polyphagous species (Campion 1984). Sweetpotato weevils are oligophagous; consuming
plants primarily in the Convolvulaceae (Austin et al. 1991). If alternative hosts are
removed from areas surrounding areas as suggested by Talekar (1991) as a method of
weevil control, then immigration into fields by gravid females or nondisrupted males,
should be limited. In addition to all host removal, the larger the area treated or the
more isolated the field, the lesser the chance of immigration (Campion 1984). Sweetpotato
weevil males can traverse considerably distances (280 m/16h) when no distractions are
present (food, mates, etc.) (Mason et al. 1990). However, when marked males are
released in a sweetpotato field with a resident sweetpotato weevil population, traversal
distances drop to a maximum of 80 m/40 h in a 3 month old field to less than 60 m/40 h
in a field ready for harvest (6 months old) (Mason & Jansson unpubl. data). Thus,
sweetpotato weevil have limited dispersal abilities, especially as a field ages, and mating
disruption may have potential.
Evidence from this work indicates that mating disruption has potential. Further
behavioral studies to verify a reduction in mating under field conditions as well as testing
without the use of insecticides should confirm its control potential. Current and future
studies are focused on combining mass trapping, mating disruption, and en-

Scientific Note

15 Sept

I Trial 2
500 4-
400 I\

300 \T T

200 +


0 2 Jl 25 Aug 30
28 Jul 25 Aug 30 Sept

Pheromone --
No Pheromone A- A

3 Nov

Fig. 1. Average number of sweetpotato weevil males caught per trap per week in
pheromone traps baited with 1 ng of pheromone placed in pheromone-treated and non-
pheromone treated sections of a commercial sweet potato field. IN = Insecticide appli-
cation; CF = Cold front.

tomopathogenic nematodes for a comprehensive control program for this serious pest
of sweetpotato in the tropics.
We thank T. Albinana, S. Fertig, T. Bailey, R. Lance, and S. Lecrone for their
assistance in data collection; R. Heath and B. Dueben for loading of pheromone; V and
S Growers for the use of their sweetpotato field; J. Pena and E. Mitchell for the helpful
suggestions; and the anonymous reviewers for their constructive suggestions. This re-
search was supported, in part, by the U.S. Department of Agriculture under CRSR
Special Grants 87-CRSR-2-3107 and 88-34135-3571 (to R. K.J.) managed by the Caribbean
Basin Advisory Group (CBAG). This is Florida Agricultural Experiment Station No.


AUSTIN, D. F., R. K. JANSSON, AND G. W. WOLFE. 1991. Convolvulaceae and Cylas:
a proposed hypothesis on the origins of this plant/insect relationship. Trop. Pest
Manage. (In Press).
CAMPION, D. G. 1984. Survey of pheromone uses in pest control, pp. 405-449 in H.
E. Hummel and T. A. Miller [eds], Techniques in pheromone research, Springer-
Verlay, N.Y.





00 ^

Trial 1 Pheromone 0--
No Pheromone A -A



1 -Y-

14 Jul

11 Aug

20 Oct

472 Florida Entomologist 74(3) September, 1991

CAMPION, D. G., B. R. CRITCHLEY, AND L. J. MCVEIGH. 1989. Mating disruption,
pp. 89-119, In A. R. Jutsum and R.F.S. Gordon [eds], Insect pheromones in plant
protection. John Wiley and Sons, N.Y.
CHALFANT, R. B., R. K. JANSSON, D. R. SEAL, AND J. M. SCHALK. 1990. Ecology
and management of sweetpotato insects. Ann Rev. Entomol. 35: 157-180.
J. H. TUMLINSON. 1986. Identification of sex pheromone produced by female
sweetpotato weevils, Cylasformicarius elegantulus (Summers). J. Chem. Ecol.
12: 1489-1503.
JANSSON, R. K. 1991. Biological control of Cylas spp, pp. 169-201 in R. K. Jansson
& K. V. Raman [eds], Sweet potato pest management: a global perspective.
Westview Press, Boulder, Colorado.
JANSSON, R. K., AND K. V. RAMAN. 1991. Sweetpotato pest management: a global
overview, pp. 1-12 in R. K. Jansson and K. V, Raman [eds.], sweetpotato pest
management: a global perspective. Westview Press, Boulder, Colorado.
JANSSON, R. K., L. J. MASON, AND R. R. HEATH. 1991. Development of a sex
pheromone monitoring program for sweetpotato weevil, pp. 97-138, In R. K.
Jansson & K. V. Raman [eds], Sweet potato pest management: a global perspec-
tive. Westview Press, Boulder, Colorado.
KYDONIEUS, A. F., AND M. BEROZA [eds]. 1982. Insect suppression with controlled
released pheromone systems. CRC Press, Boca Raton, Florida, Vol 1, pp 274;
Vol 2, pp 32.
1982. Field trials of potential navel orangeworm mating disruptants. J. Econ.
Entomol. 75: 547-550.
MASON, L. J., R. K. JANSSON, AND R. R. HEATH. 1990. Sampling range of male
sweetpotato weevils (Cylas formicarius elegantulus) (Summers) (Coleoptera:
Curculionidae) to pheromone traps: influence of pheromone dosage and lure age.
J. Chem. Ecol. 16: 2493-2502.
1972. Sex pheromones of Lepidoptera. XXXI. Disruption of sex pheromone
communication in Pectinophora gossypiella with hexalure. Environ. Entomol. 1:
ROTHSCHILD, G.H.L. 1981. Mating disruption of lepidopterous pests: current status
and future prospects, pp 207-228, In E. R. Mitchell [ed], Management of insect
pests with semiochemicals: concepts and practices. Plenum Press, N.Y.
SAS INSTITUTE. 1985. SAS users guide; statistics, version 6 ed. SAS Institute, Cary,
SCHWALBE, C. P., AND V. C. MASTRO. 1988. Gypsy moth mating disruption: dosage
effects. J. Chem. Ecol. 14: 582-588.
pheromones of Lepidoptera. XXX. Disruption of sex pheromone communication
in Trichoplusia ni as a possible means of control. Environ. Entomol. 1: 641-645.
SOWER, L. L. 1982. Douglas-fir tussock moth disruption, pp. 165-174, In A. F.
Kydonieus & M. Beroza [eds], Insect suppression with controlled released
pheromone systems. Vol. 1. CRC Press, Boca Raton, Florida.
SUTHERLAND, J. A. 1986. A review on the biology and control of the sweetpotato
weevil Cylas formicarius (Fab.). Trop. Pest Management 32: 304-315.
TALEKAR, N.. S. 1991. Integrated control of Cylasformicarius, pp. 157-167 in R. K.
Jansson and K. V. Raman [eds], Sweet potato pest management: a global perspec-
tive. Westview Press, Boulder, Colorado.
URITANI, I., T. SAITO, H. HONDA, AND W. K. KIM. 1975. Induction of furano-ter-
penoids in sweet potato roots by the larval components of the sweet potato
weevils. Agric. Biol. Chem. 37: 1857-1862.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs