(An International Journal for the Americas)
Volume 64, No. 4 December, 1981
TABLE OF CONTENTS
BEUGNON, G.-Orientation of Southern Mole Cricket, Scapteriscus
acletus, Landing at a Sound Source ..-...-.................................... 463
STILING, P. D., AND D. R. STRONG-A Leaf Miner (Diptera: Ephy-
dridae) and Its Parasitoids on Spartina alterniflora in North-
west Florida ...----------------------.........._. ---468
WILLIAMS, D. F., AND C. S. LOFRGEN-Eli Lilly EL-468, A New Bait
Toxicant for Control of the Red Imported Fire Ant ...--........-.._. 472
TRYON, E. H., JR., AND S. L. POE-Developmental Rates and Emergence
of Vegetable Leafminer Pupae and Their Parasites Reared from
Celery Foliage .------- ---......------------................. 477
SAKIMURA, K.-A Review of Frankliniella bruneri Watson and De-
scriptions of F. kelliae, n. sp. (Thysanoptera: Thripidae) --------- 483
FRANK, J. H., AND G. A. CURTIs-Bionomics of the Bromeliad-Inhabit-
ing Mosquito Wyeomyia vanduzeei and Its Nursery Plant Til-
landsia utriculata ............------------........ ............... 491
FATZINGER, C. W.-Activity Rhythms, Influence of Host Plant on Mat-
ing and Oviposition, and Rearing of the Southern Pine Cone-
worm (Lepidoptera: Pyralidae) ------------................................... ....... 506
WISEMAN, B. R., N. W. WIDSTROM, AND W. W. MCMILLIAN-Effects of
'Antigua 2D-118' Resistant Corn on Fall Armyworm Feeding
and Survival --- -------.--------.................................. 515
PECK, S. B.-Community Composition and Zoogeography of the In-
vertebrate Cave Fauna of Barbados ...----............................ 519
JOHNSON, D. W., E. R. MITCHELL, J. H. TUMLINSON, AND G. E. ALLEN
-Velvetbean Caterpillar: Response of Males to Virgin Females
and Pheromone in the Laboratory and Field .-........---......---....-- 529
KHALAF, K. T.-Multiparasitism of Puss Caterpillars (Lepidop-
tera: Megalopygidae) by a Wasp and Fly Species ............ 534
HARPER, J. D.-Citrus Pulp Bait Insecticide Formulations for
Control of Velvetbean Caterpillar and Bean Leaf Beetle
on Soybeans -- --- -----.--- ---------....... ..... 538
Continued on Back Cover
Published by The Florida Entomological Society
THE FLORIDA ENTOMOLOGICAL SOCIETY
OFFICERS FOR 1981-82
President .---......-..........----..............------ --.. W. L. Peters
Vice-President ..-----......................... ------------------Abe White
Secretary .... ----........-.........-----------------------.. F. W. Mead
Treasurer ................ ..--------...................-- ------------. D. P. Wojcik
E. C. Beck
W. Keith Collinsworth
Other Members of Executive Committee --.... R. W. Flowers
R. G. Haines
D. C. Herzog
C. A. Musgrave Sutherland
Editor --............------------ --------..----- C. A. Musgrave Sutherland
Associate Editors .....- ---.-.....-.........----- --------- F. W. Howard
J. E. Lloyd
J. R. McLaughlin
C. W. McCoy
A. R. Soponis
H. V. Weems, Jr.
Business Manager ..--..................... ----- -------------- D. P. Wojcik
FLORIDA ENTOMOLOGIST is issued quarterly-March, June, September,
and December. Subscription price to non-members is $15.00 per year in
advance, $5.00 per copy. Membership in the Florida Entomological Society,
including subscription to Florida Entomologist, is $10 per year for regular
membership and $2 per year for students. Inquiries regarding membership
and subscriptions should be addressed to the Business Manager, P. O. Box
12425, University Station, Gainesville, FL 32604. Florida Entomologist is
entered as second class matter at the Post Office in DeLeon Springs and
Authors should consult "Instructions to Authors" on the inside cover of
all recent issues while preparing manuscripts or notes. When submitting a
paper or note to the Editor, please send the original manuscript, original
figures and tables, and 8 copies of the entire paper. Include an abstract and
title in Spanish, if possible. Upon receipt, manuscripts and notes are ac-
knowledged by the Editor and assigned to an appropriate Associate Editor
who will make every effort to recruit peer reviewers not employed by the
same agency or institution as the authors(s). Reviews from individuals
working out-of-state or in nearby countries (e.g. Canada, Mexico, and
others) will be obtained where possible.
Manuscripts and other editorial matter should be sent to the Editor,
C. A. Musgrave Sutherland, 4849 Del Rey Blvd., Las Cruces, NM 88001.
Business matters for other Society officers can be sent to that individual at
the University Station address above.
This issue mailed January 15, 1982
Beugnon: Orientation of Scapteriscus acletus 463
ORIENTATION OF SOUTHERN MOLE CRICKET,
LANDING AT A SOUND SOURCE1
Department of Entomology and Nematology
University of Florida
Gainesville, FL 32611 USA
Mole crickets, Scapteriscus acletus, attracted to a synthetic male calling
song, can use visual cues as they end their flight. The insects landed signif-
icantly (P<0.001) more often on stripes than on a contrasting background.
The visual form of the stimulus was more important in attracting mole
crickets than the contour (perigrammotaxis) or the low reflectance (scoto-
taxis) of the patterns. In windy weather the proportion of crickets landing
on top of the sound source is less.
Los cortones, Scapteriscus acletus, atraidos por la llamada sint6tica del
macho, pueden guiarse visualmente segin terminan su vuelo. Significativa-
mente, los insects aterrizaron (P<0.001) mas frequentemente sobre las
rayas que en el fondo contrastante. La forma visual del estimulo fu6 de m6s
importancia para atraer los cortones que el contorno ("perigrammotaxis")
o el bajo reflejo ("scototaxis") de los patrons. Durante tiempos ventosos, la
proporci6n de cortones que aterrizaron sobre el lugar de origen del sonido fu4
Little is known of visual orientation in mole crickets. Pardi (1975) pro-
posed that the European mole cricket, Gryllotalpa gryllotalpa (L.), used
sun-compass orientation. The European mole cricket and 2 southern United
States species, Scapteriscus vicinus Scudder and S. acletus Rehn and Hebard,
are attracted to UV light traps (Hosney and Khattab 1959, Beck and Skin-
ner 1967, Ulagaraj 1975). At Gainesville, Florida, large flights of S. acletus
occur from April to July ca. 1 hour after sunset (Ulagaraj 1975). This
period of flight is correlated with the male calling period (Forrest 1980).
Adult mole crickets use the calling song of males, singing in their burrow,
to locate possible mates, suitable habitats, or oviposition sites. These dif-
ferent purposes for orientation may cause the insects to end their flights
differently. Average landing distances from the sound source are less for
females than for males (Ulagaraj 1975) and for crickets (males and fe-
males) entering a calling male's burrow than for those that do not (T. G.
Forrest, personal communication). Though orientation of flying mole crickets
is guided by phonotaxis (Ulagaraj and Walker 1973, 1975), the use of visual
cues, another possible means of orientation during landing, has not been in-
vestigated nor have the effects of weather on the accuracy of landing in
'Florida Agricultural Experiment Station Journal Series No. 2803.
'Current address: Laboratoire de Neuro6thologie, Universit4 Paul Sabatier, 31077, Toulouse,
Florida Entomologist 64 (4)
relation to the sound source. Information regarding the visual orientation
of landing mole crickets may provide insight into their basic behavior and
be useful in developing traps for mole crickets landing at simulated male
MATERIAL AND METHODS
All experiments were conducted outdoors from April to June 1980 at
Green Acres Farm, University of Florida, 19 km northwest of Gainesville.
Wind conditions and air temperatures at 40 cm above the ground were re-
corded during each observation. Mole crickets were attracted during their
early evening flight period by broadcasting a constant artificial S. acletus
call (2.7 kHz, 50 pulses/s., 106 dB(A) at 15 cm) from a small electronic
sound unit. Landings were observed on a wooden horizontal circular plat-
form (diameter 1.2 m) placed around the sound source( Fig. 1). To test
for visual preferences, I painted the platform with contrasting alternate
stripes (black or white). Mole crickets landing on top of the sound source
were captured in a cylindrical screen cage (diameter 20 cm) closed above
with a screen funnel (Fig. 1).
Scapteriscus acletus often occurs near the edges of lakes, ponds and
streams (Ulagaraj 1975). Therefore, a wading pool (diameter 1.5 m) divided
into 4 equal parts (2 containing sand and 2 with water), with the sound
unit in the center, was used to study landing preferences on sand versus
Fig. 1. Experimental device for study of landing orientation in mole
crickets attracted to a synthetic calling song. Crickets landing on the hori-
zontal circular wooden board (a) (diameter 1.2m), painted with alternate
black and white stripes, were trapped in the partitioned 19-liter bucket (b).
Metal bands prevented the insects crossing to a different area after landing
and guided them to the corresponding pitfall. Crickets landing on top of the
sound source were captured in the cylindrical screen cage.
Beugnon: Orientation of Scapteriscus acletus
The attractiveness of the different visual stimuli was measured by the
number of insects landing on them. A x2 square test was used to compare
the observed frequency of landing to an expected frequency based on a
radially uniform landing distribution.
Tests for visual orientation at landing.
1. Black stripes against a white background: In this situation (Fig. 2)
6 black stripes, sustaining 200 each, received 40% (348 of 870) of the in-
sects landing on the board. This significant tdeviation from expected (x2 =
17.40, P<0.001) could be due to (a) an orientation toward the low reflecting
black surface (i.e. "scototaxis"; Alverdes 1930); (b) attraction toward the
visual form of stripes standing against a background; (c) attraction toward
the edges of the figures (i.e. "perigrammotaxic orientation"; Michieli 1959).
obs. o. I
exp. P <.001 e. P <.001
exp. e. -
33% 33 %
white edge black
30 35 35 % 54%
33 33 33 % 50%
Fig. 2-5. Landing frequencies. 2) With 6 black stripes (covering 6x200)
on white background (6x400). 3) With 6 white stripes (covering 6x200) on
black background (6x400). 4) With 3 black and 3 white stripes of same
width (covering 6x600). Orientation to edges was evaluated by trapping
mole crickets landing on 6 half black and half white stripes covering 200
each. 5) In a partitioned wading pool (diameter 1.5 m) with 2 liters of sand
and 2 liters of water.
Florida Entomologist 64(4)
To clarify among these assumptions, the following experiments were
2. White stripes against a black background: In this reversed situation
(Fig. 3), 45% of the landings (1134 of 2508) occurred on the white stripes.
This significantly greater than expected number of landings (x2 = 159.34;
P<0.001) on the white refutes the role of a scototaxis in the former experi-
ment and reveals the importance of stripes as an attractive stimulus.
3. No form discrimination available: In this control experiment (Fig. 4),
3 black areas alternated with 3 white areas of the same size (600 each).
Consequently, no visual form could be distinguished from its background.
Edge orientation was estimated by trapping insects landing on the 6
boundaries (200 each) between each black and white area (400 each). Ob-
served landings (: = 616) occurred with similar frequencies on each of the
3 patterns (x2 = 3.65; P>0.05).
These data show the importance of the stripe/background feature of the
visual stimulus in the 2 first experiments. This importance was confirmed
in the sand vs. water situation (Fig. 5) where no significant differences in
the frequency of landings were observed (x2 = 0.90, P>0.05). A slight pref-
erence for white surfaces appeared in all the experiments: 45% for white
stripes against 40% for black ones, Fig. 2, 3; and 35% for white areas
against 30% for black ones, Fig. 4.
Tests for the effects of weather conditions on landing.
Temperatures were similar for all experiments: 18.6-22.20C. The base
area of the cylindrical screen cage over the sound source was 2.78% of total
landing area. With no wind, a significant number of "excess" landings oc-
curred on top of the sound source (86/2224= 3.87% with x2 = 9.77;
P<0.005). With winds 1.6-4.8 kmph, the number of excess mole crickets
that landed in the center was not significant (43/1248 = 3.45% with x2 =
2.07; P>0.05). With winds >4.8 kmph, no mole crickets landed in the center
(0/315; x2 = 8.99; P<0.005).
Landing reactions of some insect species have been studied under lab-
oratory conditions. Expansion of the stimulus in the visual field of the fly
(Goodman 1960) or a decrement of the environmental light level (Taddei
and Fernandez 1973) elicit the landing responses. These authors define the
landing reaction as an all-or-none reaction to a contrasting object that is
approached in flight (Fernandez and Taddei 1975). However, for a given
insect, attractiveness of a visual pattern may be due to different stimulating
characteristics (size, shape, brightness, contrast, etc.). In bright light con-
ditions, a photonegative wood cricket, Nemobius sylvestris, orients toward
a black stripe against a white background by scototaxis (Campan and
Medioni 1963). When in the same experimental conditions, a photopositive
wingless insect, Lepismachilis targionii, orients toward the black stripe
mainly by perigrammotaxis (Goulet et al. 1976). These studies show that
determinants in visual orientation are relative to overall experimental condi-
tions. In dim illumination, as at night, the optimum signal characteristic is
white reflectance (Hailman 1977). Observed percentage of landings in S.
Beugnon: Orientation of Scapteriscus acletus
acletus are greater on white areas than on black areas. This could also be
due at night to UV radiations mainly provided by the skylight and reflected
by white surfaces (Mazokhin-Porshnyakov 1969). Observations of mole
crickets landing on stripes show that they were able to distinguish this
optical signal from its background. They landed significantly more often
(P<0.001) on stripes placed on a contrasting background, whatever the sign
of the contrast (black/white or W/B), without showing any preferences to
the periphery. Study of landing on top of the sound source shows the nega-
tive influence of wind conditions on the accuracy of the landing. Even if the
wind allows perception of the sound source at a greater distance, it may
affect flying insects by increasing energy costs of upstream flights toward
the sound source and by decreasing the ability of an accurate landing on top
of the singer's burrow.
These results clearly demonstrate that, during landing, S. acletus can
display an orientation involving form vision discrimination, and this sug-
gests a need for further visual orientation studies in this and other species.
I thank T. J. Walker for advice and critically reviewing the manuscript.
I appreciate the cooperation and assistance of Tim Forrest throughout the
study. Susan Jungreis and C. O. Calkins generously helped with the manu-
script. William Oldacre, Oldacre Engineering, developed and manufactured
broadcasting units in consultation with T. J. Walker.
ALVERDES, F. 1930. Die lokomotorische Reacktionen von decapoden Kresben
auf Helligkeit und Dunkkelheit. Z. Wiss. Zool. 137: 403-75.
BECK, E. W., AND J. L. SKINNER. 1967. Screening of insecticide against the
change and southern mole cricket attacking seedling millet. J. Econ.
Ent. 60: 1517-9.
CAMPAN, R., AND J. MEDIONI. 1963. Sur le comportement "scototactique" du
grillon des bois, Nemobius sylvestris (Bosc.) C. R. Soc. Biol. 157:
FERNANDEZ PEREZ DE TALENTS, A., AND C. TADDEI-FERRETTI. 1975. Landing
and optomotor responses of the fly Musca. Pages 490-501 in G. A.
Horridge, ed. Compound eye and vision of insects. Academic Press,
FORREST, T. G. 1980. Phonotaxis in mole crickets: its reproductive signif-
icance. Florida Ent. 63(1) : 45-53.
GOODMAN, L. T. 1960. The landing responses of insects. I. The landing re-
sponse of the fly, Lucila sericata and other calliphorinae. J. Exp. Biol.
GOULET, M., R. CAMPAN, AND R. MORVAN. 1976. Preliminary study of
scototactic orientation in a wingless insect, Lepismachilis targiornii
(Grassi). Biol. Behav. 1: 367-80.
HAILMAN, J. P. 1977. Optical signals. Page 362 in Animal communication
and light. Indiana Univ. Press, Bloomington, IN.
HOSNEY, M. M., AND A. A. S. KHATTAB. 1959. The possible effect of ultra-
violet light traps in reducing the population level of mole crickets,
Gryllotalpa gryllotalpa L., in Egypt. Ent. Mon. Mag. 95: 109-10.
MAZOKHIN-PORSHNYAKOV (G. A.). 1969. Insect vision. T. H. Goldsmith,
(ed.). Plenum Press, New York, NY. 306 p.
Florida Entomologist 64 (4)
MICHIELI, S. 1959. Analiza Scotokticnih (Perigramotakticnih) reakcij pri
Arthropodih. Acad. Sci. et arti slovenica, Cl IV: Historia naturalis
et medicine: 237-86.
PARDI, L. 1975. Some observations on orientation in Talorchestia fritzi
(Stebbing), Grapsus grapsus (L.) and Amblyrhynchus cristatus
(Bell.) from the Galapagos Islands. Museum Zool. Univ. Firenze. 19 p.
TADDEI-FERRETI, C., AND A. FERNANDEZ PEREZ DE TALENS. 1973. Landing
reaction of Musca domestic. III. Dependence of the luminous char-
acteristics of the stimulus. Z. Naturforshteil C. Biophys. Biol. Virol.
28(10) : 568-78.
ULAGARAJ, S. M. 1975. Mole crickets: Ecology, behavior, and dispersal flight
(Orthoptera: Gryllotalpidae Scapteriscus). Environ. Ent. 4: 265-73.
AND T. J. WALKER. 1973. Phonotaxis of crickets in flight: attraction
of male and female crickets to male calling songs. Science 182: 1278-9.
AND 1975. Responses of flying mole crickets to three param-
eters of synthetic songs broadcast outdoors. Nature 253: 530-2.
A LEAF MINER (DIPTERA: EPHYDRIDAE) AND ITS
PARASITOIDS ON SPARTINA ALTERNIFLORA
IN NORTHWEST FLORIDA
PETER D. STILING AND DONALD R. STRONG
Department of Biological Science, The Florida State University,
Tallahassee, FL 32306 USA
The shore fly, Hydrellia valida Loew (Diptera: Ephydridae), commonly
mines the leaves of salt marsh cord grass, Spartina alterniflora Loisel, at
Oyster Bay, northwest Florida. Its larvae are frequently parasitized by
Opius sp. (Hymenoptera: Braconidae), and sometimes by Pteromalus sp.
(Hymenoptera: Pteromalidae) and Sympiesis sp. (Hymenoptera: Eulo-
La mosca de la costa, Hydrellia valid Loew (Diptera: Ephydridae), mina
frecuentemente las hojas de Spartina alterniflora, hierba de la cienaga
salada, en Oyster Bay, Florida noroeste. Sus larvas son parasitadas fre-
cuentemente por Opius sp. (Hymenoptera: Braconidae), y a veces por
Pteromalus sp. (Hymenoptera: Pteromalidae) y Sympiesis sp. (Hy-
The shore flies constitute a medium sized family of acalyptrate Diptera.
They exhibit several different modes of life, mostly in aquatic habitats.
Hydrellia Robineau-Desvoidy and Lemnaphila cresson are the only known
genera of leafmining ephydrids (Deonier 1971). This is the first record of
an ephydrid attacking Spartina alterniflora Loisel in Florida. Hydrellia
valida Loew was the only leafminer reared from this host in our intensive
studies of the arthropods of S. alterniflora in Oyster Bay, Wakulla County,
Stiling & Strong: Leaf Miner on Spartina
Eggs appear similar to those depicted for Hydrellia griseola (Fallen)
(Grigarick 1959). They are pale white, cylindrical, elongate and about 0.15
x 0.6 mm. The chorion is sculptured with delicate irregular longitudinal
ridges. Eggs are deposited longitudinally between or on the leaf ribs of the
adaxial surface of the leaf. They are generally laid towards the axil and
central part of the blade of the younger leaves of a stem (Table 1). Although
Grigarick (1959) observed up to 52 eggs of H. griseola per rice blade, we
found a maximum of 17 eggs per Spartina leaf, with a mean of 4.9 4.0
Larvae are white and may reach 4 mm in length. They create straight,
narrow mines (Fig. 1). This is common behavior for Hydrellia larvae mining
grasses (Deonier 1971) in contrast to the tortuous and blotch-like mines of
larvae feeding in broad leaves. Of 111 larvae, 85% were found mining
towards the leaf apex. The upper epidermal layer of the mine rises, similar
to a blister, and the bulge of the larva is easily observed or felt if the leaf
is slid between the fingers. The mean number of larvae per leaf was 1.68,
maximum 4. This is considerably less than the number of eggs per leaf, in-
dicating substantial mortality in either the egg or early larval stages.
Mortality could derive from natural enemies, parasitism, competition be-
tween larvae, or from other sources. However, different mines often merged
with no apparent adverse competitive effect on larvae, and several larvae
were found inhabiting the same mine.
Puparia (Fig. 2) are between 3.5 and 4 mm long, pale brown when
healthy or dark brown when parasitized. The pupal period ranged from 7
to 9 days, with an average of 8.05 (n=18). The maximum number of
puparia observed per leaf was 4, mean 1.4. Three to 5 puparia per leaf is
not uncommon for this genus (Grigarick 1959). Puparia were often located
within the older or more basal leaves of a stem (Table 1). As in most leaf-
mining ephydrids, pupation occurs within the mine. This behavior is dis-
tinctly adaptive on S. alterniflora, an intertidal host that can be twice daily
innundated by the sea (McCoy and Rey unpublished). Larvae pupating on
the ground could easily could be drowned or washed away. Adult flies emerge
through the operculum of the puparium. The operculum initially remains
attached posteriorly because the ecdysis cleaveage line extends only around
the front and side (Deonier 1971).
Hydrellia valida is multivoltine with several sequential generations dur-
TABLE 1. LOCATION OF LIFE STAGES OF Hydrellia valida ON LEAVES OF
Spartina alterniflora, 11 MAY ,1980.
Leaf no., from base to apex
Stage 1 2 3 4 5
Egg 1 23 65 29 0
Larva 49 109 44 4 0
Puparium 81 64 11 0 2
Florida Entomologist 64(4)
Fig. 1-5. 1) Typical mine. 2) fydrelia valida and puparium. 3) Opius
sp. (Hymenoptera: Braconidae). 4) Pteromalus sp. (Hymenoptera: Ptero-
malidae). 5) Sympiesis sp. (Hymenoptera: Eulophidae). Scale lines are
ing spring and early summer. At Oyster Bay, Florida, mines first appear in
mid April, peak in May, and have usually disappeared by the end of June
(Table 2). Mines are generally concentrated in the most shoreward plants,
where over 90% of the plants and 65% of their leaves may be mined. Many
members of the genus Hydrellia are polyphagous (Deonier 1971) and H.
valida may overwinter either as larvae on other hosts that grow further
inland, or as adults.
Of 187 insects reared from Hydrellia larvae or puparia in 1980, 88.3%
were parasites. Parasitism by Opius sp. (Hymenoptera: Braconidae) (Fig.
3) was 78.1%; by Pteromalus sp. (Hymenoptera: Pteromalidae) (Fig. 4)
9.1%; and by Sympiesis sp. (Hymenoptera: Eulophidae) (Fig. 5) 1.1%.
Stiling & Strong: Leaf Miner on Spartina 471
TABLE 2. TEMPORAL VARIATION IN ABUNDANCE AND PARASITISM RATES OF
Hydrellia valida ON SHORELINE Spartina alterniflora AT BOGGY
ISLAND, OYSTER BAY, NORTHWEST FLORIDA, 1980.
Date No. per 20 plants
(day, month) Larvae Pupae % Parasitism
27-IV 14 4 59.5
18-V 45 32 75.0
2-VI 10 12 94.1
9-VI 23 12 95.0
30-VI 3 1 100.0
Braconids commonly exert the greatest degree of parasitism on Hydrellia
(Grigarick 1959). Sympiesis sp. are believed to be external larval feeders
because they pupate free of their host. For Opius sp. and Pteromalus sp.,
only 1 parasite was seen to occupy and emerge from each host puparium.
At emergence, these parasites cut their own exit holes rather than use the
host's puparial operculum. Parasitism by all species was first seen during
the host's larval stages. Throughout most of the spring, parasitism was
about 90%, after a sharp increase in April and early May (Table 2).
Grigarick (1959) reports a similar seasonal increase in parasitism of H.
griseola, from less than 50% in April to nearly 90% in July.
We thank Dr. W. N. Mathis, Department of Entomology, Smithsonian
Institution, and Drs. E. E. Grissell and P. M. Marsh, Systematic En-
tomology Laboratory, United States Department of Agriculture, for identi-
fication of the leafminer and its parasitoids. Parasitoids could only be
identified to genus. We are also grateful to Mr. K. Womble for photo-
graphic expertise and to Jacqui Stiling for typing the manuscript. This
work was supported by National Science Foundation grant DEB 7921828 to
Dr. D. R. Strong and by the Department of Biological Science, Florida State
DEONIER, D. L. 1971. A systematic and ecological study of Nearctic Hydrel-
lia (Diptera: Ephydridae). Smithsonian Contributions to Zoology 68:
GRIGARICK, A. A. 1959. Bionomics of the rice leafminer, Hydrellia griseola
(Fall6n), in California (Diptera: Ephydridae). Hilgardia 29: 1-80.
472 Florida Entomologist 64(4) December, 1981
ELI LILLY EL-468, A NEW BAIT TOXICANT
FOR CONTROL OF THE RED IMPORTED FIRE ANT1,2
DAVID F. WILLIAMS AND CLIFFORD S. LOFGREN
Insects Affecting Man and Animals Research Laboratory,
Agricultural Research Service,
USDA, Gainesville, FL 32604 USA
Eli Lilly EL-468, a phenylenediamine, possesses the delayed-toxicity
necessary for a bait toxicant to control colonies of the red imported fire ant,
Solenopsis invicta Buren. In laboratory tests it gave delayed action over
more than a 10-fold range of concentrations; in field tests, baits formulated
on pregel defatted corn grits and applied at 20 and 40 g AI/ha produced as
high as 86 and 91% control.
Eli Lilly EL-468, una fenilenodiamina, tiene la toxicidad retardada que
se nesecita para poder usarlo con cebos para controlar las colonies de la
hormiga roja importada, Solenopsis invicta Buren. En pruebas de laboratorio
se mostr6 actividad retardada en un rango de concentraciones de 10 incre-
mentos. En pruebas de campo, se obtiuvieron control de hasta 86 y 91% con
cebos formulados con maiz a medio moler con la manteca extraida y la
condici6n "pre-gelatinosa" que fue aplicada a 20 y 40 g IA/ha.
Large-scale control programs for the red imported fire ant (RIFA),
Solenopsis invicta Buren, were conducted with mirex baits from the early
1960's through 1977 (Lofgren et al. 1963, 1964, Banks et al. 1973, Alley
1973). On 30 June 1978 the registrations of mirex were cancelled by the
Environmental Protection Agency. Justification for the action was attributed
to (1) the discovery of mirex residues in the environment (Ludke et al.
1971, Mehendale et al. 1972, Baetcke et al. 1972); (2) its toxicity to estua-
rine organisms (Lowe et al. 1970, 1971); and (3) its carcinogenic properties
(Innes et al. 1969, Mrak 1969).
Laboratory and field studies by Williams et al. (1980) and Banks et al.
(1981) led to the conditional registration of AC 217,300 (2(1H)-pyrimi-
dinone, tetrahydro-5,5-dimethyl-, [3[4- (trifluoromethyl) phenyl]-1-[2-[4- (tri-
fluoromethyl)phenyl]ethenyl]-2-propenylidene]hydrazone) by the Environ-
mental Protection Agency for the control of RIFA in August, 1980. Pres-
ently, this is the only chemical registered for area-wide control of the RIFA.
In an effort to find delayed-action toxicants for use in RIFA baits, USDA
(ARS) scientists have evaluated hundreds of chemicals each year, with
more than 6,500 chemicals evaluated since 1957. Very few have showed the
delayed toxicity required for a bait toxicant (Banks et al. 1977) and only
mirex and AC 217,300 were registered for commercial use.
'Mention of a pesticide, commercial or proprietary product does not constitute an endorse-
ment or recommendation by the USDA.
Williams & Lofgren: New Bait for Fire Ants
Recently a new chemical, EL-468, (N-[2-amino-3-nitro-5-(trifluoromethyl)
phenyl]-2,2,3,3-tetrafluoropropanamide) produced by Eli-Lilly and Company
has shown promise in our laboratory and field studies. In this paper we re-
port the results of these tests.
MATERIALS AND METHODS
Laboratory evaluation procedures for bait toxicants were described by
Williams et al. (1980). Chemicals are first evaluated against worker ants
in a primary screening test. Those chemicals showing delayed toxicity (de-
fined as <15% kill at 1 day but >89% after 14 days) are given secondary
screening tests against whole laboratory colonies containing a queen and
workers in all developmental stages. The chemicals that cause >90% mor-
tality in addition to killing or sterilizing the colony queen are selected for
small-scale field tests to prove efficacy against natural populations of RIFA.
A brief description of the primary screening test is as follows. Twenty
worker ants from laboratory colonies that had been without food for 14 days
were placed in 30 ml disposable plastic medicine cups ca. 24 hrs preceding
the test. Candidate chemicals were dissolved directly in once-refined soybean
oil (SBO) and offered to the ants on cotton swabs placed in small vial caps.
The ants were allowed to feed on the treated SBO for 24 hrs, then the
swabs and vial caps were removed from the cups and the ants remained
without food for an additional 24 hrs. New vial caps containing cotton
swabs saturated with untreated fresh SBO were placed in the cups and left
for the remainder of the test period. Knockdown and mortality counts were
made at intervals of 1, 2, 3, 6, 8, 10, and 14 days following initial exposure.
Each test consisted of 3 replications at 3 concentrations, 1.0, 0.1, and 0.01%.
The secondary screening tests were as follows. The chemicals were dis-
solved at 1.0, 2.5, and 5.0% by weight in once-refined SBO and then impreg-
nated on pregel defatted corn grits (Lauhoff Grain Company, Danville,
Illinois) at 30% by weight of total formulation. Five grams of the formu-
lated bait were placed in disposable plastic weighing boats (100 ml cap.)
and then each boat was placed in a laboratory colony starved for 5 days
prior to the test. The ants were allowed to feed on the bait for 96 hrs, then
the bait was removed and replaced with the standard laboratory diet of
honey-water (1:1) and Bank's diet (Williams et al. 1980). General observa-
tions on the condition of the colony and mortality counts (estimation of dead
ants and brood) were recorded weekly. The tests were continued until the
queen, brood, and >90% of the workers were dead (in the case of the queen,
complete sterility was sufficient) or the colony had recovered and returned
to a normal condition. This condition was considered reached after the
queen resumed egg laying and all stages of immatures were present. Each
concentration was tested against 2 laboratory colonies consisting of 60 to
120 thousand workers and 50 to 60 ml of brood.
The field tests were conducted in Hamilton County, Florida. The chem-
ical was dissolved at 2.5 and 5.0% by weight in once-refined SBO. Pregel
defatted corn grits were impregnated with the oil solution at 30% by weight
of total formulation to yield baits containing 0.75 to 1.5% active ingredient.
The baits were applied with a tractor-mounted auger applicator (Williams
et al. unpublished data) at 20 and 40g AI/ha (8.4 and 16.5 g AI/A).
Amdro, which is a formulated bait of AC 217,300 from American Cy-
474 Florida Entomologist 64 (4) December, 1981
anamid Company, Princeton, New Jersey, was applied as a standard at 10 g
AI/ha (4.3 g AI/A) for comparison. All treatments were conducted on 0.4
ha plots (3 replications per treatment) located in non-grazed permanent
pasture. Controls were 3 untreated plots (0.4 ha each).
Pre- and post-treatment counts of the number of active nests, the size
of the nests, and the presence (normal) or absence (abnormal) of worker
brood were recorded from 0.2 ha (0.5 acre) circles within each treatment
The percentage control in the field tests was determined by making post-
treatment evaluations at 6, 12, and 21 wk intervals using the method de-
scribed by Harlan et al. (1981). This method uses a 10-point system to rate
each colony based on the absence (categories 1-5) or presence (categories
6-10) of worker brood. A normal colony of RIFA should contain worker
brood, particularly during the warmer months of the year, thus a colony
without worker brood would indicate the absence of a queen or the presence
of a sterile queen. Also, each colony was rated on its size (no. of worker
ants) ranging from <100 workers (categories 1 and 6) to >50,000 (cate-
gories 5 and 10). Therefore, categories 1 and 6 (<100), 2 and 7 (100-1000),
3 and 8 (1000-10000) 4 and 9 (10000-50000) and 5 and 10 (>50000) are
all the same size but differ in whether worker brood is present or absent.
The number of each category in a plot multiplied by the total number of
colonies recorded for that category yields the "population index" for each
plot. The difference between the pre- and post-treatment population indexes
was used to calculate the percentage control for each treatment.
RESULTS AND DISCUSSION
In the laboratory primary screening tests EL-468 exhibited delayed
toxicity at the 0.1 and 1.0% concentrations (Table 1), but it did not show
toxicity at the 0.01% concentration. In comparison, mirex gave delayed kill
at all 3 concentrations. When tested in a bait against entire laboratory
colonies (Table 2) EL-468 killed 95 to 100% of the individual ants within
each colony in 1 wk and all colonies were dead after 8 wks.
The results of field tests (Table 3) suggest that EL-468 gave better
control with both concentrations than the standard, Amdro@, at 6 wks and
as good as or slightly better control at 12 wks; however, the differences were
TABLE 1. RESULTS OF PRIMARY SCREENING TESTS WITH ELI-LILLY EL-468
AGAINST RED IMPORTED FIRE ANTS (AVG. OF 6 TESTS).
Cone. Percent knockdown and mortality
(%) in after indicated number of days
Treatment SBO 1 2 3 6 8 10 14
EL-468 0.01 1 1 1 1 4 8 18
0.1 0 4 12 55 79 89 97
1.0 4 33 78 100
Mirex standard 0.01 0 1 1 7 39 84 98
0.1 1 2 17 66 84 92 100
1.0 0 57 90 100
Soybean oil check 0 0 0 1 2 4 7
Williams & Lofgren: New Bait for Fire Ants
TABLE 2. EFFECT OF CONCENTRATION OF EL-468 AGAINST LAB COLONIES OF
RIFA (5 GMS BAIT/COLONY); 30% SBO ON PREGEL DEFATTED CORN
GRITS; 1 COLONY/TREATMENT.
Percent knockdown and mortality
Cone. (%) after indicated wks*
in SBO 1 2 3 4 8
1.0 95 97 98 99 D
1.0 99 D
2.5 99 99 99 D
2.5 99 D
5.0 (AC 217,300)** 99 D
Check 0 0 0 1 5
*D = death of colony.
**AC 217,300 (Amdro) was used as the standard.
not statistically significant. Evaluations for 21 wks post-treatment showed
no significant differences but it appeared that the low concentration of
EL-468 was not as effective as the standard or the higher concentration of
*EL-468. EL-468 was applied at 2X and 4X the rate of AC 217,300 in the
standard (Amdro@) in these field tests, because a preliminary study indi-
cated an equivalent (IX) rate per hectare gave a maximum of 60% control.
However, more recent tests in Mississippi (Banks et al. unpublished data)
have shown good control of RIFA with rates as low as 3.2 g/ha. Also,
Williams et al. (1980) reported laboratory screening data for AC 217,300
that suggests that at the equivalent concentrations, the 2 compounds are
equally toxic. It is possible that the apparent variations in results reflect
formulation differences which will be resolved in subsequent studies under
an Environmental Use Permit.
An obvious difference between EL-468 and AC 217,300 is that the former
does not exhibit the specific effects on the colony queen that were noted with
TABLE 3. CONTROL OF RED IMPORTED FIRE ANTS WITH EL-468 IN FIELD TESTS
IN HAMILTON COUNTY, FLORIDA. EACH TREATMENT APPLIED TO 8
Cone. (%) Application rate after indicated wks**
Chemical in oil* kg/ha g AI/ha 6 12 21
EL-468 2.5 2.8 20 75a 86a 72a
EL-468 5.0 2.8 40 77a 91a 86a
(Standard) 2.5 1.4 10 66a 88a 80a
Untreated control 7b 11b 9b
*30% Soybean oil on pregel defatted corn grits.
**Avg. 3 replications. Means in the same column not followed by the same letter are sig-
nificantly different at the 5% level of confidence based on Duncan's multiple range test.
Florida Entomologist 64(4)
the latter compound (Williams et al. 1980). Despite these facts, our data
show that EL-468 is an effective delayed-action toxicant which could be an
additional useful tool for the control of imported fire ants and possibly other
pest ant species.
EL-468 can be referred to chemically as a phenylenediamine and thus it
represents the second new class of chemicals that we have discovered that
exhibit delayed toxicity.
The authors gratefully acknowledge the assistance of J. K. Plumley and
D. M. Hicks in conducting the tests.
ALLEY, E. G. 1973. The use of mirex in control of the imported fire ant.
J. Environ. Qual. 2: 52-61.
BAETCKE, K. P., J. D. CAIN, AND W. E. POE. 1972. Residues in fish, wildlife,
and estuaries: Mirex and DDT residues in wildlife and miscellaneous
samples in Mississippi-1970. Pestic. Monit. J. 6: 14-22.
BANKS, W. A., B. M. GLANCEY, C. E. STRINGER, D. P. JOUVENAZ, C. S.
LOFGREN, AND D. E. WEIDHAAS. 1973. Imported fire ants: Eradication
trials with mirex bait. J. Econ. Ent. 66: 785-9.
C. S. LOFGREN, C. E. STRINGER, AND R. LEVY. 1977. Laboratory and
field evaluation of several organochlorine and organophosphorus com-
pounds for control of imported fire ants. USDA, ARS, (Ser.) ARS-S-
169, 13 p.
H. L. COLLINS, D. F. WILLIAMS, C. E. STRINGER, C. S. LOFGREN, D. P.
HARLAN, AND C. L. MANGUM. 1981. AC-217,300, a new amidinohy-
drazone bait toxicant for control of the red imported fire ant. South-
west. Ent. 6: 158-64.
HARLAN, D. P., W. A. BANKS, H. C. COLLINS, AND C. E. STRINGER. 1981.
Evaluation of an amidinohydrazone bait for control of imported fire
ants in Alabama, Louisiana, and Texas. Southwest. Ent. 6: 150-57.
INNES, J. R., B. M. ULLAND, M. G. VALERIO, L. PETRUCELLI, L. FISHBEIN,
E. R. HART, A. J. PALLATTA, R. R. BATES, H. L. FALLS, J. J. GART,
M. KLEIN, I. MITCHELL, AND J. PETERS. 1969. Bioassay of pesticides
and industrial chemicals for tumorigenicity in mice: a preliminary
note. J. Natl. Cancer Inst. 42: 1101-4.
LOFGREN, C. S., F. J. BARTLETT, AND C. E. STRINGER. 1963. Imported fire ant
toxic bait studies: Evaluation of carriers for oil baits. J. Econ. Ent.
S ----, AND W. A. BANKS. 1964. Imported fire ant toxic
bait studies: Further tests with granulated mirex-soybean oil bait.
J. Econ. Ent. 57: 695-8.
LOWE, J. I., P. R. PARRISH, A. J. WILSON, JR., P. D. WILSON, AND T. W.
DULSE. 1971. Effects of mirex on selected estuarine organisms. Trans.
North American Wildl. Nat. Res. Conf. 36: 171-86.
-- P. D. WILSON, AND R. B. DAVISON. 1970. Effects of mirex on crabs,
shrimp, and fish. United States Dept. Inter. Cir. 335: 22-3.
LUDKE, J. L., M. T. FINLEY, AND L. LUSK. 1971. Toxicity of mirex to cray-
fish, Procambarus blandingi. Bull. Environ. Contam. Toxicol. 6: 89-96.
MEHENDALE, H. M., L. FISHBEIN, M. FIELDS, AND H. B. MATTHEWS. 1972.
Fate of mirex-14C in the rat and plants. Bull. Environ. Contam.
Toxicol. 8: 200-7.
Williams & Lofgren: New Bait for Fire Ants
MRAK, E. M., CHAIRMAN. 1969. Report of the Secretary's Commission on
Pesticides and Their Relationship to Environmental Health. Parts I
and II. 677 p. United States Dept. Health, Ed. and Welfare.
WILLIAMS, D. F., C. S. LOFGREN, W. A. BANKS, C. E. STRINGER, AND J. K.
PLUMLEY. 1980. Laboratory studies with nine amidinohydrazones, a
promising new class of bait toxicants for control of red imported fire
ants. J. Econ. Ent. 73: 798-802.
DEVELOPMENTAL RATES AND
EMERGENCE OF VEGETABLE LEAFMINER1
PUPAE AND THEIR PARASITES
REARED FROM CELERY FOLIAGE2
EARL H. TRYON, JR. AND SIDNEY L. POE3
Department of Entomology and Nematology
IFAS, University of Florida
Gainesville, Florida 32611 USA
Vegetable leafminers, Liriomyza sativae Blanchard, were reared from
'2-14' celery foliage held at different constant temperatures to determine the
influence of temperature on pupal development and on host and parasite
emergence. Significantly (P-=0.05) greater numbers of leafminer pupae
were obtained from celery leaflets held at 32.20C than at 15.60C. The percent
emergence of adult leafminers was significantly greater (P-0.05) at the
higher rearing temperature (75% emergence above 200C and 45% emer-
gence below 200C). Larval parasite emergence (Diglyphus intermedius
(Grit.) and Chrysonotomyia formosa (Westwood)) was significantly (P=
0.05) greater at temperatures below 230C (43%) than at higher tempera-
tures (29%). Leafminer pupal developmental time under constant rearing
temperatures increased from 5-7 days at 32.20C to 21.0 days at 15.60C.
Pupal development required a mean 127.8 degree-days, with an estimated
lower threshold temperature of 10.00C. The degree-day requirements for
pupal development were not significantly different at constant rearing tem--
peratures of 15-350C.
Los minadores de hojas de vegetables, Liriomyza sativae Blanchard fueron
criados en el follaje de apios '2-14' mantenidos a distintas temperatures con-
stantes para determinar la influencia de la temperature sobre el desarrollo
pupal y el surgimiento del hubsped y el parasito. Un nimero significativa-
mente mayor (PL0.05) de pupas de minadores de hojas fueron obtenidos en
ojitas de apio mantenidas a 32.20C que de las mantenidas a 15.60C. El por-
centaje de production de adults de los minadores fu6 significativamente
mayor (P=0.05) a temperatures de cria mis altas (75% a mas de 200C, y
45% bajo 200C). El surgimiento de pardsitos larvales (Diglyphus inter-
WLiriomyza sativae Blanchard.
2Florida Agricultural Experiment Station Journal Series No. 1958.
3Current address: S. L. Poe, Head, Department of Entomology, Price Hall, Virginia Poly-
technic Institute and State University, Blacksburg, VA 24061.
Florida Entomologist 64 (4)
medius (Grit.) y Chrysonotomyia formosa (Westwood)) fu6 significativa-
mente mayor (P=0.05) a temperatures bajo 230C (43%) que a tempera-
turas mas altas (29%). El tiempo de desarrollo de pupas de mina doras de
hojas bajo temperatures de cria constantes subieron de 5-7 dias a 32.20C, a
21.0 dias a 15.6C
El desarrollo pupal tom6 un promedio de 127.8 dias de grades, con un
minimo de temperature estimado a 10.00C. El requerimiento del desarrollo
pupal no fue significativamente distinto a temperatures constantes de
Only general information is available concerning the biology of serpen-
tine leafminers, Liriomyza (Diptera: Agromyzidae) and their parasites,
largely because of the difficulty in identifying the leafminer species. Several
temperature regimes are used in this study to provide data to calculate de-
velopmental rate, percent emergence and estimation of degree-days for the
leafminer pupae, Liriomyza sativae Blanchard, and its parasite, Opius spp.
Information about the influence of temperature on the biology of the leaf-
miner is essential for explaining leafminer population dynamics. Such data
may explain not only fluctuations in field populations but help evaluate the
influence of laboratory rearing environments on samples used to estimate
The rate of insect development often approximates a linear function of
temperature and this relationship is the basis for computing accumulations
(Abrami 1972). The theory of temperature summation defines an index for
heat energy required to complete a given stage of development (Eckenrode
and Chapman 1972). Degree-days, or heat units, have been useful methods
in predicting developmental times for insects (Williams and MacKay 1970,
Reid and Laing 1976). Allen (1976) provides a modified sine wave method
for calculating degree-days and modifications for local geographical biases.
In Florida, nocturnal temperatures are relatively constant, thus the sine
wave model tends to over-estimate degree-day values.
MATERIALS AND METHODS
Vegetable leafminers were obtained from leaflet samples collected from
commercial fields of '2-14' celery in May 1978 at Belle Glade, Florida. These
samples were placed in 165 pint-size paper sherbet containers each with 12
trifoliolate celery leaflets and were taken to the laboratory at the University
of Florida, Gainesville. Fifteen containers were placed at each of 11 tem-
perature regimes in growth chambers. Eight samples were held at constant
temperature of 32.20, 29.40, 26.70, 23.90, 22.2, 21.10, 18.20, or 15.60C
(30 each). Two other samples were exposed to fluctuating temperatures.
One regime simulated Orlando, Florida during January 1978 and was based
on historical data from U.S. Weather Bureau statistics with an average
daily high and low temperature of 160C and 2C, respectively, and with 7
nights of subfreezing temperatures (-0.5 to -4.0 C). The other regime,
a 160C day and 20C night temperature cycle (12 h each temperature)
eliminated the influence of freezing on leafminer emergence and develop-
ment. The average R.H. in each of the growth chambers was 54%. The 11th
regime was in an outdoor rearing cage in Gainesville, Florida, during June
1978. The average daily high and low temperatures in this cage were 32.40C
Tryon & Poe: Vegetable Leafminers & Parasites 479
and 21.90C, respectively with R.H. ranging from 74% to 99% (average was
All newly formed leafminer pupae were removed daily from the rearing
containers, placed in separate 4 oz. plastic vials, and were held at the same
temperature regime from which the pupae were collected. Each day both
containers and vials were observed and emerged leafminer adults and their
parasites were removed, identified and counted. Degree-day values were de-
termined for each experimental temperature trial using Allen's (1976)
Fortran program. Development rates (1/time) were calculated for each of
the 8 constant temperatures.
RESULTS AND DISCUSSION
Temperature was a significant factor influencing the mean number of
leafminer pupae collected per celery leaf (Table 1). The leafminer pupae
parasitized by the pupal parasite, Opius spp., were not identified and sep-
arated until adults of both insects had emerged. Percent emergence of adults
from leafminer pupae (leafminer and Opius spp.) was greater at constant
temperatures above 21.1C (75% emergence). Of the total adults (leafminer
and Opius spp.) emerging from leafminer pupae, leafminer adults formed
94% of the emerging adult population. The fluctuating Gainesville regime
resulted in 62% emergence. This lower value was attributed to decay of
foliage in rearing containers and vials. The Orlando regime with 7 nights
of subfreezing temperatures reduced adult emergence to 13.9%. At the
TABLE 1. EMERGENCE OF ADULT VEGETABLE LEAFMINER, Liriomyza sativae
BLANCHARD, AND ITS PARASITES FROM '2-14' CELERY FOLIAGE HELD
AT 11 TEMPERATURE REGIMES.
Mean number % Adult
Rearing leafminer % Adult' parasite2
temperature pupae/ emergence emergence % Total3
regimes (Co) trifoliolate from pupae from pupae parasitism
32.2 7.2a7 84.7a 6.9ab 28.2b
29.4 6.lab 86.8a 4.6ab 22.8b
26.7 5.6b 88.9a 5.5ab 23.4b
23.9 5.4b 77.4a 3.5b 18.5b
22.2 5.6b 78.3a 5.0ab 43.7a
21.1 5.3b 78.5a 6.2ab 49.5a
18.2 5.2b 47.1b 8.2ab 53.7a
15.6 3.4c 43.7b 11.7a 54.3a
Orlando4 0.7- 13.9- 2.3- 51.5-
Gainesvilles 12.6- 62.1- 9.7- 21.9-
16.26 1.8- 43.27 17.7- 70.1-
'Leafminer and Opius spp. adults.
20pius spp. only.
'Includes Opius spp., Chrysonotomyia formosa and Diglyphus intermedius.
4Regime simulated daily Orlando, Florida high and low temperatures during 4-30 January
5Reared in protected cages outdoors in Gainesville, Florida during 1-19 June 1978.
612h:12h temperature cycle of 160C day and 20C night.
'Means followed by the same letter in each column are not significantly different according
to Duncan's multiple range test (P=_0.05). Means without letters were not included in the
Florida Entomologist 64(4)
cooler temperatures the number of Opius spp. adults emerging from leaf-
miner pupae was significantly greater (P<0.05) than the adult leafminer
emergence. Under 160:20C temperature regime there was significantly (P<
0.05) more Opius spp. parasitization (17.7%) than under the Orlando
regime (2.3%). This suggests that cool and freezing temperatures affected
survival of the parasite or of the parasitized host.
Total parasitism included Opius spp. (3.2%), and the 2 larval parasites
Chrysonotomyia formosa (Westwood) (74.4%) and Diglyphus intermedius
(Girault) (22.4%). These latter 2 parasites kill leafminer larvae and
emerge as adults directly from celery foliage. The proportion of total para-
sites significantly (P<0.05) increased as rearing temperatures decreased
and ranged from 18-28% parasitization above 23.90C and 43-70% at tem-
peratures below 23.90C. Chrysonotomyia formosa was the most numerous
of the 3 species of parasites reared. Both C. formosa and Opius spp. had
greater emergence than the leafminer at cooler temperatures. The freezing
temperatures of the Orlando regime increased preadult mortality of Opius
spp. parasities (in vials). Under field conditions leafminer pupae para-
sitized by Opius spp. would be insulated by a layer of soil. This may protect
the parasitized pupae from subfreezing (--4C) temperatures. Previously,
Harding (1965) observed a high rate of parasitization and an associated
decrease in numbers of miners during the colder months in Texas.
Developmental times for leafminer and Opius spp. increased significantly
as temperatures decreased (Table 2). Developmental times for leafminer
pupae ranged from 5.7 days at 32.20C to 21.9 days at 15.60C. In the same
temperature range, Opius spp. emergence ranged from 6.4 days to 27.4 days.
In all trials this parasite required a longer average time (1 day) to mature
than did the leafminer.
The longer developmental times for Opius spp. relative to the leafminer
may provide it a competive advantage over other leafminer parasites when
the life cycle of the leafminer and parasite population are synchronized.
Opius spp. females preferentially parasitize 1st instar leafminers (Lema
and Poe 1979). A slight delay in adult parasite emergence would result in
an increase in available leafminer larvae. The stages in the life cycle of
leafminer field populations are generally synchronized only during initial
periods of invasion.
Emergence of adult leafminers and parasites occurred during a longer
period of peak emergence (70%) of time at cooler temperatures (7-9 days
when reared at 32.2C, 10-16 days at 23.90C, 19-39 days at 15.60C and 21-53
days at the 16C:20C temperature regimes). Emergence of adult Opius spp.
did not demonstrate such a large increase in the number of days to peak
emergence (70%) (9-10 days at 32.20C, 13-14 days at 23.9C, and 31-36
days at 15.60C). Five percent of the total adult C. formosa and D. inter-
medius emerged 7 weeks or later after collection when reared at the cooler
Orlando or 160:20C temperature regimes. Chrysonotomyia formosa was
predominant (97.9% of total adult insects) among late emerging parasites.
Leafminer adults at lower temperatures rarely emerged 30 days after col-
lecting the leaf samples (0.8%). The cooler temperatures of the winter
months in central Florida permit leafminer and parasite population sur-
vival by significantly extending the length of time for their pupal matura-
Tryon & Poe: Vegetable Leafminers & Parasites
t1 Lcooo c0 O O
N 00000000 IV
000i C4 00 000 M LO
0) C C) C) CD CO
o0 a 00 u o 0a cd
C.0O--ji CM O 0
co CI NH L 001 O 0000
00 C4 0D C0 It O 100
0aa.a0 3 0L0i C;
COI t- t- M Cq T- CI 0
Si ; 0 0' 00* L am0 l
a u Z o u0C cN I I I
01 Cj C; 0001 rld 0 0 10 > CO
0) 1 1 1
c j 8 I
o -c l
loj 0 C
0 0M 0
ftrd l t-Cf
>3 0) a0
00F '0 "
00 0-. >
boCC CC 0
OO CiC 0
0Fr o 0- 1
Florida Entomologist 64(4)
The growth curve for the leafminer was calculated using % development/
day from degree-day values (Fig. 1). Degree-day calculations require a low
temperature threshold for pupal development and 10.00C used. This low
threshold temperature for development of leafminer pupae appeared realistic
when substantiated by Jensen and Koehler (1970) who found 24% emer-
gence of adult Liriomyza sp. at 12.80C, by data available on other Diptera
(Allen 1978, pers. commun.), by L. sativae adult emergence of 54% at 15.6C
(Table 1) and 0% at 8.10C (pers. obs.).
The average heat unit requirements for 2500 leafminer pupae and 210
Opius spp. reared at 8 different constant temperature regimes were 127.8
and 141.2 degree-days' C respectively. These 2 means were not significantly
different (P=0.05) from degree-day values for each constant temperature
regime. These degree-day values reflect the delay in emergence of Opius spp.
adults relative to the adult leafminer. The only inconsistent degree-day
estimations were with pupae reared at the 160:20C and Orlando regimes
with degree-day values of less than 100. Apparently subthreshold tempera-
tures (100) affect critical developmental mechanisms within pupae. There
is no indication of diapause.
Degree-day requirements and developmental rates for all stages of
10 15 20 25 30 35
C REARING TEMPERATURE
Fig. 1. Influence of constant temperatures on developmental rates of
Liriomyza sativae Blanchard pupae reared from '2-14' celery foliage (Belle
Glade, Fla., Spring 1978, N = 15, 12 trifoliolate leaves per sample; y = de-
velopmental rate; x = C).
Tryon & Poe: Vegetable Leafminers & Parasites
Liriomyza will help in predicting population densities vital to developing
effective pest management tactics.
ABRAMI, G. 1972. Optimum mean temperature for plant growth calculated
by a new method of summation. Ecology 53: 893-900.
ALLEN, J. C. 1976. A modified sine wave for calculating degree days.
Environ. Ent. 5: 388-96.
ECKENRODE, C. P., AND R. K. CHAPMAN. 1972. Seasonal adult cabbage mag-
got populations in the field in relation to thermal-unit accumulation.
Ann. Ent. Soc. America. 65: 151-6.
HARDING, J. A. 1965. Parasites of the leafminer Liriomyza munda in the
Winter Garden area of Texas. J. Econ. Ent. 58(3): 442-3.
JENSEN, G. L., AND C. S. KOEHLER. 1970. Seasonal and distributional
abundance and parasites of leafminers of alfalfa in California. J.
Econ. Ent. 63: 1623-8.
LEMA, KI-MUNSEKI, AND S. L. POE. 1979. Age-specific mortality of
Liriomyza sativae due to Chrysonotomyia formosa and parasitization
by Opius dimidiatus and Chrysonotomyia formosa. Environ. Ent. 8:
REID, J. A. K., AND J. E. LAING. 1976. Developmental threshold and degree-
days to adult emergence for overwintering pupae of the apple maggot
Rhagoletis pomonella (Walsh) collected in Ontario. Proc. Ent. Soc.
Ont. 107: 19-22.
WILLIAMS, G. D., AND K. H. MACKAY. 1970. Tables of daily degree-day
above or below any base temperature. Canada Dept. of Agr. Pub.
1409. 3 p.
A REVIEW OF FRANKLINIELLA BRUNERI WATSON
AND DESCRIPTION OF F. KELLIAE, N. SP.
Bishop Museum, Honolulu, HI 96819
Frankliniella bruneri Watson, 1925 was rediscovered for the first time
in the past 55 years to be a valid species, and is rediagnosed. Frankliniella
pseudotritici Priesner, 1932 and inornata Moulton, 1936 are newly found
synonyms. Frankliniella kelliae, n. sp., mounted together with the type series
of bruneri, has otherwise long been confused with another well known spe-
cies, F. difficilis. Both F. bruneri and kelliae are widely spread and common
general flower feeders in the Caribbean area.
Frankliniella bruneri Watson, 1925, fue descubierto de nuevo despues de
55 anos como una especie valida, y se diagnostic de nuevo. Frankliniella
pseudotritici Priesner, 1932 y F. inornata Moulton, 1936 son sin6nimos
nuevos. Frankliniella kelliae n. sp. montada junta con la series tipica de F.
Florida Entomologist 64(4)
bruneri, ha sido confundido por largo tiempo con otra especie bien conocida,
F. difficilis. F. bruneri y F. kelliae se alimentan de las flores de diferentes
species y son comunes y distribuidas largamente en la region del Caribe.
Frankliniella bruneri Watson, 1925 from Cuba has remained completely
obscure for the past 55 years because of its initial misplacement and in-
sufficient specific description by Watson. Now it is rediscovered to be a valid
species and furthermore found to be a widely spread and very common gen-
eral flower feeder in the western Caribbean area. Frankliniella pseudotritici
Priesner, 1932 from Mexico and F. inornata Moulton, 1936 from Cuba are
newly found synonyms of the species. Another Frankliniella sp. mounted
together with the type series of bruneri is another widely spread and very
common general flower feeder in the central Caribbean area, and is described
here as a new species. The type series of bruneri studied is a loan from the
Florida State Arthropod Collection (FSAC).
Frankliniella bruneri Watson, REINSTATED, REVISED STATUS
Frankliniella tritici: Watson 1924:46 misidentificationn).
Frankliniella cephalica bruneri Watson, 1926:54; Moulton 1948:91.
Frankliniella cubensis: Jacot-Guillarmod 1974:768, 772 (in part, mis-
Frankliniella gemina pseudotritici Priesner, 1932:179, NEW SYNONYMY.
Frankliniella pseudotritici: Jacot-Guillarmod 1974:808.
Frankliniella inornata Moulton, 1936:62; Moulton 1948:63, 94, fig.; Jacot-
Guillarmod 1974:782, NEW SYNONYMY.
No second collection of this species has been on record, and no subse-
quently so determined specimen remains in either USNM or CAS (Cali-
fornia Academy of Sciences). Moulton (1948) who had apparently not seen
the type series accepted the species as valid, but Jacot-Guillarmod (1974)
summarily synonymized with F. cubensis, without seeing the type series.
This synonymy has no merit.
Type Series: The type series designated for F. cephalica bruneri con-
sists of a long series of specimens (30 9 9, 4 $ ) collected by S. C. Bruner
at La Jaula on Peninsula de Duanahacabibes (2 slides) and Santiago de las
Vegas (9 slides), Cuba from February to April., 1924 from various hosts:
orange, avocado, Hibiscus rosasinensis, Phaseolus helvolus, Moringa sp., and
Rosa sp. Watson (1924) first determined these as F. tritici (Fitch), but 2
years later redetermined them as F. cephalica bruneri, n. sp. (1926). The
presently remaining type series, which is probably in its entirety, in FSAC
is 35 adults and 4 larvae mounted together on 9 slides with Watson's in-
scription of "type and paratypes" or "paratypes", and 2 other topotypic
slides lacking the type status inscription but with same collection data and
Watson's writing of bruneri, n. sp. All the specimens on these slides are de-
termined as shown in Table 1.
Watson (1926) was deficient in describing the specific diagnostic char-
acters, but then few points mentioned, such as yellowish orange ocellar
crescent, dimensions of antennal segments II-IV with diminutive dorsoapical
Sakimura: Review of Frankliniella Spp.
V 7 A
\ '. \ I \ '
\ ^-/ 4
1- '00 U I Figs. 1, 5.
I[ 0 I Others.
_______ SO- Fig. 7.
Fig. 1-8. 1-4, Frankliniella bruneri (specimens all from Jamaica): 1.
Head and prothorax (9); 2. Antenna, right (9); 3. Tergite V (9); 4.
Tergite IX ( $). Fig. 5-8, F. kelliae: 5. Head and prothorax (paratype 9);
6. Antenna, left (paratype 9); 7. Pedicel of antennal III (paratype 9);
8. Tergite IX (allotype $).
production on II and undilated pedicel ("narrow peduncle") of III make
Watson's bruneri fit only Sp. No. 1 (Table 1). Species No. 1 is also pre-
dominating in numbers among all the specimens mounted, and is uniformly
represented on every one of the 11 slides, except one that is paired with the
other slide of the same collection. Species Nos. 2 and 3 have bright red
ocellar cresents, Nos. 2, 3 and 4 all have dilated pedicels, No. 4 has a
slender antenna, and No. 5 has a dark body. Watson's placement (1926) of
Florida Entomologist 64(4)
TABLE 1. SPECIES OF THRIPS MOUNTED TOGETHER WITH THE TYPE SERIES OF
8 paratype A slide with
and 2 "type and
Species slides inscribed* Combined
1. F. bruneri Watson, 1926 14 9 9,3 & 5 3 9 9 17 9, ,3 a $
2. F. cubensis Hood, 1925 5 9 9 1 9 6 9 9
3. F. difficilis Hood, 1925 1 $ 1 $
4. F. kelliae, n. sp. 2 9 1 9, 2 5 3 9 9,2 3
5. F. insularis (Franklin, 1908) 2 a a 2 &S
6. F. sp. (damaged) 1 9 1 9
2299,5 $ 599,3 3 $ 2799,8 5
*Collection data: La Jaula, orange and avocado, 6-IV-1924, S. C. Bruner.
this species in the cephalica group was in error. This error was the primary
cause for a long obscurity in which this species had remained until today.
HOLOTYPE: Among the 3 females of bruneri on the "type and paratypes"
slide, the only one in dorsal mount is selected for the holotype, and a new
label is additionally placed. The holotype is redescribed as follows: Color as
is described in Diagnosis. Body 1610 (all measurements in micrometers;
denotes length only unless otherwise stated; w. = width) (contracted to
950); head about 105, w. 150; interocellar seta 48, postocular seta iv 39.
Antenna 268, length (width) of segments: 23(26), 42(24.5) (dorsoapically
elevated a little but hardly produced), 53(21) pedicell undilated), 43(20),
34(18), 45(17), 6.5(6.5), 10.5(5). Pronotum 130, w. 189; setae: antero-
marginal 64, anteroangular 62, posteroangular i 82, ii 71, posteromarginal
ii 49; pterothorax 200; forewing 750, hind vein with 16 setae; hind tibia
170. Abdomen 1040; comb on tergite VIII indiscernible because of insuf-
ficient clearing (see Diagnosis); ovipositor 230; caudal setae: IXi 105, ii
109, iii 115, Xi 112, ii 102.
DIAGNOSIS: A total of 36 9 9 and 11 $& from Cuba, Jamaica and
Mexico were measured for this diagnosis. Specimens from Mexico are
generally far larger than others. The male form has heretofore not been
described. Intonsa group-runneri complex. Female: Yellow body without
any gray shading or blotching (see next paragraph); antenna I pale, II
grayish brown, basally pale (almost totally pale when general), III-V pale
yellow, distal 1/3-1/2 of III, 3/4 of IV, 1/4-1/3 of V grayish brown, VI-VIII
grayish brown; ocellar crescent brown to reddish brown; wing pale yellow;
major setae dark brown to dark grayish brown; integumental sculpture in-
conspicuous everywhere. Body 1.41-1.86 mm (all distended). Head (Fig. 1)
105-122, broad and rather short (1.4-1.5 times as wide as long); mouth cone
fairly long, straight and tapered; interocellar seta about 1/2 as long as
head, between anterior and posterior ocelli, postocular seta iv about 2/3 of
interocellar seta. Antenna (Fig. 2) 252-303, 2.4-2.6 times as long as head,
II somewhat dorsoapically elevated but not produced, III 2.4-2.6 times as
long as wide, pedicel simple, style stout (VII+VIII/VII w.: 2.6-2.9). Prono-
tum (Fig. 1) 1.1-1.2 times as long as head, all major setae subequal, about
Sakimura: Review of Frankliniella Spp.
1/2 as long as pronotum, posteroangular seta i somewhat longer, postero-
marginal seta ii about subequal to interocellar seta, 4 minor setae between
anteromarginal setae. Forewing with 14-17 setae on hind vein. Abdomen:
ctenidia and chaetotaxy of tergites normal for the genus; comb on tergite
VIII complete, teeth medium in length, sometimes shorter at middle, sparse;
tergites IX-X with all major setae nearly subequal to each other in length,
all longer than tergite X. Male: Paler than female, frail, antenna colored
as female but II often pale and VI sometimes weakly pale at base. Body
1.01-1.32 mm; head 87-105, about 1.4 times as wide as long; antenna 222-260,
2.5-2.6 times as long as head; sternites III-VII with large subcircular grand-
ular area (30-45 pm across); tergite VIII with comb complete but teeth
very small; tergite IX (Fig. 4) with 2 pairs of thick major setae on dorsum,
outer pair 2.4-3.3 times as long as inner pair; 2 other major setae on side
and clasper of sternite in thick and long thorns, sometimes not quite thick-
ened when general.
Muscle Marks: Among many specimens examined, some females were
seen with an ill-defined, small, circular, faint, brown mark (Fig. 3) near
both anterolateral corners of tergites III-VII. These newly discovered marks
among Frankliniella are not related to ordinary grayish brown blotchings
on tergites. These are always in the mesal area along antecosta. According
to the interpretation by B. Heming, the muscle marks are localized heavier
sclerotization of the cuticle at the origins of some of the lateral abdominal
tergo-sternal muscles (Priesner 1926: 60; Hood 1932: 1-6). The muscle
marks of this species are usually very faint and discernible only on well
cleared specimens, but rarely become fairly conspicuous. The incidence of
markings was found persistently among every series from Cuba (but not
seen on the type series), Jamaica and Mexico. Similar muscle marks were
also found in F. kelliae and difficilis. (See pages 488 and 490.)
RELATIONSHIPS: Frankliniella bruneri is one of the congeners of the
runner complex of the intonsa group. All congeners are common in possess-
ing a normal head shape, undilated antennal pedicel, and pale to yellow
body without any blotching on abdomen. They are all poorly characterized
from each other, and diagnostic characters are very limited in most of the
cases. A workable key to species is desperately needed. Frankliniella bruneri,
together with gossypiana Hood, belongs to a small group that is charac-
terized by fully developed postocular seta iv, anteromarginal pronotal seta
subequal to anteroangular seta, antennal style stout, and straight and
tapered mouthcone. Frankliniella gossypiana differs by having a narrower
head (1.3 times as wide as long), larger comb of tergite VIII, and narrowly
spaced occipital striae. Another closely related species is F. gemina Bagnall,
a common general feeder from South America, that differs by having a
distally recurved and bulky mouthcone with its tip not extending beyond
the mentum in dorsal view.
SYNONYMS: While this species remained in obscurity, it was twice de-
scribed under different names, F. pseudotritici Priesner from Mexico (Fed-
eral District, Veracruz) and inornata Moulton from Cuba. The type series
of both species on loan from Senckenberg Museum (SMF) and CAS were
compared with that of bruneri.
DISTRIBUTION AND BIONOMICS: A long series from Jamaica (Sakimura)
and other long series from Cuba and Mexico (USNM, CAS) revealed that
Florida Entomologist 64(4)
bruneri is widely distributed on the islands of Cuba and Jamaica, and from
southern Texas to southern Mexico along the Gulf coast, further south from
Yucatan to Costa Rica (SMF), and probably also into Panama and Colom-
bia. So far no specimen of this species from southern Florida has been seen.
However, its distribution is very likely extended over southern Florida. This
is a common and abundant general flower feeder with a very large host
range including many ornamental, crop and uncultivated plants. This spe-
cies has long been known to be common among cut flowers imported from
Mexico to the U. S. All specimens were misidentified in the USNM Collection
as F. trehernei Morgan by the late D. Moulton. Additional notes on its bio-
nomics are being prepared by the author.
Frankliniella kelliae Sakimura, NEW SPECIES
This species had long been confused with F. difficilis Hood, 1925, and re-
mained unstudied until quite recently. The earliest specimens discovered in
the USNM Collection were collected in 1912 at Key West, Florida, and again
in 1915 at Santiago de las Vegas, Cuba. The latter (Cardin 1918: 58) was
submitted for determination to Hood who left them unnamed. Several other
subsequent specimens (USNM, CAS) were found misidentified as F. dif-
ficilis or tritici (Fitch), or left undetermined. The first critical study was
made by Kellie O'Neill in 1965 with the material collected by the present
writer from Jamaica. Her conclusion was that they represented a closely
related but separate species from dififcilis. By a mere coincidence, 5 speci-
mens of this species were found mounted together with the type series of
F. bruneri collected in 1924 at La Jaula, Cuba (Table 1). A new species is
designated here, and is named after Kellie O'Neill in appreciation of her
broad knowledge on the vast numbers of the American species of Frank-
liniella. Measurements given in the following description are based on
samples from 11 different populations (12 9 9, 3 S& ) from Jamaica and 2
others (4 9 9, 2 $ $) from Cuba.
BRIEF DIAGNOSIS: Tritici group-difficilis subgroup. Yellow body with-
out any abdominal blotching, ocellar crescent brownish, antennal III with
mushroom form pedicel, anteromarginal pronotal seta subequal to antero-
angular seta, comb of tergite VIII always complete but sometimes very
short, caudal setae nearly subequal to each other, male with discal seta on
tergite IX well developed and far larger than seta IXi, without comb on
tergite VIII. Easily confused with F. difficilis in female, but distinct in
DESCRIPTION: Female including holotype: Pale yellow body without any
brown blotching on abdomen, muscle marks on tergites III-VII as in F.
bruneri (Fig. 3), clearly visible but rarely not; antennal I pale, II-V yellow,
distal 2/3 of II grayish brown washed, distal 1/3 of III and extreme of V
brown washed, distal 3/4 of IV dark grayish brown, VI-VIII dark grayish
brown; ocellar crescent brownish; wings pale yellow; major setae brown to
dark grayish brown; integumental sculptures inconspicuous except at sides
of tergites and laterotergites. Body 1370-1660 (1610) (measurement in
parentheses following the range denotes holotype or allotype); head (Fig.
5) 97-112 (112), w. 140-158 (156), 1.4-1.5 times as wide as long; mouth
Sakimura: Review of Frankliniella Spp.
cone rather short, straight and tapered; interocellar 39-54 (43), 0.3-0.5 of
head, variable, near the outer common tangent of anterior and posterior
ocelli; postocular iv 28-39 (39), 0.6-0.9 of interocellar, variable. Antenna
(Fig. 6) 256-292 (292), 2.5-2.6 times as long as head; II 37-41 (41), w.
22-24 (24), 1.6-1.8 times as long as wide, dorsoapically somewhat elevated
but hardly produced; III 49-59 (56), w. 18-21 (20), 2.6-3.0 (2.8) times as
long as wide, pedicel (Fig. 7) in regular mushroom form; style stout (VII+
VIII / VII w. = 2.6-2.9); length (width) of all segments of holotype: 28
(28). 41(24), 56(20), 50(18.5), 38(18.5), 49(17), 7(7), 10(5.5). Pronotum
(Fig. 5) 117-133 (133), w. 153-191 (186), 1.4-1.5 times as wide as long, 1.1-
1.2 of head; anteroangular 51-64 (64), 0.4-0.5 of pronotum, anteromarginal
44-59 (59), 0.8-1.0 of anteroangular, usually subequal but rarely a little
smaller, somewhat variable, 4 minor setae between anteromarginals; postero-
angular i 60-79 (76), ii 51-68 (68), subequal, 0.4-0.6 of pronotum, postero-
marginal ii 33-50 (50), 0.3-0.4 of pronotum. Pterothorax 184-212 (204);
forewing 600-790 (750), with 14-18 setae but usually 15-16 on hind vein;
hind tibia 153-179 (179). Abdomen 870-1070 (1070), ctenidia and discal setae
on tergites normal for the genus, comb on tergite VIII always complete but
variable in length (5-12), ovipositor 198-225 (222); caudal setae: IXi 77-102
(91), ii 90-120 (105), iii 89-115 (100), Xi 99-113 (111), ii 84-105 (102),
subequal each other.
Male including allotype: Body paler than 9, antennal I pale, II-III and
V yellow, nearly without brown or gray wash at all, IV yellow with distal
1/2 grayish brown, VI-VIII grayish brown, VI sometimes lighter at extreme
base; without muscle marks. Body 920-1180 (1180), head 101-106 (106), w.
130-138 (138), 1.3 times as wide as long, interocellar 36-40 (39), postocular
iv 24-33 (29), antenna 238-252 (252), both II-III similar in length / width
to 9. Pronotum 145-153 (153), w. 105-111 (111), 1.4 times as wide as long,
anteroangular 40-49 (45), anteromarginal 41-44 (43). Comb on tergite VIII
absent, glandular areas on sternites III-VII large, transverse, oblong;
genitalia about 93. Caudal setae (Fig. 8) : IXi 15-19 (19), discal 54-60 (54),
IXii 45-53 (53), iii 39-46 (43), clasper 62-73 (73), the first 2 in thick seta
form with discal far larger than i, the last 3 in blackish thorn form and
thickness somewhat variable among specimens.
TYPE: Holotype 9, allotype $, Jamaica, Kingston, Hope Garden,
Tithonia diversifolia fl., 15-XI-1964, K. Sakimura (Saki 4408-1 9, -3 $)
USNM Type 100069). Paratypes: same data, 1 9, 2 $S ; same locality and
date, Blighia sapida fl., Bauhinia triandra fl., Calophyllum inophyllum fl.,
and Caesalpinia coriaria fl., 16 9 9; Bryan Castle, Eupatorium odoratum
fl., 1 9 ; Bengal, Lantana camera fl., 1 9 ; Falmouth, Canavalia rosea fl.,
3 9 9; Round Hill, Barnett, Bengal, and Woodstock, all from Cocos nucifera
male fl., 7 9 ; Hope Garden, Cassia siamea fl., 3 9 9; all these 32 9 9,
2 S 9 paratypes: 15-XI-8-XII-1964, Sakimura (Saki 4404a, 4408, 4410c,
4411, 4416, 4449c, 4454a, 4455a, 4476, 4477, 4520a, 4530a), R. Latta (Saki
OTHER SPECIMENS EXAMINED: Jamaica: Bog Walk, "net", 2-II-1937, E. A.
Chapin, 1 9 (USNM). Cuba: Santiago de las Vegas, citrus fl., and Carissa
acuminata fl., 23, 26-IV-1915, P. G. Cardin (Hood 93, 94, 99), 3 9 9, 2 S $
(USNM); La Jaula, orange and avocado, 6-IV-1924, S. C. Bruner (mounted
together on 2 slides of Watson's type series of F. bruneri), 3 9 9, 2 & $
490 Florida Entomologist 64(4) December, 1981
(measured) (FSAC); Baragua, citrus fl., and Nerium oleander fl., 7, 10-IV-
1927, L. C. Scaramuzza, C. F. Stahl (Moulton 1850, 1851), 1 9, 1 & (meas-
ured) (CAS). Florida: Key West, 23-IV-1912, G. A. Runner and A. C.
Morgan, 1 9 (USNM); Big Pine Key, "flowers", 27-XII-1950, J. D. Hood
(Hood 1928), 1 9 (USNM). Bahamas: Nassau, lantana, VI-1956, N. L. H.
Krauss, 5761, 1 9, 1 3 (USNM). Dominican Republic: Porto Libertador,
citrus, 14-VI-1952, C. Dowling #44, 52-6604, 7 9 9, 1 & (USNM). Puerto
Rico: Rio Piedras, 8-IX-1947, G. N. Wolcott, 47-12970, 2 9 9 (USNM).
DISTRIBUTION AND BIONOMICS: The above data indicate that collections
were made from Florida, Bahama, Cuba, Jamaica, Dominican Republic, and
Puerto Rico. The species is apparently distributed widely in the central
Caribbean area, and is a very common general flower feeder. Additional
notes on its bionomics are being prepared by the author.
RELATIONSHIPS: Frankliniella kelliae and difficilis are closely related and
have only a few distinguishing features such as color of the ocellar crescent
and, in the male, the comb of tergite VIII and chaetotaxy of tergite IX.
Muscle marks are present in both species but apparently are less common
in difficilis. When the male is not collected together with the female and the
ocellar crescent is faded on the mounted specimens, there has been and will
be no way to make a reliable separation between them. Frankliniella dif-
ficilis differs from kelliae by having a red ocellar crescent, and, in the male,
by having a short but complete comb with both IXi and discal setae being
in a thorn form of subequal lengths (about 25-30 jm long).
Another allied species is F. tritici that is often lacking the brown
abdominal blotchings of the pale form in the Caribbean area. It differs from
kelliae by having shorter anteromarginal pronotal seta than anteroangular
seta, and incomplete comb on tergite VIII.
Acknowledgments are gratefully made to H. A. Denmark, Florida De-
partment of Agriculture and Consumer Services, Gainesville; Steve Naka-
hara and Kellie O'Neill, Systematic Entomology Laboratory, U. S. Depart-
ment of Agriculture, Beltsville, MD; Richard zur Strassen, Senckenberg
Museum, Frankfurt a. M., Germany; D. H. Kavanaugh and P. H. Arnaud,
Jr., California Academy of Sciences, San Francisco; and B. Heming, Uni-
versity of Alberta, Edmonton, Canada for cooperation in allowing loans of
type material, providing advice and information upon inquiries, or extending
generous cordiality and assistance while I was visiting the collections.
CARDIN, P. G. 1918. Notas entomologicas. Mem. Soc. Cubana Hist. Nat.
"Felipe Poey" 3: 53-61.
HOOD, J. D. 1932. Tergo-sternal muscles in the Thysanoptera. Bull. Brooklyn
Ent. Soc. 27: 1-6.
JACOT-GUILLARMOD, C. F. 1974. Catalogue of the Thysanoptera of the world.
Ann. Cape Prov. Mus. (Nat. Hist.) 7(3) : 517-976.
MOULTON, D. 1936. New American Thysanoptera. Bull. Brooklyn Ent. Soc.
-- 1948. The genus Frankliniella Karny, with keys for the determina-
tion of species (Thysanoptera). Rev. Ent. (Rio de Janeiro) 19: 55-
Sakimura: Review of Frankliniella Spp.
PRIESNER, H. 1926. Die Thysanopteren Europas, Part 1. Fritz Wagner,
Wien, 238 p.
.1932. Neue Thysanopteren aus Mexiko, gesammelt von Prof. Dr.
A. Dampf. I. Teil. Wiener Ent. Zeitung 49: 170-85.
WATSON, J. R. 1924. Adiciones a los Thysanoptera de Cuba. Mem. Soc.
Cubana Hist. Nat. "Felipe Poey" 7: 46-7.
1926. Two new Thysanoptera from Cuba. Florida Ent. 9: 53-5.
--*-^---- -- *- -- -^-^- --
BIONOMICS OF THE BROMELIAD-INHABITING
MOSQUITO WYEOMYIA VANDUZEEI AND
ITS NURSERY PLANT TILLANDSIA UTRICULATA1
J. H. FRANK2 AND G. A. CURTIS3
Aquatic stages of the mosquito Wyeomyia vanduzeei Dyar & Knab in-
habit leaf axils of the epiphytic bromeliad Tillandsia utriculata L. in south-
Rough-barked trees provide a substrate for T. utriculata and a habitat
for W. vanduzeei. The volumetric capacity of T. utriculata leaf axils can be
estimated from length of longest leaf volumetricc capacity in ml = 0.003251
x leaf length in cm2.779). In each plant, older, larger axils provide most of
the volumetric capacity, but the outermost axils lose their ability to retain
water. Water entering bromeliad axils is throughfall, i.e. rainfall reduced in
volume but enriched with nutrients as it penetrates the tree canopy
(throughfall mm = -0.2715 + (0.8825 x rainfall mm) in a high marsh
study area). Organic debris from the tree canopy (estimated annual mean
2.4 g/m2/day in a high marsh study area), trapped and decomposed in the
bromeliad axils, also provides nutrients for aquatic organisms and for the
bromeliad. The perennial bromeliad provides a stable habitat for mosquito
The number of mosquito eggs laid in axils is influenced by the size, water
content, senescence and flower production of the bromeliad; as many are laid
in the smaller, inner axils as in the larger, outer axils. At 270C, most float-
ing eggs hatched within 48 and 96 h. Eggs stranded out of water suffered
loss in numbers and in viability, with combined loss at 10 days of 49%, at
20 days of 77%, and at 30 days of 95%
En el sur de la Florida (EUA), los estadios acuaticos del mosquito Wy-
eomyia vanduzeei Dyar & Knab habitan en las axilas de las hojas de la
bromelia epifitica Tillandsia utriculata L.
Arboles con la corteza rugosa proven un substrato para T. utriculata y
un habitat para W. vanduzeei. La capacidad volum6trica de las axilas de
las hojas de T. utriculata puede ser estimada del largo de la hoja mas larga
(capacidad volum6trica en ml = 0.003251 x largo de la hoja en cm2.7799). En
'Institute of Food and Agricultural Sciences, University of Florida, Journal Series No. 2917.
2Florida Medical Entomology Laboratory, Institute of Food and Agricultural Sciences, Uni-
versity of Florida, P. O. Box 520, Vero Beach, FL 32960-0520, USA.
3Indian River Mosquito Control District, P. O. Box 670, Vero Beach, FL 32961-0670, USA.
Florida Entomologist 64(4)
cada plant, las axilas mas viejas y mayores proven la mayoria de su
capacidad volum6trica, pero las axilas exteriores pierden la capacidad de
retener agua. El agua que entra en las axilas es la caida pasante, es decir,
lluvia reducida en volume, pero enriquecida con nutrients al penetrar la
b6veda del foliage (caida pasante en mm -0.2715 + (0.8825 x mm de
lluvia) en una zona de studios en la parte mas elevada de una marisma.)
Escombro orgAnico de la b6veda de los arboles (caida media annual 2.4
g/m2/dia en una zona de studios en la parte elevada de una marisma),
atrapado y descompuesto en las axilas de las bromelias tambi6n provee
nutrients a los organismos acuaticos y a las bromelias. La bromelia perenne
provee un habitat stable para larvas de mosquitos.
El nimero de huevos de mosquitos depositado en las axilas es influenciado
por el tamafio, contenido de agua, senectud, y producci6n de flores de la
bromelia; la misma cantidad de huevos es depositada en las mas pequefias
axilas interiores que en las mayores axilas exteriores. A 27C, la mayoria
de los huevos flotantes eclosionan dentro de un period de 48-96 h. Huevos
varados fuera del agua sufrieron p6rdidas en nfmeros y en viabilidad, con
ptrdidas combinadas en 10 dias de 49%, en 20 dias de 77%, y en 30 dias de
Immature stages of hundreds of mosquito species have been recorded
from water-containing leaf axils of numerous bromeliad species in the
neotropics. Mosquito-bromeliad associations are less complex in southern
Florida where the neotropical fauna and flora are depauperate. The mos-
quitoes Wyeomyia vanduzeei Dyar & Knab and Wyeomyia mitchellii (Theo-
bald) alone represent more than 100 neotropical Wyeomyia species. Tilland-
sia utriculata L. is 1 of 2 common, indigenous bromeliads holding water
in their axils, and represents a family of over 2,000 neotropical species. This
paper describes aspects of the ecology of W. vanduzeei and its nursery plant,
Tillandsia utriculata As A MOSQUITO NURSERY
DISTRIBUTION AND HABITATS
This bromeliad occurs in southeastern Georgia, Florida, the West Indies,
Mexico, Central and South America (Small 1933). In southern Florida it is
an abundant epiphyte of rough-barked trees such as oak (Quercus app.),
cypress (Taxodium spp.) and buttonwood (Conocarpus erecta L.) (Butler
1974). Our observations concern mainly high marsh (Provost 1974) near
Vero Beach, Florida (270 38' N, 800 23' W), but some relate to the adjoin-
ing landward area. The mature buttonwoods in our study area do not ex-
ceed 4 m in height, while oaks of the adjoining landward area exceed 12 m.
Thus, T. utriculata has a greater vertical range on oak. The tree trunk and
branches supply anchorage, and the canopy supplies nutrients and protection
from frost. Even though some tillandsioid bromeliads are well-adapted to
high levels of solar radiation (Pittendrigh 1948, Benzing and Renfrow 1971),
T. utriculata normally were exposed to little direct solar radiation in our
study area because of the evergreen buttonwood canopy. The live oaks
(Quercus virginiana Mill.) of the landward area also are evergreen. Ab-
normally low temperatures occurred in our study area in January 1977, but
no damage to the bromeliads was seen, whereas canopies of buttonwood were
Frank & Curtis: Mosquito & Its Nursery Plant
damaged severely by frost. Tillandsia utriculata plants left above the level
of the frost damage were directly exposed to solar radiation, and nutrient
input eventually was reduced; many lost greenness, many died, and others
flowered at an usually small size. The bromeliad appears normally to grow
only under tree canopies.
After storms, the ground under live oaks is littered with branches and
bromeliads. The latter seldom survive long, and are eaten by rabbits (east-
ern cottontail, Sylvilagus floridanus Lyon). Casual observation suggests that
buttonwoods suffer few storm-lost branches, but if bromeliads fall from
buttonwood, they are also eaten by rabbits (marsh rabbit, Sylvilagus
palustris (Bachmann)). The abundance of T. utriculata probably is limited
by available habitat, climate, and dispersive abilities of seed.
Tillandsia utriculata plants (Fig. 1) mature several years after germina-
tion and after one season's seed production. The growing plant produces
leaves at its apex in a clockwise spiral, where each subtends an angle of
about 1370 to its antecedent with reference to the plant's axis. The newest
leaves are small and slender regardless of the plant's size. Older leaves may
measure more than 10 cm across their bases and may exceed 1 m in length.
As the plant grows, the oldest leaves die sequentially; thus the base of a
plant may be shrouded in dead leaves which hold humus in their axils.
The leaf bases overlap broadly and fit closely at their edges. The living
leaves, except the outermost, trap water in their axils. Each small body of
water has a crescent-shaped surface area determined by the curvature of
the leaf base around the plant axils, and each body of water is narrowest
at its deepest point, where the leaf is attached. In a plant containing visible
quantities of water, the surface is at a lower level in older, larger axils than
in newer, smaller ones. Water poured into the center of a dry plant fills each
axil in turn, and the excess flows sequentially into the larger axils below.
In older plants, bases of about 50% of the green leaves, the inner ones,
are fleshy and brittle; these axils together have less than 20% of the total
water-holding capacity, and the water bodies of the innermost majority of
them are not discrete. The bases of the remaining 50% of the green leaves
are fibrous; the outermost half of these hold no more than drying humus,
so that up to 80% of the water-holding capacity is due to only about 25%
of the axils; these 25% hold discrete bodies of water and form clusters of
phytotelmata (Varga 1928). Very small plants have few leaves and trap no
water. Small plants hold water only during wet weather. Large plants hold
water at all times, even though water level may be low during drought. Total
water volume in axils of the largest plants may exceed 1.3 liters.
The length of the longest leaf was used as a measure of plant size, be-
cause this was determined easily without disturbance of axil contents. Total
volumetric capacity of axils was compared by measuring longest leaf
length to the nearest 1 cm and volumetric capacity to the nearest 5 ml in 98
experimental plants (Frank et al. 1977). A power curve regression (y =
a' xb) gave a better fit to the data (Fig. 2) than did either a linear (y =
a' + bx) or exponential (y = a' ebx). The accepted power curve regression
(a' = 0.003251, b = 2.7799, N =: 98, F = 277.46) gives estimated volumetric
capacity (y) when leaf length (x) is known.
Florida Entomologist 64(4)
Fig. 1. Tillandsia utriculata plant cut to show form of leaf axils.
Growth of 100 plants of different sizes was followed for 536 days. Plants
growing on the north side of live oak trunks were considered separately
from those on the south side. Five plants were selected and marked in each
size range 0-5, 5-10, . 45-50 cm, based on length of longest leaf, giving a
total of 50 plants each in positions with northern and southern aspects, at
various heights from the ground. Longest leaf measurements were taken on
14-XI-1974 and on days 109, 214, 337 and 536 thereafter. During the entire
Frank & Curtis: Mosquito & Its Nursery Plant
* I. .
200- - *.
O : ..
a ) /*.'
100 2 4 6 1
F .leaf length (cm)
0 20 40 60 80 100
Fig. 2. Scattergram with fitted regression line (y = a' Xb) relating
volumetric capacity of Tillandsia utriculata to length of longest leaf.
period, 17 plants died or fell from their anchorages from natural causes.
There was no significant correlation between mortality and size (N = 10,
r = 0.2596, P > 0.4), and no significant difference in mortality or dislodge-
ment of plants on the northern vs. the southern aspects (N = 20, ti = 0.5797,
P > 0.5). Annual mortality (within each of the size ranges separately), was
roughly 14% (SD 14.3).
496 Florida Entomologist 64 (4) December, 1981
A linear regression of leaf length vs. time was calculated and plotted
for each of the 10 size ranges. There was no evidence of curvilinearity, and
none of the standard errors exceeded 4% of the value of each coefficient.
These regression coefficients were plotted against mean leaf length of plants
in each size range, and another linear regression was calculated (a' =
0.0017, b = 0.00078, SE of b = 3.206 X 10-7). Daily percentage increment of
leaf length is thus 0.078%, and future leaf length = x, = xo (1 +b)n, where
x, = leaf length at time t in the future, xo = present leaf length, b = rate
of increase (0.00078), n = period (days) from time xo to xt. Table 1 was
compiled from estimates made by this formula, according plants of 2 cm
leaf length to time zero years. Actual leaf length at germination is about
The canopy of the high marsh study area consists of intermingled foli-
age of buttonwood, white mangrove (Laguncularia racemosa Gaertn.) and
black mangrove (Avicennia germinans (L.)), and is sparse relative to that
of oak in the adjoining landward area. Rainfall and throughfall were meas-
ured using 20 galvanized metal rain gauges, precise to the nearest 5 ml
(0.2 mm). Using a stratified random distribution system 5 gauges were
placed in the open to measure rainfall, and 15 under the tree canopy to
measure throughfall, the latter away from gaps in the foliage. Gauges in
the open were examined every morning during a period of 6 weeks in
September-November 1976 and, when water was present, measurement was
made of the volume in all 20 gauges.
Standard deviations of means of each day's rainfall measurements were
very low, so each mean (x) for rainfall was treated as a single value when
linear regression of throughfall (multiple y values) on rainfall (x) was
TABLE 1. PARTIAL LIFE TABLE DATA FOR Tillandsia utriculata.
Length of Volumetric Population
Time longest leaf capacity Survival structure
(years) (cm) (ml) (%) (%)
0 2.00 0 100 16
1 2.66 0 86 13
2 3.53 0 74 12
3 4.70 0 64 10
4 6.24 1 55 9
5 8.30 1 47 7
6 11.03 3 41 6
7 14.66 6 35 5
8 19.49 13 30 5
9 25.91 28 26 4
10 34.43 61 22 3
11 45.77 134 19 3
12 60.84 296 16 3
13 80.87 654 14 2
14 107.49 1442 12 2
Frank & Curtis: Mosquito & Its Nursery Plant
performed (a' = 0.2715, b = 0.8825, Nx = 11, Ny = 11 X 15, F = 3,540).
The linear regression explains almost all of the difference between sets, and
a better fit would not be expected by curvilear regression. However, the value
of a' is very small and analysis of variance showed that it is not signif-
icantly different from a zero intercept. If the intercept really is zero, then
throughfall must be considered a constant percentage of rainfall. Reynolds
and Leyton (1963) gave evidence that when rainfall occurs, it must exceed a
small value (the value of a', as a positive figure, termed canopy saturation),
before any penetrates the tree canopy. Accepting this interpretation, then
the relationship of throughfall as a percentage of rainfall to rainfall is
curvilinear, and rises to an asymptote. This asymptote is calculated as 100
X the value of b (i.e. 88%), and compares with values of about 79%
(winter) and 81% (summer) obtained by Trimble and Weitzman (1954)
for throughfall in a deciduous forest in West Virginia (Geiger 1971). Ap-
parently, the sparser tree canopy in our study area permitted more through-
fall at the asymptote. The value for canopy saturation was about 0.3 mm in
our study area, and in the West Virginia study it was about 0.2 mm in
winter and about 0.4 mm in summer (Trimble and Weitzman 1954). In our
study area, throughfall is available for catchment by bromeliads situated
under the tree canopy, and derives from annual average rainfall of ca.
In the high marsh area in 1976, 400 selected T. utriculata plants of
maximum leaf length between 40 and 50 cm were used for an investigation
of a W. vanduzeei population. Plants were hung, in pairs, by nylon string
from the limbs of buttonwood and white mangrove. The perimeter of each
plant was taken as the position where the curvature of large leaf blades
became horizontal, and average diameter was about 25 cm. Thus, catch-
ment area was about 491 cm2.
Estimates were made of plant debris (leaves, twigs, seeds) falling over
a circle of 25 cm diameter beneath the tree canopy. Plastic plant pots of
diameter 25.4 cm and depth 24 cm were used as catchment baskets. Drain-
age holes in the pots were taped over to retain fine particles, and a pot was
tied adjacent to and at the same height as each of the 200 pairs of T.
utriculata. From January 1976, debris was removed from each pot at
fortnightly intervals, and 10 of the 200 collections, chosen by stratified
random sampling plan, were stored separately and air-dried for 3 weeks at
room temperature. Dry samples were weighed, and means and 95% con-
fidence limits were calculated; these values are further converted to g/m2/
day in Fig. 3.
Measurements ended in November 1976, less than a full year later, be-
cause of time restrictions. Results were thus from 21 sets, each of 10
samples. These were assigned to spring (the 7 sets up to and including 7
May), summer (the next 7 collections), and autumn (final 7 collections).
The spring samples weighed significantly less than summer samples (ti =
3.8642, P < 0.001), and significantly less than autumn samples (t, = 3.5272,
P < 0.001), but there are no significant differences (ti = 0.0343) between
summer and autumn samples.
Pool et al. (1975) reported that highest rates of litterfall in mangrove
498 Florida Entomologist 64 (4) December, 1981
/ 5 mm
day 4 150
0- -. 0
F M A M J J A S 0 N
Fig. 3. Histogram showing mean and 95% confidence limits of fortnightly
collections of organic debris fallen from tree canopy in high marsh study
area. Dry weight is given as g/m2/day. Dots indicate total rainfall in mm
during the same fortnightly intervals.
forests corresponded with annual peaks of wet season and frequent wind
storms. Total fortnightly rainfall (in mm), from a standard rain gauge
about 0.6 km from the study area, is indicated in Fig. 3. Mean debris sample
weight does not quite achieve a significant correlation with rainfall (1-
tailed, r = 0.3610, P < 0.1). A similar procedure performed on a daily basis
might have yielded a significant correlation.
Pool et al. (1975) measured litterfall in mangrove forests at several
localities in southern Florida and Puerto Rico, and found that annual fall
ranged between 0.2 and 3.96 g/m2/day between localities, with a mean of
2.2 g/m2/day over all localities. The present data do not cover an entire
year, but the mean litterfall for the period considered is 3.62 g/m2/day. If
litterfall over the period from November to January is assumed to be at the
same rate as from February to May, the annual figure for the study area
was about 2.4 g/m2/day. According to Pool et al. (1975), the rate of litter-
fall in mangrove ecosystems is higher than in surrounding ecosystems.
Thus, bromeliads in mangrove ecosystems should trap more debris than
should those in adjacent inland areas.
Tillandsia utriculata plants use nutrients from debris, leachates, and
possibly stemflow. Solid materials trapped include leaves, twigs, seeds and
pollen of trees, as well as excreta of lepidopterous larvae feeding on the
trees. Additional materials may come from associated vines and from other
trees intermingled in the canopy. Use by bromeliads of leachates from tree
canopies was demonstrated by Tukey (1970). Types of nutrients in through-
Frank & Curtis: Mosquito & Its Nursery Plant
fall were investigated by Voigt (1960), Carlisle et al. (1966), and Tukey
(1970), but not for mangrove ecosystems. Carlisle (1967) studied oak
stemflow, which also contains nutrients. Some T. utriculata plants may be
suitably positioned on oak trunks to be able to trap stemflow, but those on
the more horizontal branches of oak, and on buttonwood seem poorly posi-
tioned to use this resource. Throughfall from the salt-coated canopy of black
mangrove apparently is deleterious to T. utriculata (Frank and Curtis
1977b), and in the present study bromeliads have not been found growing
on black mangrove, yet Lugo and Snedaker (1974) stated that this associa-
The chemistry of water held in axils of bromeliads was investigated by
Laessle (1961), McWilliams (1970), Benzing (1970, 1973), Benzing, Derr
and Titus (1972) and Benzing and Renfrow (1974). The water is acid, and
in most instances bromeliads absorb nutrients by means of trichomes (ab-
sorbing scales). An ecological classification of bromeliads according to
method of nutrient uptake was initiated by Pittendrigh (1948) and expanded
by Rauh et al. (1973) and Benzing (1977).
In the laboratory, organic debris added to bromeliad infusion water
produces a bacterial scum, but addition of aquatic insect larvae, e.g. mos-
quito larvae, causes the scum to be removed, at least in part by feeding by
the insects. Continued presence of insect larvae eventually causes exhaus-
tion of the nutrient supply, and the infusion water becomes clear. Aquatic
organisms may compete with bromeliads for nutrients, and emerging adult
insects bring about a removal of nutrients from the microhabitat. At the
same time, digestion of particulate matter by the insects must contribute to
the partial breakdown of nutrients, making these more readily available to
Adult W. vanduzeei of both sexes perch on the most shaded side of tree
trunks. Preference for shade tends to keep mosquitoes under tree canopies,
in the vicinity of bromeliads. Females lay their eggs singly while in flight.
Oviposition flight is jerky and bobbing and is performed in and above
bromeliad axils, presumably allowing a rapid response of the female mos-
quito to external stimuli. The air space above the central axils of T.
utriculata is not limited by any part of the plant, except when a flower
spike is produced. The air space above the larger, outer axils may be limited
by downcurvature of leaves above, thus hindering oviposition.
Mosquito flight was observed in relation to a caged T. utriculata plant of
about 120 ml capacity from which the distal portions of the longer leaves
had been cut. Successive bobbing flights'over the axils were timed, some
over the inner axils and some over the outer axils. Seventeen flights of av-
erage duration 14 sec (SD 14) occurred over the inner axils and 19 flights
of average duration 10 sec (SD 13) over the remaining axils. Thus, inner
axils, containing a much smaller volume of water, attracted a similar num-
ber of flights of similar duration as did outer axils. Disproportionately
large oviposition in inner axils is likely to increase the probability of egg
survival, because central eggs are in less danger of being washed from the
Florida Entomologist 64(4)
plant during heavy rainfall (Frank and Curtis 1977b). We envisage a
progression of eggs and larvae from inner to outer axils, partly as a result
of rainfall so distributing the eggs, partly as a result of plant growth, and
in the absence of any indication that either eggs or larvae move in the
To test plant quality on oviposition, we selected pairs of dying and
healthy plants of similar (within 5 ml) volumetric capacity. Each dying
plant had roughly 70% of leaf surfaces browned (necrotic). A comparison
(Table 2) of number of eggs laid was performed using the methods of
Frank et al. (1976). Significantly more (P < 0.001) were laid in healthy
Because T. utriculata plants die after flowering, the presence of a flower
spike, even if a plant has no browned leaves, portends approaching death.
Using the same method as above, oviposition was compared (Table 2) in
pairs of matched plants where 1 member of each pair had a flower spike,
and the second did not. Both members appeared healthy and without
browned leaves. Significantly more (P < 0.05) were laid in plants without
We have demonstrated (Frank et al. 1976) that oviposition is affected
by plant size and by presence of water in the leaf axils, but that presence of
organic infusion, or infusion plus organic debris, appears to have no affect.
Wyeomyia vanduzeei eggs float, aided by a waxlike flotational device
(Frank et al. 1981). Advantages of floating may be associated with dis-
RESULTS OF OVIPOSITION TESTS USING CAGED MOSQUITOES AND PAIRS
OF PLANTS MATCHED FOR VOLUMETRIC CAPACITY. MORE EGGS WERE
LAID IN HEALTHY PLANTS THAN IN DYING PLANTS (td = 4.5663,
P < 0.001). MORE EGGS WERE LAID IN PLANTS WITHOUT FLOWER
SPIKES THAN IN PLANTS WITH FLOWER SPIKES (td = 2.5840, P <
Number of mosquito eggs laid
Health of plant flower spike
Pair Trial Healthy Dying Without With
1 1 168 46 634 82
2 326 97 326 420
3 476 114 486 385
4 651 124 576 512
2 5 161 83 594 531
6 225 112 1362 683
7 142 50 266 288
8 201 33 1025 794
3 9 142 45 154 137
10 114 16 933 297
11 163 13 595 462
12 154 47 267 227
Frank & Curtis: Mosquito & Its Nursery Plant
tribution towards outer axils, or with avoidance of submerged predators, or
with respiration. Other advantages of the device may be in plastron respira-
tion if trapped below the water surface, in protection from terrestrial pred-
ators, or in reduced desiccation if stranded above the water line. Stranded
eggs do not hatch until inundated by rainfall.
To test affects of stranding above water on viability, a series of plants
of similar volumetric capacity was assembled. Replicated treatments using
methods similar to those of Frank et al. (1976) compared number and
viability of eggs recovered from bromeliads stored for 10, 20 or 30 days
after oviposition with paired controls. Before experimental use, plants were
cleaned and submerged in water for 24 h to remove their fauna, then
shaken in air to remove most of the water, but allowing sufficient to remain
to stimulate oviposition by caged mosquitoes. After 24 h in a cage of mos-
quitoes, eggs in the control plant were extracted, counted and tested for
viability by attempting to hatch them in a standard way in dishes of water
Mean viability of eggs extracted from control plants was 88.3 6.2%
(30 replicates). Most of these eggs hatched within a period between 48 and
96 h from oviposition (Fig. 4). In the stored plants, the recoverable per-
centage of eggs was approximately halved every 10 days, and although the
viability of eggs recovered was little reduced after 10 days, it was reduced
sharply by 30 days (Table 3). The stored eggs which hatched did so within
minutes of placement in water, suggesting that embryonation occurred dur-
ing storage and thus out of water. A negligible percentage of eggs could be
expected to survive a drought of 30 days under field conditions, though sur-
vival might be better at lower winter temperatures.
BEHAVIOR OF LARVAE AND PUPAE
Dissections of digestive tracts of late instar larvae revealed mainly
fragments of organic, particulate matter, with algae, fungal spores, and
pollen in lesser quantities. Density of larvae, but not eggs is dependent upon
plant size, suggesting density-dependent competition (Frank et al. 1977).
By food-dependent delay of development, especially in instar III, the larvae
spread the risk of starvation in time (Frank and Curtis 1977a).
If density of larvae should vary from one axil to another, it is relevant
to know whether the larvae are able to move from axil to axil and thus dis-
cover additional food. O'Connor (1923) wrote that the larva of the mos-
quito Aedes kochi (Doenitz), found in plant leaf axils in the Indo-Pacific
region, can leave the water and "creep like a caterpillar", accomplished by
special antennal structures. Peters (1956) reported that the larva of the
mosquito Uranotaenia ornata Theobald, which inhabits plant leaf axils in
Liberia, can crawl out of the water of a container, browse along under the
water film that follows it, then turn round and redescend into the container.
Although some insect larvae present in T. utriculata axils can climb out of
water containers (watch glasses, specimen vials), Wyeomyia larvae were not
observed to do this, and it seems improbable that they can move from one
axil to another unless the leaves forming the axils are damaged. However,
their ability to delay development according to food availability is surely an
adaptation to a relatively stable habitat with intermittent food supply. They
Florida Entomologist 64 (4)
24 48 72 96 120
TIME IN HOURS
Fig. 4. Histogram showing cumulative % of Wyeomyia vanduzeei eggs
hatching vs. time after oviposition, based on 30 groups each of 100 eggs at
Frank & Curtis: Mosquito & Its Nursery Plant
O-i t- 10
*^f Iz Oi
eo 6 o-
+1 +1 +1
o co- CO
00 t- CO
L0 <'d 00
CO 00 v
UO L6 00
+1 +1 +1
D M4 C
to 00 CO
cO 0 c1
T-I CI CO
+ +1 +1
1 0I 00
CO CO "di
o cd cd
C)l C CO
504 Florida Entomologist 64 (4) December, 1981
are thus adapted for the bromeliad axil habitat just as the locomotive abil-
ities of A. kochi and U. ornata adapt these larvae for other plant axils.
Wyeomyia larvae and pupae react to disturbance by diving, similarly to
larvae and pupae of other mosquitoes. Because of the shape of T. utriculata
axils (Fig. 1), the mosquitoes are hard to see. Because heavy rainfall con-
stitutes a disturbance, the diving reaction may enable larvae and pupae to
avoid being washed out of the plants (in contrast to eggs) (Frank and
PREDATORS AND PARASITES
The larger axils of bromeliads hold discrete bodies of water. Amphibious
predators, able to move from one axil to another, may be better adapted to
feed on mosquito larvae. In Florida, aquatic larvae of a muscid fly identified
as Neodexiopsis sp. (D. Fish, pers. comm.) are amphibious predators in T.
utriculata axils, but the predatory larvae of the mosquito Toxorhynchites
r. rutilus (Coquillett) are not amphibious. Tree frogs (Hyla) are sometimes
present in the larger plants (Neill 1951). In the neotropics, but unknown in
Florida, Dytiscidae (Balfour-Browne 1938) and Hydrophilidae (Champion
1913) are amphibious in bromeliads. Terrestrial predators may be able to
take advantage of stranded aquatic organisms during times of low water
level in bromeliad axils. Staphylinidae and Carabidae, well-represented in
bromeliads in the neotropics (Champion 1913, Zaragoza 1974), seem to be
absent from this habitat in southern Florida, but ants and spiders are
present and probably fulfill a similar role.
Pilosporella fishi Hazard and Oldacre, a microsporidan parasite of W.
vanduzeei larvae, seems too rare to have a significant effect on the mosquito
population (Frank and Curtis 1977c). The fungal parasite Coelomomyces
sp. also is rare (Hall and Anthony 1979).
We thank D. Fish, J. R. Linley, L. P. Lounibos, E. McCoy, G. F. O'Meara,
D. R. Strong and M. J. Way for critical comments on versions of this manu-
script. J. R. Rey kindly translated the abstract into Spanish. This is contri-
bution No. IX in a series entitled: On the biononics of bromeliad-inhabit-
BALFOUR-BROWNE, J. 1938. On two new species of bromeliadicolous Copelatus
(Col., Dytiscidae). Ent. Mon. Mag. 74: 100-2.
BENZING, D. H. 1970. Foliar permeability and the absorption of minerals
and organic nitrogen by certain tank bromeliads. Bot. Gaz. 131: 23-31.
-- 1973. The monocotyledons: their evolution and comparative biology.
1. Mineral nutrition and related phenomena in the Bromeliaceae and
Orchidaceae. Quart. Rev. Biol. 48: 277-90.
1977. Bromeliad trichomes: structure and function. J. Bromeliad
Soc. 27: 122-8.
J. A. DERR, AND J. E. TITUS. 1972. The water chemistry of micro-
cosms associated with the bromeliad Aechmea bracteata. American
Midl. Nat. 87: 60-70.
Frank & Curtis: Mosquito & Its Nursery Plant 505
---- AND A. RENFROW. 1971. The significance of photosynthetic efficiency
to habitat preference and phylogeny among tillandsioid bromeliads.
Bot. Gaz. 132: 19-30.
AND 1974. The mineral nutrition in the Bromeliaceae. Bot.
Gaz. 135: 281-8.
BUTLER, J. 1974. Pineapples of the treetops. Florida Nat. 47: 13-7.
CARLISLE, A. 1967. The nutrient content of tree stemflow and ground flora
litter and leachates in a sessile oak (Quercus petraea) woodland. J.
Ecol. 55: 615-27.
---, A. H. F. BROWN, AND E. J. WHITE. 1966. The organic and nutrient
elements in the precipitation beneath a sessile oak (Quercus petraea)
woodland. J. Ecol. 54: 87-98.
CHAMPION, G. C. 1913. Coleoptera, &c., in bromeliads. Ent. Mon. Mag. 49:
FRANK, J. H., AND G. A. CURTIS. 1977a. On the bionomics of bromeliad-
inhabiting mosquitoes. III. The probable strategy of larval feeding in
Wyeomyia vanduzeei and Wy. medioalbipes. Mosquito News 37: 200-6.
---- AND 1977b. On the bionomics of bromeliad-inhabiting mos-
quitoes. IV. Egg mortality of Wyeomyia vanduzeei caused by rainfall.
Mosquito News 37: 239-45.
-- AND 1977c. On the bionomics of bromeliad-inhabiting mos-
quitoes. VII. Incidence and effect of Pilosporella fishi, a parasite of
Wyeomyia vanduzeei. Mosquito News 37: 487-9.
--, G. W. ERDOS, AND E. A. ELLIS. 1981. On the bionomics of
bromeliad-inhabiting mosquitoes. VIII. The flotational structure of
Wyeomyia vanduzeei eggs. J. Med. Ent. 18: 337-40.
S- AND H. T. EVANS. 1976. On the bionomics of bromeliad-
inhabiting mosquitoes. I. Some factors influencing oviposition by
Wyeomyia vanduzeei. Mosquito News 36: 25-30.
S AND 1977. On the bionomics of bromeliad-inhabiting
mosquitoes. II. The relationship of bromeliad size to the number of
immature Wyeomyia vanduzeei and Wy. medioalbipes. Mosquito News
GEIGER, R. 1971. The climate near the ground. (Translation of 4th German
edition). Harvard University Press, Massachusetts.
HALL, D. W., AND D. W. ANTHONY. 1979. Coelomomyces sp. (Phycomycetes:
Blastocladiales) from the mosquito Wyeomyia vanduzeei (Diptera:
Culicidae). J. Med. Ent. 16: 84.
LAESSLE, A. M. 1961. A micro-limnological study of Jamaican bromeliads.
Ecology 24: 499-517.
LUGO, A. E., AND S. C. SNEDAKER. 1974. The ecology of mangroves. Annu.
Rev. Ecol. Syst. 5: 39-64.
MCWILLIAMS, E. L. 1970. Comparative rates of dark CO2 fixation and
acidification in the bromeliaceae, Euphorbiaceae and Orchidaceae. Bot.
Gaz. 131: 285-90.
NEILL, W. T. 1951. A bromeliad herpetofauna in Florida. Ecology 32: 140-3.
O'CONNOR, F. W. 1923. The life-history and habits of Finlaya kochi. Res.
Mem. London School Trop. Med. Hyg. 4: 56-7.
PETERS, W. 1956. The mosquitoes of Liberia. A general survey. Bull. Ent.
Res. 47: 525-51.
PITTENDRIGH, C. S. 1948. The bromeliad-Anopheles-malaria complex in
Trinidad. 1-The bromeliad flora. Evolution 2: 58-89.
POOL, D. J., A. E. LUGO, AND S. C. SNEDAKER. 1975. Litter production in
mangrove forests of southern Florida and Puerto Rico. Proc. Int.
Symp. Biol. Management Mangroves 1: 213-37.
PROvosT, M. W. 1974. Mean high water mark and use of tidelands in Flor-
506 Florida Entomologist 64(4) December, 1981
ida. Florida Sci. 36: 50-66.
RAUH, W., R. SCHILL, N. EHLER, AND W. BARTHLOTT. 1973. Some remarks
on the water supply of bromeliads. J. Bromeliad Soc. 12: 89-111.
REYNOLDS, E. R. C., AND L. LEYTON. 1963. Measurement and significance of
throughfall in forest stands. Pages 127-141 IN A. J. Rutter and
E. F. H. Whitehead (eds.). The water relations of plants. A Sym-
posium of the British Ecological Society. Wiley, New York.
SMALL, J. K. 1933. Manual of the southeastern flora. University of North
Carolina Press, Chapel Hill.
TRIMBLE, G. R., AND S. WEITZMAN. 1954. Effect of a hardwood forest canopy
on rainfall intensities. Trans. American Geophys. Union 35: 226-34.
TUKEY, H. R. 1970. Leaching of metabolites from foliage and its implica-
tion in the tropical rain forest. Pages H.155-H.160 IN H. T. Odum
and R. F. Pigeon (eds.). Tropical rainforest. Division of Technical
Information, United States Atomic Energy Commission.
VARGA, L. 1928. Ein interessanter Biotop der Biocbnose von Wasseror-
ganismen. Biol. Zentralbl. 48: 143-62.
VOIGT, G. K. 1960. Alteration of the composition of rainwater by trees.
American Midl. Nat. 63: 321-6.
ZARAGOZA, C., S. 1974. Coleopteros de algunas bromelias epifitas y doce
nuevos registros de species para la fauna Mexicana. An. Inst. Biol.
Univ. Nac. Auton. Mexico, Ser. Zool. 45: 111-8.
*^- -- --- -- --^-^--i --I-- --^
ACTIVITY RHYTHMS, INFLUENCE OF HOST PLANT
ON MATING AND OVIPOSITION, AND REARING
OF THE SOUTHERN PINE CONEWORM
CARL W. FATZINGER
USDA, Forest Service
Southeastern Forest Experiment Station
Olustee, FL 32072 USA
The southern pine coneworm, Dioryctria amatella (Hulst), (Lepidoptera:
Pyralidae) is nocturnal and reproductive behavior occurs only during the
last 5 h of scotophase (12L:12D photoperiod). Mating does not begin until
moths are 2 days old and mating and oviposition are dependent on the
presence of the host plant. Satisfactory rearing required surface steriliza-
tion of eggs with sodium hypochlorite, autoclaving the dietary wheat germ,
inclusion of sorbic acid in the larval rearing medium, and a larval rearing
density of no more than 2 per 30 ml container. The described methods have
been used to rear more than 40 successive generations of the insect on the
El gusano de pifi6n, Dioryctria amatella (Hulst), (Lepidoptera:
Pyralidae) es nocturno y su funcionamiento reproductive toma lugar
solamente durante las fltimas 5 horas del period de oscuridad (12L:12D)
fotoperiodo). El apareamiento no comienza hastaque las polillas tengan 2
Fatzinger: Southern Pine Coneworm 507
dias, y el apareamiento y la oviposici6n dependent de la presencia de la plant
hospedera. Una cria satisfactoria requiri6 la esterillizaci6n de la superficie de
los huevos con hipoclorito de sodio, germen de trigo esterilizado en autoclave,
inclusion de acido s6rbico en el medio de cria, y una densidad de cria larval
de no mas de 2 por cada envase de 30 ml. Los m6todos descritos han sido
usados para criar mas de 40 generaciones consecutivas de insects en el
medio de WGCS.
Larvae of the southern pine coneworm, Dioryctria amatella (Hulst)
damage southern pines by feeding on male and female strobili, on vegetative
tips, on strobili infected with the rust Cronartium strobilinum (Arth.)
Hedge. and Hahn, and on fusiform rust galls caused by Cronartium
quercuum (Berk.) Miyabe ex Shirai f. sp. fusiforme (Cumm.) Burds, et
Snow (Ebel 1963, 1965, Neunzig et al. 1964, Coulson and Franklin 1970,
Ebel et al. 1975). Larvae pass through 5 instars (Fatzinger 1970a) and
require from 3 to 5 weeks to develop to the pupal stage when reared at ca.
240 C on pine cones (Ebel 1965). The species is multivoltine and may have
from 1 to 4 generations per year in north Florida (Ebel 1965, Merkel and
Previous attempts to establish laboratory colonies were only partially
successful because of insufficient egg production between generations. Ebel
(1959) reared D. amatella from the egg to adult stage with procedures de-
veloped for Dioryctria ebeli Mutuura and Monroe (formerly D. abietella
(Denis & Schiffermiiller)), on waxed conelets of slash pine, Pinus elliottii
Engelm var. elliottii. DeBarr (1968) reared larvae of D. amatella on the
artificial medium developed by Fatzinger (1968) for D. ebeli but could not
obtain enough eggs for continuous culture. Merkel and Fatzinger (1966)
reared D. amatella on slash pine cones and obtained an average of only 30
eggs per female deposited on small fusiform rust galls cut from slash pine
branches. Ebel (1965) observed that only 16% of newly emerged females
held in mating cages deposited viable eggs during the first 2 days whereas
43% contributed by day 5. DeBarr (1968), however, found that only 10-20%
of the females mated in the laboratory whereas about 50% did so in an
The study reported here was conducted to optimize the rearing of D.
amatella by determining its temporal patterns of reproductive behavior,
requirements for host plant material to stimulate mating and oviposition,
tolerance of increased larval densities, reaction to surface sterilization of
eggs with sodium hypochlorite or peracetic acid, and acceptance of sterilized
MATERIALS AND METHODS
A laboratory colony of D. amatella was initiated in 1970 with larvae
collected from fusiform rust galls of slash pine and eggs from moths that
emerged from slash pine cones. The larvae were reared on the WGCS
medium developed for D. ebeli; constituents include wheat germ, casein, 5
sugars present in slash pine cones, salts, vitamins, 2 mold inhibitors, agar,
and water (Fatzinger 1970b). In a slight modification, the medium was
thickened by decreasing the water content 25% and the nonessential stach-
Florida Entomologist 64(4)
yose was eliminated. The hot (950 C) medium was dispensed into either 30
ml plastic condiment cups or sterilized glass test tubes and allowed to cool
(gell) prior to initiating hearings of first-stage larvae. The cups were filled
to a depth of 1 cm with medium. The test tubes were filled to a depth of 4
cm and gelled at a 300 angle to provide a thickness-dependent moisture
gradient which enabled larvae to select optimum conditions for feeding.
After the larvae were transferred, the tubes were stoppered with sterile,
nonabsorbent cotton and the cups were capped with plastic, snap-on lids. The
resulting pupae were separated by sex (Fatzinger 1968) and held on moist
paper toweling in wire screen cages (30 x 30 x 30 cm) until moths emerged.
The environment for all rearing and experimentation was 270 C, ca. 65%
RH, and 12L:12D photoperiod (0600 to 1800 photophase).
Locomotor activity, female calling, and mating behavior were recorded
automatically by time-lapse photography (Fatzinger 1973). Moths were
photographed through the glass doors of clear plastic cages at 30-min inter-
vals for 7 days.
Host material consisted of cross sections (ca. 6 mm thick x 4 cm dia.)
cut from a fusiform rust branch gall. Initially the behavior patterns of
26 9 moths and 9 pairs of moths that emerged from field-infested slash pine
cones were observed without host plant materials. Behavioral effects of
host plant material were observed with moths obtained from the first 15
successive generations reared. These included 9 $ observed without galls and
18 8, 20 9, and 14 pairs observed with galls. The general patterns of be-
havior were obtained by averaging the observations of all moths regardless
of their age. The times of the occurrence of behavioral events are reported
as means in hours and minutes angular deviations in tenths of hours
(P = 0.01). These means were obtained by the methods described by Bat-
schelet (1965) for circular normal distributions.
Over a 3-year period, 26 colonies of D. amatella were initiated to assess
the effect of larval rearing density on colony survival and duration of larval
stage. Tests included 11 colonies or 14,045 larvae reared 1 per container
(899 in test tubes and 13,146 in cups), 4 colonies or 764 larvae reared 2 per
container (140 larvae in test tubes and 624 larvae in cups), and 11 colonies
or 28,482 larvae reared 3 per container (4,776 in test tubes and 23,706 larvae
Eggs and constituents of the WGCS medium were individually plated on
sterilized malt agar and nutrient agar media in plastic petri dishes to de-
termine the sources of microbial contaminants. The dishes were held at
220 C and were observed for 2 weeks at 2-day intervals for the growth of
The efficacies of peracetic acid and sodium hypochlorite for sterilizing
the egg surfaces were compared. Eggs deposited on cheesecloth were washed
for 1 min in a detergent (0.1% sodium alkylarylsulfonate), submersed for
5 min in either 0.1% peracetic acid or 6.5% sodium hypochlorite solutions,
and rinsed twice in sterile water. Individual eggs were transferred to plates
of both culture media (30 plates of culture media containing 5 eggs each
The egg sterilization methods also were evaluated by rearing 100 larvae
that hatched from eggs treated with peracetic acid and 100 larvae from eggs
treated with sodium hypochlorite. These larvae were reared on the WGCS
Fatzinger: Southern Pine Coneworm 509
medium and the percentage of contaminated rearing containers, the average
percentage of surface contamination, and the survival percentage to the
pupal stage were compared between the treatments. Controls included 100
larvae hatched from untreated eggs and 100 rearing containers that held
WGCS medium but no larvae. The brush used to transfer larvae to rearing
containers was sterilized with Amphyl disinfectant after every third
Samples of the constituents of the WGCS medium were plated on 10
replicate dishes of each culture medium to evaluate this source of contamina-
tion. To reduce contamination in the WGCS medium, its ingredients were
treated in the following ways: (1) wheat germ was sterilized by autoclaving
at 15 psi for 30 min prior to mixing the medium, termed AUTOWG; (2)
0.25% sorbic acid was added to AUTOWG as an additional mold inhibitor,
termed AUTOWG+S; (3) the WGCS medium was autoclaved at 15 psi for
30 min, termed AUTOWGCS; (4) the wheat germ was sterilized over
propylene oxide for 24 h prior to mixing the medium, termed FUMWG; (5)
the wheat germ was sterilized with dry heat for 4 h at 1800 C prior to
mixing medium, termed DRYWG; and (6) the standard WGCS medium,
The initial generation of D. amatella reared on each experimental
medium was started with surface sterlized eggs collected from generations
27 through 34 of the laboratory colony. First-stage larvae were reared in-
dividually to the pupal stage in test tubes containing the experimental
media. The number of days until first visual evidence of contaminated media,
number of days to larval death, and duration of the larval stage were re-
Darkened pupae from each medium were held in individual 25 x 95 mm
vials that contained a piece of 1 x 6 cm screen and were fitted with a cork
stopper. The sex and weight of the pupae, and the duration of the pupal
stage were recorded for each individual. Samples of moths that emerged on
the same day from each experimental medium were transferred to wire
screen cages (30 x 30 x 30 cm) for mating and egg deposition. The moths
were provided water from a sponge held in an uncovered petri dish. The
number of cages observed for each medium depended on the timing of moth
emergence and the numbers of emerging moths. Egg deposition was recorded
in an average of 4 cages (range 1 to 6) per replicate; the grand average
was 14 cages (range 3 to 34) per treatment. The average number of moths
per cage was 4.8 (range 2 to 10) and the average sex factor (ratio of fe-
males to total) was 0.4 (range 0.2 to 0.6) per cage. Each cage contained 2
cross sections (ca. 1.5 cm thick x 4 cm dia.) cut from fusiform rust galls of
slash pine because host plant material was essential for mating and oviposi-
tion. Each cross section of gall was wrapped in 2 layers of cheesecloth upon
which moths deposited eggs. The galls were replaced every 2 days with
freshly cut cross sections and wrappers. The cheesecloth wrappers facilitated
removal of eggs from the galls and supported the eggs during surface
sterilization. Numbers of eggs deposited on the cheesecloths and numbers of
larvae hatching from each batch of eggs were determined daily. The genera-
tion time was calculated by summing the average duration of development
for female insects from egg to adult and the average preoviposition period.
Differences among treatment means were tested for statistical sig-
510 Florida Entomologist 64 (4) December, 1981
nificance by analysis of variance and Duncan's multiple range test, and are
reported as means standard errors. Prior to these analyses, data collected
as percentages were transformed to arc sin -proportion, but only the per-
centages are listed in the tables and results.
RESULTS AND DISCUSSION
BEHAVIOR OF MOTHS AND EFFECTS OF HOST PLANT MATERIAL
A t-test for paired observations revealed no significant difference (P>
0.01) between diel periodicities of locomotor activity of male and female
moths. Locomotor activity began at 1825 0.5 (N=165) and ended at 0532
1.7, i.e., it began about 25 min after dark and continued until ca. 28 min
prior to the onset of photophase. During each photophase, moths generally
remained motionless unless disturbed. The pattern was similar to that re-
ported for D. ebeli (Fatzinger 1973) and the presence or absence of a
fusiform rust gall had no effect. However, D. amatella calling, mating, and
oviposition did not occur in cages without fusiform rust galls (26 9 moths
and 9 pairs of moths). In contrast, D. ebeli readily mated and oviposited in
laboratory cages without host plant material (Ebel 1959, Merkel and
Fatzinger 1966, Fatzinger 1970b).
When the females were 2 days old, calling behavior began ca. 8 h after
dark and ended ca. 2.5 h later. Two- to 5-day-old virgin moths began to mate
ca. 9 h after dark and ceased to mate ca. an hour later (Fig. 1). These ob-
90 \- -CALLING
10 12 2 4 6 8 10 12 2 4 6 8
TIME DURING 12L: 12D PHOTO PERIOD (HR.)
Fig. 1. Periodicity of locomotor, calling, and mating activity of Dioryctria
Fatzinger: Southern Pine Coneworm
servations agree with the delayed mating reported by Ebel (1965). Multiple
mating at intervals of 1 to 5 days was observed in both males and females;
3 9 and 2 & mated twice and one pair mated 3 times.
EFFECT OF LARVAL REARING DENSITY
During this study the same quantity of food was placed in each rearing
container regardless of the number of larvae reared per container. The
average numbers of pupae recovered per container from hearings at different
larval densities were 0.58 0.04 pupae at 1 larva container, 1.25 0.14 pupae
at 2 larvae/container, and 0.91 0.10 pupae at 3 larvae/container. Survival
of pupae was the same at all densities, averaging 83.1-1.6%, and the dif-
ferences were due to larval mortality. Larval mortality at the highest
density was significantly greater than at the 2 lower densities (P<0.01).
Several partially eaten pupae were found, but cannibalism was rare.
Larval density did not significantly (P>0.01) influence the time required
to rear larvae from the first stage through pupation. The average durations
were 30.8 1.0, 26.11.8, and 28.1 0.5 days for larvae reared 1, 2, or 3
per container, respectively.
Of the 9,494 rearing containers containing 3 larvae per container, ca.
5% were heavily contaminated (95% of the medium surface was contami-
nated). Only ca. 1% of the rearing containers with 1-2 larvae per container
became heavily contaminated. The increased incidence of excessive contami-
nation for colonies reared with 3 larvae per container was probably due to
the greater chance of transferring larvae hatched from imperfectly disin-
fected eggs to a given container.
Dead larvae were found more frequently in excessively contaminated
rearing containers. Prior to death, many of these larvae remained in the
first or second stage while larvae in other containers progressed to the
fourth or fifth larval stage. The physical barrier formed by microorganisms
on the surface of the medium apparently caused the larvae to die from
starvation (Fatzinger 1970b, Bucher and Bracken 1976).
Some larvae that hatched from eggs oviposited in the fall and winter
remained in the first stage for as long as 66 days before dying (Fig. 2).
They readily crawled about if disturbed but remained in the first stage long
after the remainder of the colony pupated. In North Carolina, D. amatella
overwinter primarily as early-stage larvae (Neunzig et al. 1964). In lab-
oratory and field populations of D. amatella from north Florida, Ebel (1965)
found early-stage larvae that ceased active feeding in the fall and winter,
and either wandered about or remained inactive among cone scales or bark
crevices. The inactive first-stage larvae observed during hearings may have
been exhibiting the overwintering behavior of wild populations and they may
have died from starvation because the medium deteriorated.
SURFACE STERILIZATION OF EGGS
Eggs collected from the rearing cages were contaminated with both
bacteria and fungi. Significantly fewer eggs were contaminated (P<0.01)
after treatment with peracetic acid (33.7%) or sodium hypochorite (53.7%),
but treatment with peracetic acid was superior. However, equal production
of pupae was obtained from eggs surface-sterilized by either method (Table
Florida Entomologist 64(4)
MAXIMUM DURATION OF LARVAL STAGE
2 5 8 11 14 17 20 23 26 29 32 35 38 41 44 47 50 53 56 59 62 65
NUMBER OF DAYS IN REARING (CLASS MIDPOINT)
Fig. 2. Number of days Dioryctria amatella larvae lived after hatching
during different months and life spans of those that died after being placed
on artificial media.
1) and that production was higher than that from untreated eggs. Con-
tamination of rearing containers held without larvae was minimal.
DECONTAMINATION OF THE LARVAL REARING MEDIUM
Wheat germ and casein were the primary sources of contaminants in the
WGCS medium. All the plates containing wheat germ showed visible growth
of microorganisms but only 25% of the plates containing casein showed
TABLE 1. EFFECT OF SURFACE STERILIZATION OF EGGS ON CONTAMINATION OF
REARING CONTAINERS AND SURVIVAL OF Dioryctria amatella REARED
ON THE WGCS MEDIUM.
Contaminated Average surface Survival to
rearing area of medium pupal
Treatment containers* contaminated** stage*
Peracetic acid 39 13.2 a 80
Sodium hypochlorite 55 19.6 a 82
Non-sterilized eggs 84 41.5 b 52
WGCS medium only 22 0.8 c -
*A X2 test for independence showed significant differences among treatments at the 1%
**Visually estimated to the nearest percent in contaminated rearing containers. Any 2
means followed by the same letter are not significantly different at the 1% probability level.
Fatzinger: Southern Pine Coneworm 513
contaminants to be present. Therefore, the 5 modifications of the WGCS
medium were designed primarily to decrease contamination originating
from wheat germ. The elapsed time for visual evidence of surface contami-
nation did not differ significantly among media (Table 2). There were some
significant differences in percentages of rearing containers that became con-
taminated and in survival from the larval to the adult stages. The lowest
incidence of contaminated rearing containers and the highest survival rate
of larvae was associated with the AUTOWG+S medium. Although the
AUTOWG+S medium was not significantly different from the AUTOWGCS,
AUTOWG, or FUMWG media, it was the only medium tested that pro-
duced significantly more moths than the CHECK medium. Dry heat steriliza-
tion of wheat germ was the least effective method.
The type of medium did not significantly influence the duration of in-
dividual stages or the generation time which averaged 47.9 + 1.1 days (Table
2). The average weights of pupae, however, were significantly different
among some of the media. The number of eggs deposited by individual fe-
males (average 95.4-12.6, N=176) and percentage hatch (47.5-3.4, N=
16,725 eggs) were not significantly (P>0.01) affected by the media.
Moths of D. amatella were nocturnally active during 12L:12D photo-
periods. Female calling behavior and subsequent mating and oviposition
were dependent on the presence of host plant material. Mating behavior was
delayed for 2 days after emergence and occurred during the last 5 h of
scotophase. Multiple matings occurred, but were infrequent.
Survival of D. amatella larvae was optimized when 2 larvae were reared
per 30 ml container and microbial contaminants were eliminated. Thus,
TABLE 2. SURVIVAL FROM FIRST-STAGE LARVAE TO ADULTS, CONTAMINATION
OF REARING CONTAINERS, PUPAL WEIGHTS, AND GENERATION TIMES
OF Dioryctria amatella, REARED ON MODIFIED WGCS MEDIUM.*
Average Avg. No.
percent days to
survival % of first Pupal Generation
to adult rearing visible weight time
Medium** stage containers growth (mg) (days)t
AUTOWG+S 83.5 a 3.0 a 17.2 a 94.7 a 48.5 4.4
AUTOWGCS 70.6 a b 21.7 a b 6.1 a 91.5 a 44.0 + 1.9
AUTOWG 61.6 a b 28.3 a b c 15.3 a 88.0 a 47.5 1.5
FUMWG 45.7 a b 43.7 a b c 10.4 a 86.7 a 50.8 3.2
CHECK 42.5 b 64.5 bc 15.2 a 77.2 b 50.8 3.3
DRYWG 28.5 b 81.5 c 12.7 a 79.8 b 48.7 +9.9
*Any 2 means followed by the same letter are not significantly different at the 1% prob-
ability level by Duncan's multiple range test (N=150-652 larvae, 2-8 replicates).
**AUTOWG+S=Wheat germ autoclaved plus sorbic acid, AUTOWGS=entire medium
autoclaved, AUTOWG=wheat germ autoclaved, FUMWG= wheat germ fumigated, DRYWG
= wheat germ dry heat sterilized.
Florida Entomologist 64(4)
semiaseptic techniques should be used when preparing the medium and trans-
ferring larvae to rearing containers. Peracetic acid or sodium hypochlorite
were equally effective for surface sterilizing eggs; however, sodium hy-
pochlorite is less flammable, easier to store, and more readily available.
Sorbic acid should be added to the medium and the wheat germ autoclaved.
I thank N. C. Leppla of the Insect Attractants, Behavior, and Basic
Biology Research Laboratory; F. M. Davis of the Delta States Area Crop
Science and Engineering Research Laboratory; and E. W. Clark, G. L.
DeBarr, G. F. Fedde, and J. C. Nord of the Southeastern Forest Experiment
Station for their helpful suggestions and critical reviews of this paper.
Mention of a proprietary product does not imply its endorsement by the
U.S. Dept. of Agriculture.
BATSCHELET, E. 1965. Statistical methods for the analysis of problems in
animal orientation and certain biological rhythms. American Inst.
Biol. Sci. Monogr. 57 p.
BUCHER, G. E., AND G. K. BRACKEN. 1976. The bertha armyworm,
Mamestra configurata (Lepidoptera: Noctuidae). Artifical diet and
rearing technique. Canadian Ent. 108: 1327-38.
COULSON, R. N., AND R. T. FRANKLIN. 1970. The biology of Dioryctria
amatella (Lepidoptera: Phycitidae). Canadian Ent. 102: 679-84.
DEBARR, G. L. 1968. A study of methods for the continuous laboratory rear-
ing of Dioryctria amatella (Hulst). Unpublished Progress Rep. FS-
SE-2203 (Rev.) -1-6, on file at Southeastern Forest Experiment Sta-
tion, Asheville, NC.
EBEL, B. H. 1959. Laboratory rearing of a pine cone insect, Dioryctria
abietella (D. & S.). J. Econ. Ent. 52: 561-4.
- 1963. Insects affecting seed production of slash and longleaf pines-
their identification and biological annotation. United States Dept.
Agric. For. Serv. Res. Pap. SE-6, 24 p. Southeastern Forest Experi-
ment Station, Asheville, NC.
1965. The Dioryctria coneworms of north Florida pines (Lepidop-
tera: Phycitidae). Ann. Ent. Soc. America 58: 623-30.
T. H. FLAVELL, L. E. DRAKE, H. O. YATES III, AND G. L. DEBARR.
1975. Seed and cone insects of southern pines. United States Dep.
Agric. For. Serv. Gen. Tech. Rep. SE-8, 40 p. Southeastern Forest
Experiment Station, Asheville, NC, and Southeastern Area, State and
Priv. For., Atlanta, GA.
FATZINGER, C. W. 1968. Rearing successive generations of Dioryctria
abietella (D. & S.) (Lepidoptera: Phycitidae) on artificial media
with aspects on nutrition of the insect. Ph.D. Thesis, North Carolina
State Univ., Raleigh, 178 p.
1970a. The number of instars of 'Dioryctria amatella (Hulst) in
north Florida. United States Dep. Agric. For. Serv. Res. Note SE-142,
3 p. Southeastern Forest Experiment Station, Asheville, NC.
1970b. Rearing successive generations of Dioryctria abietella
(Lepidoptera: Phycitidae) on artificial media. Ann. Ent. Soc. Amer-
ica 63: 809-14.
- 1973. Circadian rhythmicity of sex pheromone release by Dioryctria
abietella (Lepidoptera: Pyralidae (Phycitidae)) and the effect of a
Fatzinger: Southern Pine Coneworm 515
diel light cycle on its precopulatory behavior. Ann. Ent. Soc. America
MERKEL, E. P., AND C. W. FATZINGER. 1966. Coneworms. Pages 451-60 in
C. N. Smith (ed.), Insect Colonization and Mass Production. Academic
Press, Inc., New York, NY.
MERKEL, E. P., AND C. W. FATZINGER. 1971. Periodic abundance of pine
cone-infesting Lepidoptera in black light traps and sleeve cages in
north Florida. Florida Ent. 54: 53-61.
NEUNZIG, H. H., E. D. CASHATT, AND G. A. MATUZA. 1964. Observations on
the biology of four species of Dioryctria in North Carolina (Lepidop-
tera: Phycitidae). Ann. Ent. Soc. America 57: 317-21.
EFFECTS OF 'ANTIGUA 2D-118' RESISTANT CORN ON
FALL ARMYWORM1 FEEDING AND SURVIVAL2,3
B. R. WISEMAN, N. W. WIDSTROM, AND W. W. MCMILLIAN4
The impact of a fall armyworm resistant corn, 'Antigua 2D-118', on larval
numbers was measured for 3 years in field studies. The susceptible standard,
'Cacahuacintle Crosses' (Cac.C) had ca. 2 times as many fall armyworms as
did 'Antigua 2D-118' 10 days after infestation. The resistance of 'Antigua
2D-118' was expressed in fewer larvae per plant and less leaf damage than
the susceptible entry for all 3 years.
En un studio de campo de 3 afios se evaluaron el efecto sobre el nimero
de larvas de una linea de maiz, 'Antigua 2D-118', la cual es resistente a
Heliothis zea (Boddie). Diez dias despues de la infestaci6n, el standard
susceptible, Cacahuacintle Crosses (Cac. C.) tenia cerca de double el nimero
de larvas que tenia Antigua 2D-112. En, cada uno de los 3 afios, la resistencia
de Antigua 2D-118 se media en menos larvas por plant y menos dafios a las
hojas que en el standard.
Plant resistance to insects has been reported as an ideal control method
for insect pests (Luginbill 1969). Dahms (1972) showed the role of plant
resistance in an integrated control program by demonstrating the direct and
indirect effects of all 3 mechanisms of resistance (nonpreference, anti-
biosis, and tolerance) and he gave examples of how resistant varieties re-
duced insect populations. Schalk and Ratcliff (1976) reported an evaluation
of programs of the Science and Education Administration (USDA) whereby
the use of insect-resistant cultivars has been successful in controlling insect
pests and in reducing the use of insecticides. Wiseman et al. (1978) found
"In cooperation with the University of Georgia College of Agriculture Experiment Stations,
Coastal Plain Station, Tifton, GA 31793. Accepted for publication 8 September 1981.
"Mention of a proprietary product does not constitute endorsement by the USDA.
4Southern Grain Insects Research Laboratory, ARS, USDA, Tifton, GA 31793 USA.
516 Florida Entomologist 64 (4) December, 1981
that when corn earworms, Heliothis zea (Boddie), were placed on silks of
resistant corn, only one-fourth as many emerged as compared to a susceptible
Increased emphasis on fall armyworm, Spodoptera frugiperda (J. E.
Smith), plant resistance research has occurred since the development of
artificial infestation procedures (Wiseman and Davis 1979, Wiseman et al.
1980, Wiseman and Widstrom 1980). Germplasm resistant to the fall army-
worm has been discovered (Wiseman et al. 1966, Wiseman et al. 1976, Davis
et al. 1978). In a recent study, the nonpreference and antibiosis mechanisms
of resistance were both found in selected resistant corn entries (Wiseman
et al. 1981). This paper reports on a 3-year study of the effects of a resistant
corn on fall armyworm numbers.
MATERIALS AND METHODS
'Antigua 2D-118' and 'Cacahuacintle Crosses' (Cac.C), representing re-
sistant (R) and susceptible (S), respectively, to fall armyworm leaf feed-
ing, were planted on 10 April 1978, 9 April 1979, and 16 April 1980. Tests
each year were arranged in a split-plot design for each of 10 replications.
Whole plots (6.1 m rows 0.76 m apart) were days larvae were counted after
initial infestation and subplots were the corn entries. Plants within treat-
ments were infested at ca. the 10-leaf stage of plant development using the
modified bazooka (Wiseman et al. 1980). First instars were dispensed in a
mixture of corncob grits (#2040 grit-o'-cobs@) at the rate of 20 larvae/
plant in 0.2 ml of grits. Larval counts were made on 10 individual plants at
6, 8, 10, and 12 days after infestation in 1978 and 4, 6, 8, and 10 days after
infestation in 1979 and 1980. Damage ratings were made at 15 days after
infestation using a visual damage rating scale of 0-9 where 0 = no damage
and 9 = plant perishes. Analyses of the data were made for each year and
Duncan's multiple range test was used to separate means.
RESULTS AND DISCUSSION
For 2 of the 3 years, test numbers of fall armyworm larvae declined by
at least 50% from initial infestation until 10-12 days after infestation
(Table 1). Significant differences were found among means in 1978 for 6, 8,
TABLE 1. COMBINED MEAN NUMBERS OF FALL ARMYWORM LARVAE/PLANT AT
4, 6, 8, 10, OR 12 DAYS AFTER INFESTATION. 1978-80.
Days after Mean number of larvae*
infestation 1978 1979 1980
4 8.3 a 13.0 a
6 2.4 a 6.7 b 9.0 b
8 1.6 b 6.4 b 6.9 c
10 1.2 c 4.4 c 4.5 d
12 1.0 c -
Means within a column for a given year followed by the same letter are not significantly
different at P = 0.05. Combined means are averages of the number of larvae over both
resistant and susceptible entries at days after infestation.
Wiseman et al.: Corn Resistant to FA W 517
and 10 days after infestation; but no differences were detected between the
mean number of larvae at 10 and 12 days after infestation. Since this was
the case, larval numbers were counted beginning at 4 days past initial in-
festation in 1979-80 instead of 6 days. Thus, we could better understand the
earlier effects of the resistant vs the susceptible on the fall armyworm. In
both years, there was a significant decline in the number of larvae found at
6 days compared with 4 days. Generally, during the 1978 testing period, no
rainfall occurred in and around the period of initial infestation. However,
for both 1979 and 1980, rainfall occurred throughout the testing period.
Rainfall occurring immediately following the initial infestation may be a
contributory factor in the higher establishment in both 1979 and 1980 than
The resistance of 'Antigua 2D-118' to the fall armyworm relative to
'Cac.C' was maintained throughout the testing years (Table 2), even though
the number of larvae/plant was almost twice as much in 1980 for 'Antigua
2D-118' as compared to 1979. Damage ratings also supported the fact that
'Antigua 2D-118' maintained its resistance; it was significantly less damaged
in all cases than 'Cac.C.'
The apparent differences in the ratios of larvae to plant damage ratings
for 1978 and 1979-80 may be because initial larval counts were not made
until the fourth day after infestation in 1978 and/or that rainfall during
the periods of infestation for 1979-80 reduced adverse environmental in-
fluences, which permitted a higher larval establishment. Thus, more larvae
were found at each recording period for 1979-80.
The effects of the 2 corn entries on fall armyworm numbers at 4, 6, 8,
10, or 12 days after initial infestation of 20 larvae/plant are shown in Table
3. In 1978, 'Antigua 2D-118' had significantly fewer larvae than 'Cac.C' at
6, 8, and 10 days after infestation, but not at 12 days, even though 'Cac.C'
had almost twice as many larvae per plant (Table 3). Ratios of larvae per
plant on the resistant vs the susceptible for 1978 were: 1:3.4 at 6 days,
1:3.0 at 8 days, 1:3.0 at 10 days, and 1:1.8 at 12 days for an overall ratio
for 1978 of 1:2.9. In 1979-80, 'Antigua 2D-118' had significantly fewer larvae
at every count as compared with 'Cac.C.' Ratios of larvae per plant on the
resistant vs the susceptible for 1979 were: 1:1.9 at 4 days, 1:3.2 at 6 days,
1:3.4 at 8 days, and 1:3.4 at 10 days, with an overall ratio of 1:2.7 larvae
for the resistant vs the susceptible. Ratios of larvae per plant for the re-
sistant vs the susceptible for 1980 were: 1:1.3 at 4 days, 1:1.8 at 6 days,
1:1.9 at 8 days, and 1:1.8 at 10 days, with an overall ratio of 1:1.6. There-
TABLE 2. MEAN NUMBER OF FALL ARMYWORM LARVAE AND MEAN DAMAGE
RATING FOR A RESISTANT 'ANTIGUA 2D-118' AND A SUSCEPTIBLE
No. larvae/plant Mean damage rating
Entry '78 '79 '80 '78 '79 '80
'Antigua 2D-118' 0.8 a 3.5 a 6.4 a 1.4 a 2.7 a 4.0 a
'Cac.C' 2.3 b 9.4 b 10.3 b 2.5 b 5.5 b 5.8 b
*Means within a column for a given year followed by the same letter are not significantly
different at P = 0.05.
Florida Entomologist 64(4)
TABLE 3. EFFECTS OF A RESISTANT AND SUSCEPTIBLE CORN ON FALL ARMY-
WORM NUMBERS. 1978-80.
Number of larvae/plant*
Days after 1978 1979 1980
infestation A C A C A C
4 5.8 a 10.8 a 11.3 a 14.7a
6 1.1 a 3.7 a 3.2 b 10.1 b 6.4 b 11.6 b
8 0.8 a 2.4 b 2.9b 9.8 b 4.7bc 9.0 c
10 0.6 a 1.8 bc 2.0 c 6.8 c 3.2 c 5.7 d
12 0.7a n.s. 1.3 c -
*Means within a column for a given year and entry followed by the same letter are not
significantly different at P = 0.05. Horizontal means within a given year with between
the corn entries are significantly different at P = 0.05. A = resistant 'Antigua 2D-118' and
C = susceptible 'Cac.C.'
fore, for the 3-yr testing period, 'Cac.C' had ca. twice as many fall army-
worms per plant as 'Antigua 2D-118.'
Wiseman et al. (1981) found in laboratory tests that 'Antigua 2D-118'
only had a slight level of antibiosis and that most of its resistance was due
to the mechanism of larval nonpreference. Thus, the nonpreference type of
resistance of 'Antigua 2D-118' can impact detrimentally on fall armyworm
We thank J. L. Skinner of this laboratory for his assistance in this study.
DAHMS, R. G. 1972. The role of host plant resistance in integrated insect
control. Pages 152-67 in M. G. Jotwani and W. R. Young (eds.). Con-
trol of the sorghum shootfly. Oxford and IBN Pub. Co., New Delhi.
DAVIS, F. M., W. P. WILLIAMS, AND G. E. SCOTT. 1978. Development of corn
genotypes resistant to the southwestern corn borer and fall army-
worm. Mississippi Bus. Rev. 39: 9-11.
LUGINBILL, P., JR. 1969. Developing resistant plants-the ideal method of
controlling insects. USDA Prod. Res. Rep. No. 111. 14 p.
SCHALK, J. M., AND R. H. RATCLIFF. 1976. Evaluation of ARS program on
alternative methods of insect control: Host plant resistance to insects.
Bull Ent. Soc. America 22: 7-10.
WISEMAN, B. R., AND F. M. DAVIS. 1979. Plant resistance to the fall army-
worm. Florida Ent. 62: 123-30.
F. M. DAVIS, AND J. E. CAMPBELL. 1080. Mechanical infestation
device used in fall armyworm plant resistance programs. Florida Ent.
W. W. MCMILLIAN, AND N. W. WIDSTROM. 1978. Potential of re-
sistant corn to reduce corn earworm production. Florida Ent. 62: 92.
R. H. PAINTER, AND C. E. WASSOM. 1966. Detecting corn seedling
differences in the greenhouse by visual classification of damage by the
fall armyworm. J. Econ. Ent. 59: 1211-4.
AND N. W. WIDSTROM. 1980. Comparison of methods of infesting
Wiseman et al.: Corn Resistant to FAW
whorl-stage corn with fall armyworm larvae. J. Econ. Ent. 73: 440-2.
--- W. W. MCMILLIAN, AND W. D. PERKINS. 1976. Greenhouse
evaluations of leaf feeding resistance in corn to the corn earworm.
J. Georgia Ent. Soc. 11: 63-7.
---, W. P. WILLIAMS, AND F. M. DAVIS. 1981. Fall armyworm: Re-
sistance mechanisms in selected corns. J. Econ. Ent. 74: (In press).
COMMUNITY COMPOSITION AND ZOOGEOGRAPHY OF
THE INVERTEBRATE CAVE FAUNA OF BARBADOS
STEWART B. PECK
Department of Biology, Carleton University
Ottawa, Ontario KlS 5B6, Canada
A survey of the macroscopic invertebrate fauna of Coles Cave, the largest
undisturbed cave in Barbados, found a community of at least 44 non-
accidental species. Most of the species are supported by scavenging on bat
guano but at least 7 are predators and 1 is a parasitoid. The species richness
is mostly terrestrial, for only 7 species were found in the aquatic stream
community. Only a milliped seems to be cave-restricted, and few of the
species are endemic to Barbados. Three have been introduced by man. The
community shows little apparent niche separation and is generally composed
of species that are normally forest litter-inhabitants. They mostly arrived
at the island of Barbados by overwater dispersal since its emergence in the
Pleistocene, and then moved to the cave habitat via the forest understory.
La cueva de Cole es la mas grande sin molestar en la isla de Barbados.
La cueva tiene una comunidad, por lo menos, de 44 species que viven alli
naturalmente. Algunos de las species viven por comiendo la guano de los
murcielagos, pero 7 son de rapifia y una es parasito. Lo mas de las species
son terrestre y 7 son acu6tico. Solamente un diplopodo aparece limitado a
la cueva, y unas de las species son end6mico a Barbados. Hay tres que han
introduicido por hombre. La comunidad generalmente esta formado de
species que normalmente viven en la tierra vegetal de bosques. Las species
han dispersado sobre el mar desde la isla surgi6 en la period Pleistoceno, y
entonces pasaron a la cueva via la tierra vegetal.
Caves can serve as models for testing evolutionary, ecological and bio-
geographic hypotheses (Culver 1970, 1971; Culver et al. 1974; Vuilleumier
1973) but this has been attempted mostly for temperate zone caves. Data are
not yet adequate for such analyses for tropical zone caves, although they are
rapidly becoming available for the islands of the Caribbean (Peck 1974,
1975, 1977, 1981; Peck and Peck 1981; Orghidan et al. 1977; Taboada 1974).
This paper is offered as a data base on the invertebrate cave fauna of
Barbados, in the Lesser Antilles, as a step towards the formation of gen-
eralizations on the distributional and ecological characteristics of Neo-
Florida Entomologist 64(4)
tropical cave communities. It is part 8 in a series on the invertebrate fauna
of tropical American caves.
BARBADOS. This is a relatively small island, 32 km long and 23 km across,
with a maximum elevation of 370 m. It lies in the tropical Atlantic about
150 km east of the Windward Islands of the Lesser Antilles Island Arc
(Fig. 1 inset). It has a sub-humid to humid maritime tropical climate. The
island is situated on a north-south ridge of sediments uplifted at the eastern
margin of the Caribbean plate as it overrides the Americas Plate. The island
is unusual for the Lesser Antilles in that it is non-volcanic, and 6/7 of its
area is mantled by a succession of concentric Pleistocene reef terraces.
These have been isotope dated (Bender et al. 1973, Matthews 1973) and show
a correlation of increasing age with elevation, as the island was uplifted. A
conclusion is that land for overwater terrestrial biotic colonization has only
been available since the early Pleistocene. This gives a valuable reference
point in time by which to evaluate the dispersal and colonization potential of
M ... / COLES CAVE
0 25 50 75 100 COLES CAVE
S 100 200 300
SBARBADOS, W. I.
SURVEYED MARCH 1979 WITH BRUNTON
COMPASS AND STEEL TAPE BY M BUCK
,. E. SALM, K. TOEWS, AND D ROUTLIFFE,
DRAWN BY M. BUCK.
Fig. 1. Map of Coles Cave. Insets show position of Barbados (arrow) in
the tropical Atlantic, east of the Lesser Antilles Island Arc, and the position
of the cave on the island (dot).
Peck: Invertebrate Cave Fauna
different taxa (Peck and Peck 1980). These upraised reefs have been a site
for studies of karst development (Fermor 1972, 1977).
CAVES. Many small caves exist in the loosely consolidated faces of inland
and seaside coralline cliffs and gully walls, but few go into a zone of com-
plete darkness. Animal Flower Cave, on the NE coast, is a sea cave tourist
attraction with sea anemones. Several caves have been used as a water
source by the Government Waterworks Department, such as Harrison Cave,
Baker's Cave, and Cole's Cave. This is no longer true for Cole's Cave. Harri-
son Cave, near Welchman Hall, is now being developed as an attraction for
the tourist industry.
Cole's Cave is the largest and least disturbed cave in Barbados. It is
located in a gully bottom between Walkes Spring Plantation and Welchman
Hall near the center of the island, grid coordinates F.4, 3.7, on the 1:50,000
Barbados Topographic Map. It is reached by a dirt road entering the sugar
cane fields to the southwest, from a main road just south of Clifton Hill
Chapel. When the field to the north ends because of the steep lip of the
gully, a foot path goes down into the gully, and into the climb-down sink
(swallow hole) entrance of the cave. The cave is at approximately 200 m
elev., under the approximately 240 m high coral terrace of early or mid
The cave is essentially a medium diameter (walking) stream passage
about 1000 m in length. The stream generally flows to the southwest, on
coral rubble (Fig. 1). Two side chambers have large piles of guano pro-
duced by the bat Brachyphylla cavernarum minor Miller (Swanepoel and
Genoways 1978). The cave was once gated to exclude bats because they
were thought to be a health hazard. The upstream entrance is still closed
against bats. The fauna was most abundant in the high and narrow tunnel
descending from the main dome-pit entrances and at the guano. Fauna in
the stream was only seen in guano-laden pools beneath the bat roosts. Car-
rion baits in the stream drew no crustacean fauna. The air and water tem-
perature in the upstream half was 230C and the air at the end of the smaller
diameter downstream passage (which floods significantly) was 250C.
The macroscopic fauna of Coles Cave was surveyed in 3 trips to the cave
in February, 1979, while studying the forest litter invertebrates in Barbados
(Peck and Peck 1980). No synthetic work on the terrestrial or freshwater
invertebrate faunas is available for Barbados as it is for some other islands
(Harrison and Rankin 1976). Subterranean groundwaters of Barbados were
sampled for crustaceans in 1979 by Jan Stock, who found the interesting
Typhlatya cave shrimp (pers. comm.), which will be reported elsewhere. The
following list reports the trophic role of the species, and the level of cave
restriction (troglophile indicates that the species can exist as reproducing
populations in caves but is not limited to them). Information on the geo-
graphical distribution is given wherever possible.
ANNOTATED FAUNAL LIST
Florida Entomologist 64(4)
Dugesia sp. (?).1 Guanophage. Troglophile. These 0.75 cm long flat-
worms were on submerged guano.
Undetermined family, genus, and species. Guanophage. Troglophile.
Very small, pale, eyeless, oval flatworms were common on sub-
Genus and species undetermined. Guanophage. Troglophile. Small
pot-worms were abundant in soil and guano. The species are
Aeolosoma sp. Guanophage. Troglophile. Abundant on submerged
guano. Worms in this genus are often found in highly polluted
waters. The species are usually widespread.
Tubifex sp. Guanophage. Troglophile. Abundant on submerged
guano. Worms in this genus are often found in highly polluted
waters. The species are usually widespread.
Subulina octona (Bruquiere). Guanophage. Troglophile. Only dead
shells were found in the entrance passage. This very adaptable
species is widespread through the world tropics and was prob-
ably spread through commerce. It is frequent in Caribbean caves.
Retinella sp. (?). Guanophage. Troglophile. This low spired, pale,
and thin shelled snail occurred at the edge of guano piles.
Physa sp. Guanophage. Troglophile. This genus of snail is also
known from a guano ladden cave stream in Jamaica (Peck
Trichorhina tomentosa Budde-Lund (?). G. A. Schultz det. Guano-
phage. Troglophile. The specimens all seem to be females, and
parthenogenesis is common in the genus. These were exception-
ally abundant in guano. The species is reported from Venezuela,
and the genus occurs elsewhere in the Antilles and South Amer-
'The question marks indicate determinations that cannot be verified by a specialist for a
variety of reasons.
Peck: Invertebrate Cave Fauna
Mesocyclops sp. (?). Guanophage. Troglophile. Copepods occasion-
ally occur in guano-rich streams in caves in Puerto Rico and
Jamaica. The species are usually widespread.
Euryopis sp. One immature female, W. A. Gertsch det. Predator.
Genus and species undetermined. Three juveniles, W. A. Gertsch
det. Predator. Troglophile.
Family, genus, and species undetermined. Guanophages and predators.
Troglophiles. This is probably a rich fauna, judging by the guano
mite faunas known from Puerto Rico and Mona Island (32 and 17
Schizomus spp. J. Reddell det. Predators. Troglophiles. Two unde-
scribed species are represented by males and females. It is inter-
esting that the widespread species, S. puertoricensis (Chamber-
lin), which is often found in Caribbean caves, is not present
here (see Rowland and Reddell 1980).
Phrynus barbadensis (Pocock). Predator. Troglophile. This is also
recorded from Trinidad. These are usually predators on crickets
in caves. They are abundant in Neotropical lowland caves (Mul-
Undetermined genus and species. W. A. Shear det. Guanophage.
Troglobite (?). This seemingly new and cave-limited genus and
species was common on guano.
Eurhinocricus sp. W. A. Shear det. Guanophage. Troglophile. These
very large millipeds were exceedingly common in the entrance
tunnel, in the quano chambers, and elsewhere. The genus also
occurs in Mexico and Central America.
Scutigerella sp. Guanophage. Troglophile. Found in soil at edges of
guano piles. These omnivores are generally soil inhabitants, and
several species have been widely distributed by man.
524 Florida Entomologist 64 (4) December, 1981
Cyphoderus sp. K. Christiansen det. Guanophage. Troglophile.
Heteromurus sp. K. Christiansen det. Guanophage. Troglophile.
Dicyrtoma sp. K. Christiansen det. Guanophage. Troglophile.
Brachystomella sp. K. Christiansen det. Guanophage. Troglophile.
While collembola are frequent cave inhabitants, the above genera
are not commonly encountered in Caribbean caves.
Nicoletia sp. (?). Guanophage. Troglophile. All members of this
family are blind scavengers of hypogean habitats. Some are
known only from caves, but others have been spread by man.
Lepidocampa (Lepidocampa) sp. L. Ferguson det. Guanophage.
Troglophile. These are frequent soil inhabitants, and species in
this genus are widespread. This seems to be the first record
from a cave.
Amphiacusta sp. Guanophage. Troglophile. The genus occurs on
many Caribbean islands, and often forms large populations in
Periplaneta americana (L.) Guanophage. Troglophile. This cos-
mopolitan species is native to tropical Africa, and is common in
Caribbean guano caves.
Amnestus pusio (Stal). R. C. Froeschner det. Guanophage (?).
Troglophile. The species is widespread in the tropical Americas.
Members of the genus are frequent tropical cave inhabitants
Microvelia mimulus White. R. C. Froeschner det. Predator. Trog-
glophile. These may prey on micro-crustaceans in (or collem-
bola on) the water. They occur occasionally in tropical cave
streams. The species is widespread through the West Indies, and
Central and South America (Drake and Hussey 1955).
Genus and species undetermined. Guanophage. Troglophile. Adults
and larvae were in guano, which is a common habitat for them
in the Neotropics.
Oosternum costatum Sharp. A. Smetana det. Guanophage. Trog-
lophile. Abundant in guano. The species is widely distributed,
Peck: Invertebrate Cave Fauna 525
ranging from South America, through the West Indies and
Central America, to the southeastern United States (Smetana
Aleocharinae, genera and species undetermined. Predators. Trog-
lophiles. Two genera were present. These are frequent in guano,
probably preying on mites.
Ataenius luteomargo Chapin. H. F. Howden det. Guanophage.
Accidental (?). The species is endemic to the Caribbean, being
known from the 4 Greater and many of the Lesser Antilles
(Cartwright and Chalumeau 1978).
Pseudodiranchis sp. C. Yoshimoto det. Parasitoid. Troglophile.
These are parasitic on larvae and pupae of flies, undoubtedly
those in the guano.
Gnamptogenys sp., striatulaa complex). P. Ward det. Predator (?).
Troglophile. This large, black ponerine is native to the Antilles.
Hypoponera sp. P. Ward det. Predator (?). Troglophile. The genus
is known from other Caribbean caves.
Wasmannia auropunctata Wheeler. P. Ward det. Predator (?).
Troglophile. This tiny ant is an introduced tramp species of
Myrmecinae, genus and species undetermined. Predator (?). Trog-
Undetermined genus and species. Guanophages. Troglophiles. Four
species were present in the guano. These are common cave
Undetermined genus and species. Guanophage. Troglophile. Mem-
bers of this family are commonly found in moist guano.
At least 44 non-accidental species of invertebrates occupied Coles Cave.
More will be added if determinations become available on the mites. The
species are mostly guanophages and are supported directly by their feeding
on the abundant food source of bat guano. The diversity is mostly in the
terrestrial environment (only 7 species are aquatic but more microscopic
protistans are probably present). Predaceous species are 2 spiders, 2 schiz-
omids, the amplypygid, 2 staphylinid beetles, the Microvelia bug and prob-
ably some mites and ants. One parasitoid is present. The Subulina snail,
Periplaneta roach, and Wasmannia ant were introduced by man. Few of the
species seem to be endemic to Barbados.
Only the stylodesmid milliped appears to be a cave-evolved species. All
others are troglophilic species with broad ecological tolerances and require-
ments. This is a loosely structured community, where few species interact
Florida Entomologist 64(4)
with each other directly, and there is little apparent niche separation. The
prime interaction is seemingly directly between the guanophages and their
food resource. The richness of the community is interesting; to join it most
species had to pass or overcome 2 dispersal barriers or filters. The first and
most important was in reaching the island itself, probably mainly from
South America and (for a litter fauna) via rafting in debris of the Orinoco
River, though direct observations seem not to be available. The second was
in locating the specific habitat of the cave. No published data are available
on the occurrence or abundance of the invertebrate species in other habitats
in Barbados, but there seems to be little reason to think that they do not
occur elsewhere. It is probable, following the extensive clearing of Barbados
for sugar cane plantations by 1700, that many of the forest litter species are
now much less abundant.
Analysis of the freshwater invertebrate faunas of nearby (150 km to the
west) St. Vincent by Harrison and Rankin (1976) suggested that these
faunas are remnants (following a vicariance breakup or fragmentation) of
a richer ancestral land mass and biota (Rosen 1975). While this may be so
for St. Vincent (and other islands in the Lesser Antilles arc) the geological
story of Barbados, as now understood, seems to require initial overwater
colonization by all its biota in or since the Pleistocene.
The systematists who helped with identifications are all thanked. Marcus
Buck organized and carried out the mapping of the cave. Barbados research
facilities and accommodation were provided by the Bellairs Research In-
stitute (of McGill University) at Holetown. Field work was supported by
an operating grant of the Natural Sciences and Engineering Research
Council of Canada to study the distribution and biology of forest litter and
cave invertebrates. Professors A. D. Harrison and H. F. Howden reviewed
BENDER, M., F. T. TAYLOR, AND R. K. MATTHEWS. 1973. Helium-Uranium
dating of corals from Middle Pleistocene Barbados reef tracts. Quat.
Res. 3: 142-6.
CARTWRIGHT, O. L., AND F. E. CHALUMEAU. 1978. Bredin-Archbold-Smith-
sonian biological survey of Dominica. The superfamily Scarabaeoidae
(Coleoptera). Smithsonian Contr. Zool. 279: 1-32.
CULVER, D. C. 1970. Analysis of simple cave communities. I. Caves as
islands. Evolution 24: 463-74.
1971. Caves as archipelagoes. Natl. Speleol. Soc. Bull. 33: 97-100.
-- J. R. HOLSINGER, AND R. BAROODY. 1974. Toward a predictive cave
biogeography: the Greenbrier Valley as a case study. Evolution 27:
DRAKE, C. J., AND R. F. HUSSEY. 1955. Concerning the genus Microvelia
Westwood, with descriptions of two new species and a checklist of the
American forms (Hemiptera: Veliidae). Florida Ent. 38: 95-115.
FERMOR, J. 1972. The dry valleys of Barbados. Inst. British Geog. Trans.,
-- 1977. Karst development in Barbados. Proc. 6th Int. Cong. Speleol.,
Olomoc, 1973. II, sub-section Ba: 131-40.
Peck: Invertebrate Cave Fauna
FROESCHNER, R. C. 1960. Cydnidae of the Western Hemisphere. Proc. United
States Natl. Mus. 111: 337-680.
HARRISON, A. D., AND J. J. RANKIN. 1976. Hydrobiological studies of East-
ern Lesser Antillean Islands. II. St. Vincent: Freshwater fauna-its
distribution, tropical river zonation, and biogeography. Arch. Hydro-
biol., Suppl. 50: 275-311.
MATTHEWS, R. K. 1973. Relative elevation of Late Pleistocene high sea level
stands: Barbados uplift rates and their implications. Quat. Res. 3:
MULLINEX, C. 1975. Revision of Paraphrynus Moreno (Amblypygida:
Phrynidae) for North America and the Antilles. Occ. Pap. California
Acad. Sci. 116: 1-80.
ORGHIDAN, T., A. NUNEZ JIMENEZ, V. DECOU, ST. NEGREA, AND N. VINA
BAYES. 1977. Rdsultats des expeditions biosp6ologiques Cubano-
Romaines A Cuba. Editura Acad. Rep. Social. Romania, Bucuresti.
PECK, S. B. 1974. The invertebrate fauna of tropical American caves, Part
II: Puerto Rico, an ecological and zoogeographic analysis. Biotropica
1975. The invertebrate fauna of tropical American caves, Part III:
Jamaica, an introduction. Int. J. Speleol. 7: 303-26.
1977. Recent studies on the invertebrate fauna and ecology of sub-
tropical and tropical American caves. Proc. 6th Int. Cong. Speleol.,
Alomoc, 1973. V, sub-section Db: 185-94.
1981. Zoogeography and invertebrate cave faunas in southwestern
Puerto Rico. Natl. Speleol. Soc. Bull. 43: 59-68.
AND J. PECK. 1980. Insect field work opportunities in Barbados.
Lesser Antilles. Ent. News 91: 63-4.
AND 1981. The subterranean fauna and conservation of
Mona Island (Puerto Rico): a Caribbean karst environment. Natl.
Speleol. Soc. Bull. 43: 70-9.
ROSEN, D. E. 1975. A vicariance model of Caribbean biogeography. Syst.
Zool., 24: 431-64.
ROWLAND, J. M. AND J. R. REDDELL. 1980. The order Schizomida
(Arachnida) in the New World. III. Mexicanus and Pecki groups
(Schizomidae: Schizomus). J. Arachnol. 8: 1-34.
SMETANA, A. 1978. Revision of the subfamily Sphaeridiinae of America
north of Mexico (Coleoptera: Hydrophilidae). Ent. Soc. Canada Mem.
SWANEPOEL, P., AND H. H. GENOWAYS. 1978. Revision of the Antillean bats
of the genus Brachyphylla (Mammalia: Phyllostomatidae). Carnegie
Mus. Nat. Hist. Bull. 12: 1-53.
TABOADA, G. S. 1974. Sinopsis de la esp6leofauna Cubana. Ser. Espel. y
Carsol. no. 43, Acad. Ciencias Cuba, Habana. 65 p.
VUILLEUMIER, F. 1973. Insular biogeography in continental regions. II. Cave
faunas from Tessin, southern Switzerland. Syst. Zool. 22: 64-76.
Florida Entomologist 64(4)
VELVETBEAN CATERPILLAR: RESPONSE OF
MALES TO VIRGIN FEMALES AND
PHEROMONE IN THE LABORATORY AND FIELD
D. W. JOHNSON2, E. R. MITCHELL3, J. H. TUMLINSON3
AND G. E. ALLEN4
Observations of the sexual behavior of Anticarsia gemmatalis (Hiibner)
were made in the laboratory and field during 1976. The calling behavior of
the female is characteristic of many moth species in that the tip of the
abdomen is curved upward when calling. In the laboratory, males fly in a
zig-zag path toward calling females or to pheromone extracts. Clasper ex-
tension by some males was observed as they approached the pheromone
source. Females began calling ca. 2 h after sunset and this behavior con-
tinued intermittently throughout the scotophase. Mating and increased age
greatly reduced the attractiveness of females to males in field traps.
Pheromone was obtained from both non-calling and calling females. In
field bioassays, extracts from females collected before and after dark were
effective in attracting males, with extracts taken at 4 and 6 h after dark
capturing the most males. Seven trap designs were evaluated for effective-
ness at capturing A. gemmatalis males, but only the electric grid trap proved
efficient and warranted further use.
El comportamiento sexual de Anticarsia gemmatalis (Hubner) fue ob-
servado en el campo y en el laboratorio en 1976. El comportamiento de la
hembra en atraer a los machos es semejante a el de muchas otras species de
polillas, eso es, el punto del abdomen se encuerva para arriba. En el lab-
oratorio los machos vuelen en un rumbo de zigzag hacia las hembras atrac-
tantes y hacia extractos de feromanos. Se observaron la extension de los
agarradores de unos machos cuando estos acercaron el fuente de la feromona.
Las hembras empezaron a "llamar" (atraer machos) cerca de 2 horas
despues de la puesta del sol y este comportamiento siguia intermitentamente
durante el period de oscuridad. El apareamiento y la edad avanzada reduce
much la actractividad de las hembras en trampas. Se obtuvieron la feromona
de hembras que llamaba y los que no llamaba. En bio-ensayos de campo, los
extractos de las hembras que se capturaron antes y despues del period de
oscuridad fueron efectivos en atraer a los machos. Los extractos de las
hembras capturadas 4 y 6 h despues de volverse oscuro atrayeron el mAximo
nfmero de machos. Se evaluaron 7 disefios de trampas por efectividad en
capturar machos de A. gemmatalis. Solamente la trampa de rejilla electrica
result bastante eficiente para continuar usarlo.
'Present address: Department of Entomology, University of Kentucky, West Kentucky
Research and Education Center, Box 469, Princeton, KY 42445 USA.
'Insect Attractants, Behavior and Basic Biology Research Laboratory, Agric. Res. Serv.,
US9DA, Gainesville, FL 32604.
4Present address USDA/SEA-CR, Washington, DC. Florida Agricultural Experiment Sta-
tion Journal Series No. 1284.
Johnson et al.: Velvetbean Caterpillar
Greene et al. (1973) made observations in field cages on mating and
oviposition behavior of the velvetbean caterpillar, Anticarsia gemmatalis
(Hiibner). They suggested the existence of a female sex pheromone, after
noting that males frequently would fly around females whose abdomen tips
were pointed dorsally. Because insect sex pheromones are providing new
methods of controlling insect pests, we initiated investigations to confirm
the presence of a sex pheromone in A. gemmatalis and to identify param-
eters which may affect female attractiveness and male response.
PROCEDURES AND RESULTS
Insects used in these experiments were reared on an artificial diet accord-
ing to the methods of Greene et al. (1976). Insects were sexed in the pupal
stage (Butt and Cantu 1962) and placed in pans of moist vermiculite. Males
and females were allowed to eclose in screenwire cages in separate rooms.
Adults were maintained on 10% sucrose solution. Cages containing moths
to be used in behavioral observations were held in greenhouse rooms which
allowed penetration of sunlight (temperature ca. 240C, RH ca. 55%). Fe-
male moths which were extracted for pheromone were held in a screenwire
cage (63cm x 61cm x 91cm), under a light cycle of 12:12, photophase:
scotophase (temperature ca. 240C, RH ca. 55%).
BEHAVIORAL OBSERVATIONS: Observations of pheromone release behavior
were made in the greenhouse room containing the males. Individual virgin
females were confined in screenwire cylinders (4.5 cm, diameter x 12.7 cm,
length) suspended within a larger cage (57 cm x 54 cm x 184 cm) contain-
ing ca. 100 virgin males. Observations of the calling behavior and the sub-
sequent responses of males were made between 1730 and 0200 h on 7 con-
secutive nights. Sunset occurred at about 1800 h during these tests.
Males became active ca. 45 min. after sunset. This activity was char-
acterized by apparently indiscriminate flight around the cage, followed by
walking on the sides of the cage, and rapid fluttering of the wings. Females
exhibited similar behavior except they were unable to fly because of their
confinement. Obvious sexual activity was not noted until ca. 2 hours after
dark. Female calling consisted of wing fanning followed immediately by
dorsal elevation of the abdomen tip. Females repeated the calling behaviors
in an unbroken sequence for a time period varying between a few seconds to
several minutes. These calling sequences were repeated by each female
several times during the observation period. Calling was observed only when
females were on the vertical walls of the cage. During this activity, males
flew toward calling females. Walking or non-calling females did not elicit
Male response consisted of flying a zig-zag path toward the female.
Flight was initially erratic but became more direct as the male approached
the female. Upon reaching the female, the male flew around her several
times, and either landed or flew away. As males hovered near the female,
the claspers were usually extended.
EXTRACTION OF PHEROMONE: Calling females were vacuumed into a
1-liter jar containing sufficient ether to just cover the insects. The jar was
checked often to insure that the females were always covered with solvent.
Then they were extracted by soaking for 30 min. and the insect carcasses
Florida Entomologist 64(4)
were removed by filtration. The collection vessel was carefully rinsed twice
with small portions of ether. This wash was added to the original filtrate
by pouring it through the filtration apparatus (Mitchell et al. 1974). The
excess ether was removed by distillation under atmospheric pressure through
a 375 mm Vigreux column. This concentrated extract and the distillate were
stored at -600C until bioassay. The number of females obtained during
each collection varied due primarily to the availability of insects. On vari-
ous occasions, as few as 5 and as many as 75 females were collected. Collec-
tion required a minimum ca. 25 ml of solvent and the volume of residue
after distillation was ca. 1 to 2 ml for each 100 insects extracted.
A bioassay of the concentrated materials was conducted by placing 25
female equivalents (FE) on a 9 cm Whatman #1 filter paper and presenting
the paper to caged males. The cage and conditions were the same as those
employed while making behavioral observations. The bioassay began at 2 h
past sunset and the cage contained ca. 100 males. Using this procedure, ether
extracts from calling and non-calling females collected at night were shown
to be attractive and male responses were typical of the behavior shown by
males approaching calling females.
FIELD BIOASSAYS: Traps of 7 different designs were tested to determine
the most effective trap for capturing males: an electric grid trap (Mitchell
et al. 1972); a double-cone trap similar to that of Kaae and Shorey (1972);
an omnidirectional trap designed for pink bollworms, Pectinophora gossy-
piella (Saunders) (Sharma et al. 1973); a screenwire trap used for the
tobacco hornworm, Manduca sexta (L.) (Mitchell et al. 1972); a sticky trap
used for the sugarcane borer, Diatraea saccharalis (Fabr.) (Perez and Long
1964); a Pherocon 1C sticky trap (Zoecon Corp., Palo Alto, CA); and a
large walk-in trap constructed from screenwire used for capturing the
lesser peachtree borer, Synanthedon pictipes (Grote and Robinson) (Gentry
and Blythe 1978). Traps were baited with 3 virgin females (2- to 3-days
old) and placed ca. 30 m apart in a soybean field near Gainesville, FL.
Captured males were removed daily and the traps were rotated one position.
Each collection was considered a replicate (4 replicates). Results obtained
in this test indicated that the electric grid was the only trap effective
against the VBC. The electric grid trap captured 5.9 0.1 (-XS.E.) moths/
trap/night. All other traps captured no VBC moths. The ineffectiveness of
traps other than the electric grid for capturing fall armyworm, Spodoptera
frugiperda (J. E. Smith), and the beet armyworm, S. exigua (Hiibner),
was noted by Tingle and Mitchell (1975). Therefore, all other field assays
were conducted using electric grid traps.
The relative attractiveness of laboratory-reared and feral females was
studied to insure that laboratory-reared females would provide reliable in-
formation. Individual feral and laboratory-reared females of equivalent age
were placed in electric grid traps. The mean capture of males/trap/night by
traps baited with feral females was 6.6 -1.3 and 4.60.5 (-XS.E.) by
traps baited with laboratory-reared females (15 replications, 3 traps/female
type, for 5 consecutive nights). The mean capture by feral and laboratory-
reared females was not significantly different (t-test, t=0.407). Unbaited
traps captured no males.
The influence of age of females on pheromone production was evaluated
in 2 tests. One test utilized 3 laboratory-reared females as bait. Age groups
Johnson et al.: Velvetbean Caterpillar
tested were 1, .. .7 days post eclosion. A replicate was considered to be the
capture of males by one trap/night (3 replicates/age group). The 2nd test
utilized baits consisting of 20 FE of crude extract obtained from non-calling
laboratory-reared females of different ages. Females were collected 2 hours
after dark. Each age was replicated 4 times. All females used in these tests
were aged in screenwire cages in the greenhouse. A pan containing female
pupae was placed in a cage and allowed to remain over night. The following
day the pan containing uneclosed pupae was transferred to a second cage
leaving the newly closed females behind. Therefore, each cage contained
females which had closed during a single night.
In the tests using live females, the baits were placed in the field just be-
fore sunset and removed the following morning. Test females were held
during the day at 15C under a photoperiod approximating that in the field.
Females dying during the day were replaced with females of equivalent age.
Pheromone extracts were tested by placing 20 FE in 0.25 ml of solvent on a
9 cm #1 Whatman filter paper and suspending it within the trap. Trap loca-
tion within the field was constant. At each collection, the treatment was
replaced with a new and different one; these treatments were rotated
throughout the field. Moth capture for the extracts were recorded hourly
(replicate). Extracts for each age group (1, 2, 3, 5, 10 days old) were repli-
cated 4 times. Tests were run between 2400 and 0300 h.
In tests using live females 1-7 days of age as baits, trap capture of
males was 4519, 6612, 37.39, 43.321, 11.3+4, 10.6+4, and 51, re-
spectively (XS.E.). Linear regression analysis substantiated a decrease
of male capture with increasing age of the bait females (P=.05). In tests
using 20 FE of crude extract obtained from 1, 2, 3, 4 and 10-day-old females
capture of males was 31.6, 2.50.2, 2.3+1, 3.53 and 3.0 0.6, respec-
tively (-+XS.E.). No significant difference in attractiveness of these baits
to VBC males was apparent. This was somewhat surprising in that there was
a significant decline in the number of VBC males captured in traps baited
with live females of different ages. Thus, one might assume that not only age
but age in conjunction with exposure to the changing environment caused
the decrease in female attractiveness. The extracts were taken from females
which were not exposed to these changes as they were held under constant
temperature, humidity, light cycle and food supply.
The nocturnal sexual activity of the VBC male was monitored with an
electric grid trap equipped with an automatic sample-changing device
(Mitchell et al. 1972) that captured males and segregated samples into
hourly intervals. The trap was baited with 3 laboratory-reared females. The
bait females were placed in the trap ca. 1 h before sunset (20:15 EDT) and
removed early the following morning. Captured male VBC were counted
daily. Each night's capture was considered a replicate (8 replicates). Cap-
tures of VBC males commenced ca. 1 h after sunset and remained fairly
uniform throughout the night ( + S.E. captures/h = 4.4 0.8).
Female VBC were extracted with ether as described previously to de-
termine if the pheromone could be obtained at any time during the photo-
periodic cycle. Laboratory-reared females held under a 12:12 photophase:
scotophase were collected at 9 and 6 h before dark, immediately after the
lights were turned off and 2, 4, 6 and 9 h into the scotophase. Females were
extracted when ca. 3 days old. The extracts were bioassayed between 2400
532 Florida Entomologist 64 (4) December, 1981
and 0300 h by baiting individual electric grid traps with 20 FE of pheromone
in 0.25 ml of solvent on a 9 cm #1 Whatman filter paper. Counts of cap-
tured VBC males were recorded hourly (replicate), and the filter paper was
replaced with fresh bait. Each treatment was replicated 4 times. All treat-
ments attracted some wild VBC males (2.5, 2.0, 2.0, 0.7, 5.6, 4.6 and 1.5 X
males/treatment, respectively). However, significantly more males were at-
tracted to pheromone extracts obtained from females 4 and 6 h into the
scotophase (P= 0.05, Duncans Multiple Range Test).
The effect of mating on female attractiveness to wild males was de-
termined by comparing the capture of males by traps baited with mated
females to that of virgin females of equivalent age. Newly emerged virgin
laboratory-reared females were placed in 4 liter paper cartons with virgin
laboratory-reared males at a ratio of 1 female to 4 males. The carton tops
were replaced with screenwire and the insects were held for 2 nights in the
greenhouse under natural light. The following day, mated and virgin fe-
males were placed in traps ca. 1 h before sunset, and they were removed
early the following morning. Each trap contained 1 female. Four traps per
female type per night for 8 nights were used in this evaluation. Male cap-
tures were recorded daily (replicate). Females were dissected to confirm
mating. The mean capture of males in traps baited with virgin females was
16.8 0.6 (X-S.E.) as compared to 1.5- 0.8 (XS.E.) males captured in
traps baited with mated females (means differ significantly at the 1%
level, student's t-test). The reduction in attractiveness of mated VBC fe-
males to feral VBC males is consistent with results reported for other moth
species (Perez and Long 1964, Raulston et al. 1975).
This investigation has confirmed the presence of a female-produced sex
pheromone in A. gemmatalis. The pheromone can be extracted with ether
yielding an attractive extract to both laboratory-reared and wild males when
tested under both laboratory and field conditions.
The authors express their gratitude to N. C. Leppla and T. R. Ashley of
the Insect Attractants, Behavior and Basic Biology Research Laboratory,
USDA, ARS, Gainesville, FL, for providing insects used in this study, and
for help in analysis of the data, respectively. In addition, we wish to
acknowledge W. W. Copeland and R. W. Hines, also of the Insect At-
tractants, Behavior and Basic Biology Research Laboratory, for their tech-
nical assistance. Finally, we wish to thank G. R. Greene, University of
Florida, for providing some insects used in this research.
Mention of a commercial or proprietary product in this paper does not
constitute an endorsement of that product by the USDA. This paper is part
of a thesis prepared by the senior author as part of the requirements for
the M.S. degree presented by the Departrfent of Entomology and Nematol-
ogy, University of Florida, Gainesville, FL. Approved for publication 18
BUTT, B. A., AND E. CANTU. 1962. Sex determination of lepidopterous pupae.
USDA Tech. Bull. ARS-33-75.
Johnson et al.: Velvetbean Caterpillar 533
GENTRY, C. R., AND J. L. BLYTHE. 1978. Lesser peachtree borers and peach-
tree borers: a device for trapping, collecting and marking native
moths. Environ. Ent. 7: 783-4.
GREENE, G. L., J. C. REID, V. N. BLOUNT, AND T. C. RIDDLE. 1973. Mating
and oviposition behavior of the velvetbean caterpillar in soybeans.
Environ. Ent. 2(6) : 1113-5.
N. C. LEPPLA, AND W. A. DICKERSON. 1976. Velvetbean caterpillar:
A rearing procedure and artificial medium. J. Econ. Ent. 69(4):
KAAE, R. S., AND H. H. SHOREY. 1972. Sex pheromones of Lepidoptera.
XXIX. An improved double-cone trap for males of Trichoplusia ni.
Environ. Ent. 1(5): 675-7.
MITCHELL, E. R., J. C. WEBB, A. BAUMHOVER, R. G. ENDRIS, R. W. HINES,
AND S. MASUDA. 1972. Activity of tobacco hornworm moths as de-
termined by electrocutor grid traps baited with a blacklight and virgin
females. Environ. Ent. 1 (6) : 679-82.
J. H. TUMLINSON, W. W. COPELAND, R. W. HINES, AND M. M.
BRENNAN. 1974. Tobacco budworms: Production, collection and use
of natural pheromone in field traps. Environ. Ent. 3(4): 711-3.
PEREZ, R., AND W. H. LONG. 1964. Sex attractant and mating behavior in
the sugarcane borer. J. Econ. Ent. 57(5): 688-90.
RAULSTON, J. R., J. W. SNOW, H. M. GRAHAM, AND P. P. LINGREN. 1975.
Tobacco budworm: Effect of prior mating and sperm content on the
mating behavior of females. Ann. Ent. Soc. America 68 (4) : 201-4.
SHARMA, R. K., R. E. RICE, H. T. REYNOLDS, AND R. M. HANNIBAL. 1973.
Effects of trap design and size of hexalure dispensers on catches of
pink bollworm males. J. Econ. Ent. 66 (2) : 377-9.
TINGLE, F. C., AND E. R. MITCHELL. 1975. Capture of Spodoptera frugiperda
and S. exigua in pheromone traps. J. Econ. Ent. 68 (5) : 613-5.
Florida, Entomologist 64(4)
MULTIPARASITISM OF PUSS CATERPILLAR BY A WASP AND FLY
SPECIES (LEPIDOPTERA: MEGALOPYGIDAE)-(Note). Multipara-
sitigm is parasitism of one host individual by 2 or more different species of
parasites. The megalopygid moth, Megalopyge opercularis (J. E. Smith),
is parasitized in the cocoon stage by the ichneumonid, Lanugo retentor
(Brulle), and in the larval stage by one of 2 tachinid flies, either Carcelia
lagoae (Townsend) or Euphorocera sp. The tachinid flies glue their eggs on
the exterior of larvae. On hatching they become endoparasites. The wasp
L. retentor, on the other hand, oviposits only after the formation of the
cocoon and its larva lives as an ectoparasite on the host prepupa within the
shell of the cocoon (Khalaf 1975. Loyola Univ. Press, New Orleans, Louisi-
ana. 43 p.).
Experimental multiparasitism was induced following collection of mature
puss caterpillars (M. opercularis) at various times of the year from oak
trees in the streets of New Orleans. These were intended to serve as hosts
for the rearing and colonization of L. retentor. Many collected caterpillars
already harbored one of the tachinid parasites. Caterpillars lived in the
laboratory on oak leaves or artificial diet and in a few days formed cocoons.
A few days later, the cocoons were exposed to female L. retentor and then
were incubated for observation or for later dissection. Wasps were secured
from numerous cocoons, undoubtedly some after participating in unsuspected
multiparasitism. On the other hand, test cocoons from which flies emerged
following exposure to wasp oviposition represented known multiparasitism
and are listed in Table 1.
In case #1 pupation of the flies occurred after host cocoon formation;
thus the pupal stage of the fly was less than one month at that time of the
year. In case #2, the tachinid flies that emerged were sluggish and died
within 2 days. In case 5, the wasp penetrated the cocoon but removed the
ovipositor quickly, signifying that an egg was not laid. The wasp then pene-
trated again and even tried to penetrate for the third time when it was
finally removed. I repeatedly observed that when old host cocoons (e.g., 16
days old) containing one or more fly puparia were involved the penetrating
wasp (even in cases of multiple penetrations) would withdraw the ovipositor
quickly without laying an egg; Carcelia lagoae or Euphorocera sp. later
Multiparasitism was sometimes detrimental to both parasites (cases 7,
10, 12). In case 10, there was a scar on the puparium that could have been
caused by the wasp larva or the wasp penetrations.
Host cocoons 8-21 were penetrated twice by the wasp at various inter-
vals. The wasp accomplished the first oviposition readily since the tachinid
larvae were still within the host. However, the behavior in the second
oviposition, unless it followed the first in less than 3 days, was quite mod-
ified. Sometimes even at the 3 day interval, the second oviposition involved
1 or 2 short penetrations with a quick withdrawal (cases 15, 18); similar
behavior was encountered in the 5 day interval (case 16). In the 6 day
interval, wasp behavior involved rejection, prolonged or repeated short pene-
trations, and resting in contact with the cocoon between penetrations. Three
or 4 days after the first oviposition, the fast feeding wasp larva seemed to
drive the tachinid larvae out of the host to form puparia. Tachinid puparia
in host cocoons modified the wasp behavior, and rarely invited actual oviposi-
tion. Exceptions to this were cases 13 and 20 where the wasp laid an egg
even after the 6 day interval.
Cases 20 and 21 shed some light on the nature of competition between
the 2 parasites. The wasp larva apparently consumed the pupating fly, leav-
ing only the outer cuticle. Therefore, this case of multiparasitism became
Another method was used to study multiparasitism; live, overwintering
host cocoons were dissected on the day of collection between the first of
February and the last part of March. Some of the cocoons were multipara-
sitized by both the wasp and a tachinid fly. Some live, overwintering host
cocoons contained a large, living Lanugo larva in its own cell, and among
the walled off material, the host larval skin was found with one or more
tachinid egg shells still attached to it. In cases of numerous eggs, the
tachinid species would most probably be Carcelia since a host parasitized by
this species gives rise to many flies. There was no trace of live tachinids
within the multiparasitized cocoons, nor was there a sign of tachinid emer-
gence, indicating that the fly failed to develop, most probably because the fly
parasitism came late in the life of the host larva. The presence of tachinid
eggs on the prepupa was evidence that the tachinid eggs were sometimes laid
on the integument of the last larval instar. On the other hand, the presence
of live wasp larvae in the multiparasitized cocoons (with dead tachinid re-
mains) does not necessarily indicate the wasp's success. Success necessitates
at least the emergence of the wasp. For this reason cocoons from which the
adult wasp emerged were dissected in search for tachinid eggs which might
be attached to the host integument. Six of 41 host cocoons that overwintered
and produced wasps also contained tachinid eggs.
In the final series of investigations, obviously parasitized cocoons from the
field were dissected in a search for remains of the other parasitic species.
The wasp leaves the host cocoon by chewing a hole; the tachinid flies leave
by opening the operculum. Over 100 field-collected, deserted host cocoons
were dissected. Evidence of tachinid parasitism was detected in 2 percent
of the cocoons from which wasps emerged. Seven percent of cocoons from
which flies emerged showed traces of unsuccessful wasp infection. (This in-
vestigation received support from the Academic Grant Fund of Loyola Uni-
versity.)-K. T. KHALAF; Loyola University, New Orleans, LA 70118 USA.
Florida Entomologist 64(4)
f >> &i
rrt c a oi
w S <
e e P? s
> k k'~
0) 0)r-' r
Ii ] ULJ k
M X !> !> tD
.s ~5 S
> ow ci
0 X C1 CI C)
I I h-(
I I I
1-1 - T-
I;3C IM 1-
I-I b41I Il-4 -
o 5o so
C3 to & O
IU h ^ N
T~fmO ^1 t
4- I-p 1-4 -
o -q L L -
I 111I I
Sto t-- o00 e 0
-I -l4 l-1 C
Florida Entomologist 64(4)
CITRUS PULP BAIT INSECTICIDE FORMULATIONS FOR CONTROL
OF VELVETBEAN CATERPILLAR1 AND BEAN LEAF BEETLE ON
SOYBEAN3-(Note). Ground citrus pulp has recently been developed as a
carrier for technical methomyl under the trade name Nu-bait R4 which con-
tains 1.25% AI by weight. Martin (1979. Insecticide and Acaricide Tests
4: 126-7) and Bass (1979. Insecticide and Acaricide Tests 4: 142-3) re-
ported control of the fall armyworm, Spodoptera frugiperda (J. E. Smith),
on Coastal bermudagrass and peanuts, respectively, with this product. The
same material was considered as a potential carrier to enhance activity of
microbial insecticides. To investigate this possibility, DipelR, a commercial
formulation of Bacillus thuringiensis, was formulated by blending with
ground citrus pulp at a ratio of 1:19 wt/wt. The test described here was
conducted to determine the activity of both formulations against the velvet-
bean caterpillar (VBC), Anticarsia gemmatalis Hiibner, the bean leaf beetle
(BLB), Cerotoma trifurcata (Forster), and non-target arthropods on soy-
A small-plot field trial was initiated on soybeans on 3 October 1979, at
the Black Belt Substation (Auburn University Agricultural Experiment
Station) near Marion Junction, Alabama. An infestation of VBC (34.2
larvae/row m) was found which was uniformly distributed over a small,
late-planted field of 'Bragg' variety soybeans in the R6 developmental stage
(Fehr, W. R. and C. E. Caviness. 1977. Iowa Coop. Ext. Serv. Rep. 80:
12 p.). Significant numbers of BLB (22.4 adults/row m) were also present
at this date.
Treatments used in the test included Nu-bait at rates of 0.14 and 0.28
kg AI/ha, ground citrus pulp containing 0.56 kg Dipel/ha, methomyl
(Lannate 1.8EC) spray at 0.50 kg AI/ha, and an untreated control. A com-
pletely randomized field plot design was utilized with 2 replications. Replica-
tion number was limited by the small field size and minimal plot size re-
quired. Plots were 4 rows (0.91-m spacing) by 15.2 m. Bait formulations
were applied with a cyclone-type, hand-operated applicator (Seymour No.
75 Universal Spreader) calibrated to deliver 11.2 kg of bait/ha. Plots re-
ceiving 22.4 kg of bait/ha were treated twice at the 11.2 kg delivery rate.
Methomyl was applied with a CO2 sprayer calibrated to deliver 187 1 total
volume/ha with 3 nozzles (TX 4 cone spray tips) at 2.11 kg/cm2 (30 p.s.i.).
Target pests were sampled by the shake-cloth technique (Boyer, W. P.
1963. Coop. Econ. Insect Rep. 13: 91-2). Three 0.91 row-m samples were
taken at 2 and 5 days post-treatment in each plot. Larval counts were taken
in small and medium plus large size classes so that total VBC counts and
counts of the damaging size classes (medium plus large) could be analyzed
separately. In addition, the effects on non-target insects were measured by
taking 9.1 row-m D-vac samples (Dietrick et al. 1959. J. Econ. Ent. 52:
1085-91) in each plot on each sample date. Insects collected in the latter
samples were returned to the laboratoryand sorted into various taxonomic
categories. All insect counts from both sampling techniques were analyzed
by analysis of variance and means were compared by multiple range tests.
'Anticarsia gemmatalis Hubner (Lepidoptora: Noctuidae).
2Cerotoma trifurcata (Forster) (Coleoptera: Chrysomelidae).
3Alabama Agricultural Experiment Station Journal Series #15-81007.
4A product of Griffin Corporation, Valdosta, Georgia.
Scientific Notes 539
Populations of both VBC and BLB increased in untreated plots in the
2 days between treatment and first evaluation. Increase in VBC could be
attributed to a large egg hatch, since the majority were small larvae at 2
days post-treatment. Increase in BLB numbers may be related to movement
from immediately adjacent soybeans which were senescing rapidly.
Nu-bait at both 0.14- and 0.28-kg rates significantly reduced VBC popu-
lations at 2 days post-treatment (Table 1). At 5 days, the reduction was still
evident. Methomyl, utilized as a standard control and applied at 0.50 kg,
significantly reduced VBC populations by 93 and 88% at days 2 and 5, re-
spectively, but provided no better reduction of VBC numbers than citrus
pulp containing only 1/4 or 1/2 as much methomyl. Citrus pulp containing
1 kg of Dipel per 20 kg of formulation and applied at the rate of 11.2 kg of
formulation or 8.9 billion international units of potency/ha reduced VBC
levels by 50 and 100% at days 2 and 5, respectively. The slower mode of
action necessitates a 5-day reading for this material. All size classes of
VBC were effectively reduced by all treatments, although percent reductions
of small larvae were generally somewhat higher than reductions of medium +
BLB populations were affected in several different ways by the various
treatments. Both Nu-bait treatments significantly reduced BLB populations
at 2 days, while methomyl spray did not (Table 2). Dipel applied in citrus
pulp was ineffective against BLB. At day 2, plots treated with Dipel in citrus
pulp actually had significantly higher (by 50%) BLB populations than in
untreated plots. Dipel applied as a spray in a separate test (Harper, J. D.
1981. Insecticide and Acaricide Tests 6: 147) also was shown to have no
activity against BLB. These results suggest that the citrus pulp attracts
BLB. By day 5, all treatments had similar BLB populations, suggesting that
BLB is highly mobile and that the effects of the bait on it are rapidly masked
in small plots by interplot movement. For this reason, large plots would be
TABLE 1. COUNTS OF VELVETBEAN CATERPILLARS, Anticarsia gemmatalis
HijBNER, FOLLOWING TREATMENT ON 3 OCTOBER 1979, WITH SE-
LECTED INSECTICIDES-MARION JUNCTION, ALABAMA.*
ment Days post-treatment
Med. & Med. &
Rate large Total large Total
Material (kg AI/ha) larvae larvae larvae larvae
Untreated 34.2 46.5a 102.3a 31.7a 78.7a
Nu-Bait 0.14 10.4c 17.0b 1.6b 1.6b
Nu-Bait 0.28 8.7c 9.3b 8.7b 15.9b
methomyl 0.50 2.7c 7.7b 4.4b 9.3b
+ Dipel 0.56** 27.3b 50.9b 0.Ob 0.Ob
*Numbers within columns followed by the same letter are not significantly different
(DMRT) at the 0.05 level.
**Expressed as kg of Dipel/ha.
Florida Entomologist 64(4)
TABLE 2. COUNTS OF BEAN LEAF BEETLES, Cerotoma trifurcata (FORSTER),
PER ROW-M OF SOYBEANS FOLLOWING TREATMENT ON 3 OCTOBER
1979, WITH SELECTED INSECTICIDES-MARION JUNCTION, ALA-
Rate Pre-treatment post-treatment
Material (kg AI/ha) 2 5
Untreated 22.4 61.Sb 48.7a
Nu-Bait 0.14 24.6cd 39.4a
Nu-Bait 0.28 12.6d 33.4a
methomyl 0.50 47.6bc 54.7a
citrus pulp + Dipel 0.56** 93.0a 37.7a
*Numbers within columns followed by the same letter are not significantly different
(DMRT) at the 0.05 level.
**Expressed as Kg of Dipel/ha.
needed to determine efficacy of the methomyl baits on BLB representative of
Non-target arthropods were separated into 18 different taxonomic cate-
gories ranging from order to species. Individual taxa were generally present
in small numbers in the D-vac samples. Populations of most individual taxa
present showed no significant response to any treatment at either day 2 or
5 post-treatment. Total numbers of non-target arthropods were highest (but
not significantly) in the Dipel bait plots on both dates (Table 3). This ap-
parent concentration of organisms, which included significantly higher num-
bers of predatory nabid nymphs (Hemiptera: Nabidae) suggests that the
bait is acting as an attractant. Higher mortality in the pest species studied
suggests that it may also be functioning as a gustatory stimulant. The exact
nature of citrus pulp attractancy and its role in pest management should be
explored further.-JAMES D. HARPER, Dept. Zoology-Entomology; Auburn
University Agricultural Experiment Station; Auburn University, AL 36849
TABLE 3. NUMBERS OF ARTHROPODS OTHER THAN BEAN LEAF BEETLES, Cero-
toma trifurcata (FOSTER) AND VELVETBEAN CATERPILLAR, Anti-
carsia gemmatalis HiiBNER, COLLECTED IN D-VAC SAMPLES FROM
SOYBEAN PLOTS FOLLOWING TREATMENT ON 3 OCTOBER 1979, WITH
SELECTED INSECTICIDES AT MARION JUNCTION, ALABAMA. ALL
FIGURES ARE AVERAGE COUNTS PER 3 ROW-M OF BEANS.*
Nu-Bait Nu-Bait Citrus pulp
Day post- (0.14 kg (0.28 kg +
treatment Control AI/ha) AI/ha) Dipel Methomyl
2 38.0 35.0 28.0 62.0 21.0
5 11.5 15.0 16.5 26.5 18.5
*No significant differences occurred between treatments within dates (DMRT, 0.05).
Scientific Notes 541
"LOVE BITES" IN A LYCID BEETLE'-(Note). While copulating and
mate-guarding, male insects are sometimes dislodged from their mounts by
sexual rivals. Various clamps, claspers, spines, and gin-traps occur in the
genital region to grip the female and secure the male's position (Parker
1970, Biol. Rev. 45: 525-67). But the lycid beetle Calopteran discrepans
(Newn.), forms an unusual sexual attachment. Coupled males pierce the
female integument with their mandibles.
In a deciduous forest in Alachua Co., Florida, I watched 40 male and 4
female C. discrepans. Males were engaged in short 0.1 to 3.0 m flights
among the shrubs and herbs. After alighting they would flick open their
large black and yellow poster-like elytra. In 45 landings the number of wing
flicks varied from 0 to 13, with a mean of 1.3. Flicks may be an advertise-
ment to females or an aposematic display to tracking predators (Burke 1976,
Ent. News 87: 229-32; females also flick when disturbed and female C.
terminal appear to flick more regularly than do males; again Burke (1976).
Females were found in the leaf litter. The highly biased sex ratio, 10:1,
suggests that males remain in emergence areas to search for mates.
