Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00008
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1995
Copyright Date: 1917
Subject: Florida Entomological Society
Entomology -- Periodicals
Insects -- Florida
Insects -- Florida -- Periodicals
Insects -- Periodicals
General Note: Eigenfactor: Florida Entomologist: http://www.bioone.org/doi/full/10.1653/024.092.0401
 Record Information
Bibliographic ID: UF00098813
Volume ID: VID00008
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access
Resource Identifier: isbn - 0015-4040
issn - 1938-5102
oclc - 33223434

Full Text

Medal et al.: Predation of Spissistilus festinus 561


'University of Arkansas, Department of Entomology, Fayetteville, AR 72701

2University of Arkansas, Agricultural Statistics Laboratory
Fayetteville, AR 72701


The developmental stages of Spissistilus festinus most susceptible to predation by
young female adults of Geocoris punctipes and Nabis roseipennis and to Orius insid-
iosus of undetermined age and sex was determined in the laboratory. With G. puncti-
pes, the highest S. festinus mortality (90-100%) occurred in the early (1st, 2nd)
nymphal stages while Nabis roseipennis attacked all nymphal stages equally well. In
general, 0. insidiosus did not feed on S. festinus. This study suggests that S. festinus
nymphs are potential prey for G. punctipes and N. roseipennis in the field.

Key Words: Geocoris punctipes, Nabis roseipennis, biological control.


En pruebas de laboratorio fueron determinados los estados de desarrolo de Spis-
sistilus festinus mas susceptibles a la depredaci6n por hembras adults de Geocoris
punctipes yNabis roseipennis asi como por Orius insidiosus de edad y sexo indetermi-
nados. Con G. punctipes, la mortalidad de S. festinus mas alta (90-100%) ocurri6 en los
estados ninfales tempranos (1 y 2 ) mientras N. roseipennis atac6 todos los estados
ninfales de la misma manera. En general, 0. insidiosus no se aliment6 de S. festinus.
Este studio sugiere que las ninfas de S. festinus son presa potential de G. punctipes
y N. roseipennis en el campo.

The threecornered alfalfa hopper, Spissistilus festinus (Say), is considered a pest of
economic importance in soybean, C(. .... max (L.), and other legume crops in several
southeastern states. Researchers in Arkansas (Mueller & Dumas 1975) and Louisi-
ana (Sparks & Newson 1984, Sparks & Boethel 1987) report yield losses resulting
from feeding by adults and nymphs of this pest. The main damage is caused when the
base of the main stem is girdled resulting in dead or weakened plants.
The importance of entomophagous arthropods in preventing the increase of lepi-
dopterous pest populations in field crops, including soybean, has long been recognized
(Lopez et al. 1976, Lawrence & Watson 1979, Richman et al. 1980). A parasitoid (Jor-
dan 1952, Herting & Simmonds 1972) and several predators (Spurgeon 1992) ofS. fes-
tinus have been reported. Polyphagous predators such as Geocoris spp., Nabis spp.,
and Orius insidiosus (Say), which are frequently abundant in many soybean-growing
areas of the United States (Roach 1980, Barry 1973, Bell & Whitcomb 1963), may sup-
press S. festinus populations in soybeans. However, the impact of predators on this in-
sect pest has not been evaluated.

Florida Entomologist 78(4)

A laboratory study was designed to determine the developmental stages of S. fes-
tinus most susceptible to predation by adults of Geocoris punctipes (Say), Nabis
roseipennis Reuter, and 0. insidiosus.


Spissistilus festinus were obtained from a laboratory colony maintained on
Phaseolus vulgaris L. pods at 26 1 C, 70 to 80% RH, and a photoperiod of 14:10
(L:D). The colony was revitalized periodically with field-collected adults to overcome
adverse selection effects of laboratory rearing.
Nymphs of G. punctipes and N. roseipennis were collected with sweep nets in al-
falfa, Medicago sativa L., fields in southwestern Arkansas during the spring-summer
of 1992. They were fed in the laboratory until the adult stage on second instar soybean
looper, Pseudoplusia includes (Walker). Green bean