Males mount dorsally, between the females' slightly spread wings. Ex-
amination of coupled pairs showed the sickle-like male mandibles bite
through the humeral angle (shoulder) of the female's right elytron. Up to
3 males were found upon a female's back. When such masses were picked
up, they clung together and were separated only with some effort, leaving
bleeding wounds in the female's elytra. Interloping males inserted their jaws
into the mid regions of the elytra. These positions are similar to those of the
Nigerian lycid Metriorrhynchus semiflabellatus Thoms., in which a copulat-
ing male was seen to grip the right tibia of his mate and competing males
held the edge and corrugations of the female elytra in their mandibles
(Poulton 1913, Proc. Roy. Ent. Soc. [in trans.] LXXVIII). Burke ibidd.)
found a non-mating C. terminalis male puncturing the intersegmental mem-
branes between a female's metathorax and first abdominal segment. He
presumed the male to be feeding on her body fluids and does not mention
bites by copulating males, though a mating male had his head pressed
against the same region of his mate. When disturbed, mounted C. discrepans
raise their meso- and metathoracic legs, presumably to prevent the position-
ing of an interloper (much like the leg lifts of another cantheroid, the fire-
fly Photuris hebes (= SH), Lloyd 1977, Florida Ent. 60:63).
Single males engage in biting combats. In one case, a captive male
pierced the ventral thorax of his opponent and held him for 24 hours. A
victory this expensive in time seems Pyrrhic. It is questionable whether
fights in nature are so prolonged (more energetic and mortal use of man-
dibles occurs in the related phengodids, e.g. Tiemann 1967, Proc. California
Acad. Sci. 35: 235-64).
Love bites, with their potential damage to the female, illuminate the
different reproductive interests of the sexes. The diminished reproductive
value of an injured mate can be offset by the greater percent fertilization a
male achieves if his ejaculate is not mixed with a rival's. Thanks to
D. Wojcik for supplying references.-JOHN SIVINSKI, Dept. of Entomology
and Nematology, University of Florida, Gainesville, FL 32611 USA.
Florida Agricultural Experiment Station Journal Series No. 0000.
Florida Entomologist 64(4)
WEST INDIAN RECORD FOR ELASMUS POLISTIS BURKS (HY-
MENOPTERA: EULOPHIDAE)-(Note). Elasmus polistis, a parasitoid
of Polistes spp., was described by Burks (1971. J. Washington Acad. Sci.
61: 194-6) from specimens collected in Georgia, Maryland and Pennsylvania.
Burks suggested that the species had been introduced, but had no clue as
to its origin. Support of the introduction hypothesis is derived from the
absence of E. polistis from several studies of Polistes nests prior to 1971 and
its subsequent appearance as a common species in Pennsylvania, Maryland,
Georgia, Texas and Oklahoma (Reed and Vinson 1979. J. Kansas Ent. Soc.
Specimens determined as E. polistis by NFJ were reared from nests of
Polistes crinitus Felton by MAI at Charlotte Amalie, St. Thomas, Virgin
Islands, on 15 June 1979; and from Estate Bordeaux, St. Thomas, VI on
25 November 1979. The Charlotte Amalie population showed a high in-
cidence of parasitism; several thousand parasitoids were reared from a 2
liter container-full of nests. This is the first report of Elasmus polistis from
outside the eastern and central U.S.A. and from the endemic West Indian
Since E. polistis is probably an effective colonizer, its occurrence in the
West Indies may be the result of another introduction. However, the in-
frequency of this type of collecting in the West Indies may have allowed its
presence to go unrecorded. Polistes spp. are common here, and rearing at-
tempts should be made on other islands. Additional distribution records
may indicate a West Indian origin of this species.
Voucher specimens of Elasmus polistis from St. Thomas are deposited in
the collection of the Ohio State University.-NoRMAN F. JOHNSON AND
MICHAEL A. IVIE, Department of Entomology, 1735 Neil Ave., Ohio State
University, Columbus, OH 43210, USA.
BUFFALOGRASS, A NEW HOST FOR THE FALSE SPIDER MITE,
AEGYPTOBIA NOMUS (ACARINA: TENUIPALPIDAE) --(Note).
Buffalograss, Buchloe dactyloides (Nutt.) Engelm. (Poaceae), is a prom-
inent component of the shortgrass prairies of the American Southwest. Oc-
casionally, buffalograss is used as a low-maintenance turfgrass in South-
western lawns. The only previously reported pests are the buffalograss web-
worm, Surattha indentella Kearfott, in Kansas (Sorensen, K. A., and H. E.
Thompson. 1979. J. Kansas Ent. Soc. 52: 282-96) and the white grub of
Phyllophaga crinita Burmeister in Texas (Reinhard, H. J. 1940. J. Econ.
Ent. 33: 572-8).
Aegyptobia (Pentamerismus) nomus (Baker and Pritchard) previously
has been taken on Bouteloa (sic.) (=Bouteloua) gracilis (H.B.K.) Griffiths
(Poaceae) (blue grama), in Mandan, North Dakota (15 Oct. 1920) and in
Modena, Utah (22 Sept. 1925). It was also taken on Distichlis strict (Torr.)
Rybd. (Poaceae) (inland saltgrass) in Manden, North Dakota (15 Oct.
1920) (Baker, E. W., and A. E. Pritchard. 1953. Wasmann J. Biol. 11: 353-
'Texas Agricultural Experiment Station Journal Series TA No. 16603.
On 30 July and 25 August 1980, we found A. nomus infesting buffalo-
grass at 2 sites in Dallas, Dallas County, Texas. A record drought and heat
wave had forced unirrigated grasses into semi-dormancy, consequently leaf
color was not a useful character for damage detection. What brought the
infestation to our attention was the presence of "witches'-broom" fascicles
produced by the abortion of internodes on some terminals. Microscopic ex-
amination of such terminals disclosed populations of A. nomus under the
Because A. nomus rarely is collected, it may have only minor pest po-
tential on buffalograss. The apparent damage we witnessed may have re-
sulted primarily from the unusual weather of the summer of 1980. Con-
versely, because buffalograss historically has been subjected only occasionally
to cultivation, past infestations by this mite may merely have passed un-
We thank Dr. Edward W. Baker (USDA, SEA, Systematic Entomology
Laboratory, Beltsville, MD) and the Florida Department of Agriculture's
Division of Plant Industry for the taxonomic determination of A. nomus.-
ROBERT L. CROCKER, WILLIAM E. KNOOP, AND JOHN M. OWENS, Texas A&M
University Research and Extension Center, 17360 Coit Road, Dallas, Texas
TOXICITY OF ATMOSPHERIC GASES TO IMMATURE STAGES OF
ANASTREPHA SUSPENSA-(Note). Fumigation is one of the most
effective ways of destroying insects, and development of new and improved
techniques is a principal research effort at our station. The toxicity of
atmospheric gases in high concentration has been known to entomologists for
a long time. Carbon dioxide (CO,) in particular was used almost routinely
to anesthetize insects (to facilitate handling) until detrimental effects were
detected (Sherman. 1953. J. Econ. Ent. 46:15-9). Controlled atmospheres
(CA) are used during storage to prolong shelf-life of many fruits and vege-
tables (Hatton et al. 1975. Proc. Florida State Hort. Soc. 88: 335-8). The
lethal effects on insects plus beneficial effects on commodities stimulated our
interest in investigating the possibilities for controlling Caribbean fruit fly
with atmospheric gases.
Immature stages of the Caribbean fruit fly were exposed for various
periods of time to atmospheres of ethylene (C2H,), nitrogen (N,), and
carbon dioxide (CO,). Gases were delivered in constant flow through wide-
mouthed glass jars (2 liter) that served as CA chambers. Jars were con-
nected in continuous series for each type or concentration of gas, and could
be removed individually as needed to accommodate the time factor. Ethylene
was mixed with air to a mean concentration of 940 ppm; N2 and CO2 were
used pure (100%) or mixed with air to provide 40, 60, and 80% CO,. All of
the gases were humidified (90+% RH) by bubbling through water and were
then passed at constant flow rates (2 liters/hr for CH, and 4 liters/hr for
N2 and CO2) through the CA jars. Gas concentrations were measured by
gas chromatography before and after entering the CA jars. The technique
utilized a flame ionization detector for CH4, and a thermal conductivity
detector with appropriate columns for separating CO2, 02 and N,. A small
pump was used to pull air through the control jars at the same flow rates
Florida Entomologist 64(4)
as those used for test gases. All tests were conducted at 22-230C., except
the egg test which was at 100C. Mortality data in the treated lots were cor-
rected for natural mortality in the controls (air flow) by Abbott's formula.
Mature Larvae. Naked Caribbean fruit fly larvae in lots of 200 were ex-
posed to C2,H, CO,, and N2 for 4 to 7 hrs. Larvae were blotted dry (to
prevent escape from containers) and placed in open petri dishes which in
turn were placed in closed CA jars through which the gases flowed. Sur-
vivors were insects which were able to reach the adult stage. Ethylene at
940 ppm for 4 hr did not kill any larvae; CO2 at 50% for 4 hr and 100%
for 7 hr produced only 6.0 and 6.9% mortality, respectively; N2 at 100%
for 7 hr killed 7.4% of the larvae. A much longer exposure period would be
required to produce a significant lethal effect from these gases. The poor
performance by CH, coupled with the fact that it is explosive at 3% in air
caused us to drop it from further consideration.
Pupae. Atmospheres of 100% N, and 100% CO2 for periods ranging
from 36 to 72 hr were tested against 4 and 11-day-old pupae to measure their
ability to survive without oxygen. Pupae were exposed to the gases in lots
of 200, and adult flies which were recovered following treatment were
counted as survivors. Mortality of 4-day-old pupae ranged from 0 to 21%
in N2 and 0 to 3% in CO2. For 11-day-old pupae the corresponding values
were 0 to 17% in N2 and 11 to 78% in CO2. Nitrogen had similar but slight
effect on both age groups, but CO0 was uniformly more toxic to the older
Larvae in Artificial Diet. Five-day-old larvae were transferred (by
artist's brush) in lots of 150 from the laboratory culture into 3.5 x 6 cm.
dia. teflon-lined aluminum cups containing 40 ml of fresh laboratory diet.
The cups were then placed into the 2-liter testing jars and exposed to 100%
N, and 40, 60, 80, and 100% CO2 for periods ranging from 36 to 96 hr.
Larvae which attained full growth following treatment were counted as
survivors. Sixty hours of exposure were required to kill all larvae in 100%
N2. Although there was not a clear correlation between CO, concentration
and time of exposure, generally high mortality (95-100%) occurred at 48
hr with little added efficiency from longer exposures.
Eggs (1-day-old). Eggs from laboratory reared flies were delivered
volumetrically by pipette (about 250 per drop) into cups containing 40 ml
of fresh laboratory diet. They were then exposed to 40, 60, and 80% CO2
for 48, 72, and 96 hr. Eggs hatch within 3 days at room temperature. We
therefore conducted these tests at 100 C. to slow embryonic development
and insure that only the egg stage would be treated. Larvae which hatched
and were able to attain full growth were counted as survivors. Results were
similar to those for young larvae, with near total kill (95-99%) at 48 hr ex-
The toxicity of CO2 to eggs and young larvae, in particular, indicates
its potential for controlling this insect. Lethal effects of CO, were not clearly
proportional to concentration, suggesting that there may be no advantage
to increasing gas concentrations above a threshold level. Although CO, is
slower acting than conventional fumigants, its activity appears to fall
within the realm of practical application.-C. A. BENSCHOTER, D. H.
SPALDING, AND W. F. REEDER, Subtropical Horticulture Research Unit, ARS-
S&E-USDA, 13601 Old Cutler Road, Miami, FL 33158 USA.
MELILOTUS ALBA: A NEW HOST OF THE VELVETBEAN CATER-
PILLAR-(Note). Velvetbean caterpillar, Anticarsia gemmatalis (Hiibner)
(Lepidoptera: Noctuidae) larvae were first found on Melilotus alba Desr.
near Homestead, FL on 5 March 1980. A total of 29 larvae were collected
from this new host during March and April, 1980. Subsequently, larvae were
collected from M. alba during May, 1981. Melilotus alba is a shrubby
glabrous annual herb up to 2 m high and is found in wastelands and margins
of cultivated fields throughout the US and southern Canada (Long and
Lakela, 1971. A Flora of Tropical Florida. Univ. Miami Press). It has lush
growth during February-May in southern Florida and may play an im-
portant role in the development of a migratory A. gemmatalis population.
This research was funded in part by EPA grant No. CR-806277-02-0,
entitled "Development of Comprehensive, Unified, Economically, and En-
vironmentally Sound Systems of Integrated Pest Management for Manage-
ment for Major Corps". Florida Agricultural Experiment Stations Journal
Series No. 3252. -VAN WADDILL, University of Florida, AREC, 18905 SW
280 St., Homestead, FL 33031 USA.
L -- -* ---^- --- -- -- -- -- --
TENDING OF DEAD QUEENS BY WORKERS OF SOLENOPSIS IN-
VICTA BUREN (HYMENOPTERA: FORMICIDAE)-(Note). As part
of our normal research activity, we rear and maintain several hundred
colonies of Solenopsis invicta Buren, the red imported fire ant. If the mother
queen or her workers die they are normally discarded on the refuse pile
within a few days. However, recently, while conducting laboratory tests with
chemicals for toxic baits, we noted that worker ants from some colonies
tended their queen for many weeks after she had been killed by exposure to
the bait toxicant AC 217300 (tetrahydro-5,5-dimethyl-2(1H) [pyrimidinone
(3 [4- (trifluoramethyl) phenyl]1 (2-[- (trifluoromethyl) phenyl]ethenyl) -2-pro-
penylidene)hydrazone). This compound was reported (Williams et al. 1980.
J. Econ. Ent. 73: 798-802.) to be selectively toxic to the queen and that in
large colonies the queen died within 2 to 4 wks even though thousands of
workers survived exposure to the chemical. The queens of these colonies
were physogastric and ovipositing at the time of exposure to the bait. In 13
of these colonies the surviving workers maintained the gaster of the queen
for 3 to 9 weeks (Fig. 1), but in 4 others the workers quickly discarded the
queen. The workers were also observed with the entire queen or the thorax
and gaster, but they never maintained only the head or thorax alone or the
head and thorax together.
The behavior of workers to the dead queen's gaster was typical of that
observed with live queens, i.e., they clustered around the corpse with almost
constant grooming and licking of the cuticle. With one exception, efforts to
induce the workers to accept a new queen (2 to 4 wks old) were futile as
long as they kept the dead queen. New queens were killed just as they would
have been if a healthy queen had been present. However, once the body of
the dead queen was discarded, a new queen was usually accepted. In 1978
two of our coworkers (A. Glover and B. M. Glancey, personal communica-
Florida Entomologist 64(4)
Fig. 1. Abdomen of dead queen of S. invicta tended by worker ants.
tion) also observed a colony of S. invicta that maintained the gaster of their
dead queen for 7-8 months at which time ca. 100 workers remained alive.
Several attempts to re-queen this colony resulted in execution of the new
The observation of Glover and Glancey suggested that the tending be-
havior we observed was not related to some property of the toxicant, but
rather to the physiological condition of the queen at the time of death. This
hypothesis was strengthened when we killed 4 previously healthy queens by
freezing. One queen was discarded after 8 days, another after 14 days. The 2
remaining queens were still being maintained at 15 days when they were
removed for other tests.
Tending of dead queens by daughter worker ants for long periods of time
(many weeks) has not been reported previously. Recently, Fletcher and
Blum (1981. J. Georgia Ent. Soc. 16(3): 352-6.) reported on the use of dead
queens of S. invicta as bioassay units to assay for the presence of a deala-
tion inhibitory pheromone in mated queens. The length of time the dead
queens were maintained during these studies was not specified. E. O. Wilson
(1958. Psyche 65: 41-51.) noted that queens and some large major workers
of Pogonomyrmex badius Latreille were kept within the nest under labora-
tory conditions for 3 days after death during which time they were licked
and moved about. After this time, they were moved to the refuse pile. T. C.
Schnierla (Army Ants: A Study in Social Organization, 1971.) while ob-
serving a trail of Eciton hamatum, found the headless body of a dead queen
within a tight little ball of workers. Because of her dismembered and dried
condition, he assumed that she had been dead for many days. Workers clung
to the body when he moved it with tweezers.
Our observations suggest that the gaster-tending behavior is mediated
by a pheromone produced by the queen while she is alive, but which remains
on the cuticle or in some other organ or tissue after death. Jouvenaz et al.
(1974. Ann. Ent. Soc. America 67: 442-4.) and Glancey (1978. Proc. Tall
Timbers Res. Sta., No. 7, 173 p.) have reported that S. invicta queens pro-
duce a pheromone that attracts workers and Vander Meer et al. (1980. Ann.
Ent. Soc. America 73: 609-12.) have shown that a queen-tending phe-
romone (s) can be extracted from the poison sac. The behavior induced by
this pheromone is similar to that toward dead queens and can be mimicked
with surrogate queens (small pieces of rubber septa) treated with extracts
of the poison sac. The variability in retention time of the dead queen could
be related to the condition of the poison sac after death. If it remains intact,
the pheromone could be released slowly over a period of many weeks. Since
the venom alkaloids possess antimicrobia activity, they could delay deteriora-
tion of the poison sac. On the other hand, decreases in pheromone production
related to aging or senility or accidental destruction of the poison sac at the
time of, or shortly after death, could account for early rejection of the dead
The tending of dead queens by worker ants could be significant in plan-
ning control programs with bait-toxicants such as Amdro@ because requeen-
ing of field colonies from mating flights by newly-mated queens may not
occur as rapidly as expected. Thus, fewer treatments would be required to
maintain control. Mention of a commercial or proprietary product in this
paper does not constitute an endorsement or recommendation of this product
by the USDA.-DAVID F. WILLIAMS, CLIFFORD S. LOFRGEN AND ROBERT K.
VANDER MEER, Insects Affecting Man and Animals Research Laboratory,
ARS, USDA, P.O. Box 14565, Gainesville, FL 32604 USA.
Florida Entomologist 64(4)
The American Mosquito Control Association will hold its annual Photo
Salon in connection with the annual meetings in Sacramento, California,
18-22 April 1982. There will be a Photographic Society of America-approved
Insect Photo Salon and an AMCA Mosquito Photo Salon (which will cover
all aspects of mosquitoes and mosquito control). Both salons will be con-
ducted according to PSA standards. A silver medal will be awarded for
"Best of Show" in each salon. The competition will be open to all photog-
raphers. The entry fee will be $3.00 and the closing date for receipt of entries
will be 29 March 1982. Entry forms and complete details can be obtained
from: JOHN T. BARBER, Chairman, AMCA Photo Salon, Biology Department,
Tulane University, New Orleans, Louisiana 70118.
THE 65TH ANNUAL MEETING OF THE
FLORIDA ENTOMOLOGICAL SOCIETY
FIRST ANNOUNCEMENT AND CALL FOR PAPERS
The Florida Entomological Society will hold its 65th Annual Meeting on
10-13 August 1982 at The Colony on Longboat Key, Sarasota, Florida. The
Colony is located at 1620 Gulf of Mexico Dr., Sarasota, FL 33580. Mailing
address is P. O. Box 3949, Sarasota, FL 33578; telephone 813/383-5581. Suite
rates will be $55.00 for a one floor suite (2 sleeping areas), and $85.00 for a
two floor suite (2 BR, 2-bath, plus sleeping area). Pre-registration fee will
be $25.00 ($10.00 student). Registration at the door will be $30.00 ($12.00
student). Banquet is included in registration fee. Questions concerning local
arrangements should be directed to:
David M. Pomfret
Fahey Pest Control
P. 0. Box 1452
Sarasota, FL 33578
If you plan to present a paper the tear out sheet must be completed and
postmarked no later than 1 June, and sent to:
Abe White, Program Chairman
817 W. Fairbanks Ave.
Orlando, FL 32804
Eight minutes will be allotted for presentation of oral papers, with 2
minutes for discussion. Also, for the first time any member may elect to
present a poster paper at the evening session. Three oral student papers,
judged best on content and delivery, will be awarded monetary prizes. Stu-
dent authors must be Florida Entomological Society members and be reg-
istered for the meeting.
p p p- p-p-- a a-a a ai
SLIDE POLICY FOR ANNUAL MEETINGS
The following slide policy will govern slide presentations at the Annual
Meetings. Only Kodak Carousel projectors for 2 x 2 slides will be available.
However motion picture projectors will be available by special request to
the Local Arrangements Chairman prior to the date of the meeting.
Authors should keep slides simple, concise, and uncluttered with no more
than 7 lines of type on a rectangle 2 units high by 3 units wide. All printed
information should be readable to an audience of 300 persons.
A previewing room will be designated for author's use. A projectionist
will be available in the previewing room at least one hour before each session.
Authors are expected to give the projectionist their slides in the previewing
room prior to each session. Slides will be returned to the authors after each
session in the meeting room.
Authors are expected to organize their slides in proper order in their
personal standard Kodak Carousel slide tray (no substitution, please). Only
a few slide trays will be available in the previewing room from the projec-
tionist for hardship cases. Slides in the tray should be in correct order start-
ing with slot #1 of the tray and positioned correctly (position of slides to go
into tray: 1. upside down, and 2. lettering readable from this position upside
down and from right to left). A piece of masking tape should be placed on
the slide tray by the author and the following information should be written
on the tape: 1. author's name, 2. session date, and 3. presentation time.
Agriculture & Environment, Inc.
13616 NW 8th Ave.
Gainesville, FL 32601
American Cyanamid Co.
8475 Farrah Lane
Germantown, TN 38138
Asgrow Florida Co.
P. O. Drawer D
Plant City, FL 33566
P. O. Box 11422
Greensboro, NC 27409
Clark Pest Control, Inc.
Jacksonville, FL 32206
Dow Chemical USA
P. O. Box 22300
Tampa, FL 33622
Duncan Spray Service, Inc.
P. O. Box 2464
Lakeland, FL 33803
E. I. DuPont DeNemours
114 Overlook Dr., S. E.
Winter Haven, FL 33880
E. O. Painter Printing Co.
P. O. Box 877
DeLeon Springs, FL 32028
Entomo Agronomos, Inc.
18400 SW 134th Ave.
Miami, FL 33177
Estech General Chemicals Corp.
P. O. Drawer 1272
Winter Haven, FL 33880
Florida Pest Control Association, Inc.
3907 Rose Petal Lane
Orlando, FL 32808
1808 N. 57 St.
Tampa, FL 33619
Helena Chemical Co.
P. 0. Box 5525
Tampa, FL 33675
ICI Americas, Inc.
P. O. Box 208
Goldsboro, NC 27530
Joe Gould Tree Service
P. 0. Box 715
Sarasota, FL 33578
Lear Pest Control Co.
505 NW 103 St.
Miami, FL 33138
Ralph L. Miller, Hort. Consul.
1003 Lake Davis Drive
Orlando, FL 32806
Mobay Chemical Corporation
1007 Virginia Ave., Suite 104
Hapeville, GA 30354
P. 0. Box 631
Clemson, SC 29631
Pesticide Chemicals, Inc.
P. O. Box 369
Fort Pierce, FL 33459
Southern Liquid Fertilizer, Inc.
P. O. Box 56
Zellwood, FL 32798
Stauffer Chemical Co.
2009 Orient Road
Tampa, FL 33619
P. 0. Box 1697
Sanford, FL 32771
Thompson-Hayward Chemical Co.
P. O. Box 471
Sanford, FL 32771
Union Carbide Corp.
7825 Baymeadows Way
Jacksonville, FL 32216
Woodbury Chemical Co.
P. O. Box 4319
Princeton, FL 33032
Woolfolk Chemical Works, Inc.
P. O. Box 938
Fort Valley, GA 34030
Wright Pest Control, Inc.
P. 0. Box 2185
Winter Haven, FL 338800
822 No. Indiana Creek Dr. G-4
Clarkston, GA 30021
INDEX TO VOLUME 64
Acarina-147, 340, 542
Adams, C. T., et al., note by-199
Adams, G., article by-66
Aiken, R. B., article by-267
Anderson, D. M., et al., article by-119
Andrews, K. L., et al., article by-147
Annual meeting-203, 383, 461, 548
Anticarsia gemmatilis-529, 545
ants-138, 454, 545
Araneae-457, 276, 457
Averill, A. L., et al., article by-222
Baer, R. G., et al., article by-226
Balciunas, J. K., et al., note by-195
Baranowski, R. M., et al., note by-197
BAY SIR 8514-361
bean leaf beetle-538
Beavers, J. B., et al., note by-361
behavior-i, 4, 18, 30, 44, 66, 74, 89,
Bennett, F. D., et al., note by-197
Benschoter, C. A., et al., note by-543
Beugnon, G., article by-463
book notices, reviews-379, 460
Brookhart, J. 0., et al., article by-283
Burk, T., article by-30
Central America-147, 192, 412
Clark, R. K., Jr., obituary-382
Clark, W. E., article by-217
Coleoptera-66, 119, 138, 197, 217, 337,
control, insecticidal-472, 538
Crocker, R. L., et al., note by-542
Curtis, G. A., et al., article by-491
Dakin, M. E., Jr., et al., note by-454
Denmark, H. A., et al., article by-147,
Diptera-30, 158, 456, 468, 477, 491,
Donnelly, T. W., article by-412
Index to Volume 64
Eli Lilly EL-468-472
Entomological Society of America,
Esenbeckia (Proboscoides) -158
Fairchild, G. B., et al., article by-158
false spider mite-542
Fatzinger, C. W., article by-506
Felder, D. L., et al., note by-454
Florida-119, 138, 189, 197, 207, 244,
Florida Entomological Club-358
Formicidae-138, 454, 472, 545
Frank, J. H., et al., article by-138,
Glancey, B. M., et al., notes by-194,
Glover, A., et al., note by-454
Gordon, R. D., et al., article by-119
Greenfield, M. D., article by-4
Habeck, D. H., et al., note by-195
Harper, J. D., note by-538
Hassey, M., et al., article by-246
Hemiptera-194, 246, 519
Heppner, J. B., articles by-271, 309
Hetrick, L. A., note by-361
host plant-468, 477, 506, 542
Hymenoptera-74, 89, 138, 175, 192,
insect behavioral ecology-1
Ivie, M. A., et al., note by-542
Johnson, N. F., et al., note by-542
Johnson, D. W., et al., article by-529
Khalaf, K. T., note by-534
Knoop, W. E., et al., note by-542
Kosztarab, M., et al., article by-226
Kurczewski, F. E., article by-424
Lepidoptera-4, 18, 183,192, 195, 271,
Leppla, N. C., Presidential Address-
Lloyd, J. E., et al., photo story by-459
article by-1, 89
Lofgren, C. S., et al., note by-199, 454
-- et al., article by-472
Lower Rio Grande Valley-437, 441
Florida Entomologist 64(4)
Lucas, J., et al., article by-207
mating-4, 30, 44, 74, 89, 506
Matthysse, J. G., et al., article by-340
McCoy, E. D., et al., article by-405
McMillian, W. W., et al., article by-
---, et al., note by-198
meeting notices-383, 384
minutes, 63rd meeting-364
moths-4, 183, 195, 455, 506, 515, 529,
Muchmore, W. B., article by-189
Muma, M. H., article by-276
et al., article by-283
Noctuidae-183, 192, 395, 515
Owens, J. M., et al., note by-542
parasitism-192, 468, 477, 534, 542
Parkman, P., et al., note by-192
Peck, S. B., article by-519
pheromones-4, 222, 529
photo story-363, 459
Poe, S. L., et al., article by-477
Poole, R. W., et al., article by-183
Porter, C. C., articles by-175, 235,
260, 437, 441
Prokopy, R. J., et al., article by-222
Ray, D. H., et al., article by-385
red imported fire ant-194, 199, 454,
Reeder, W. F., et al., note by-543
Reiskind, J., note by-457
resistance, to insects-515
Rey, J. R., et al., article by-405
Robinson, G., note by-456
Sakimura, K., article by-483
Index to Volume 64
Schroeder, W. J., et al., note by-361
Schuster, J. C., note by-192
searching-18, 44, 66, 89, 529
sex pheromones-455, 529
sexual selection-4, 30, 44, 66, 74, 89
Shepard, M., et al., note by-192
Sivinski, J., note by-541
Slater, J. A., et al., article by-246
Snider, R., article by-244, 417
Solenopsis-194, 199, 454, 472, 545
southern mole cricket-463
southern pine coneworm-506
Spalding, D. H., et al., note by-543
Stange, L. A., et al., article by-207
Staphylinidae-138, 337, 399
Stark, B. P., et al., article by-468
Stiling, P. D., article by-468
Strassman, J. E., article by-74
Sullivan, R. T., article by-44
Texas-175, 260, 437
Thomas, M. C., et al., article by-138,
Todd, E. L., et al., article by-183
Tryon, E. H., Jr., et al., article by-477
Vander Meer, R. K., et al., note by-
velvetbean caterpillar-529, 538, 545
Waddill, V., note by-545
Walker, T. J., et al., article by-18
Wharton, R., note by-194
Widstrom, N. W., et al., article by-
---, et al., note by-198
Wilkerson, R. C., et al., article by-158
Williams, D. F., et al., article by-472
---, et al., note by-545
Wineriter, S. A., et al., article by-18
Wing, S., et al., photo story by-459
Wiseman, B. R., et al., articles by-
--- et al., note by-198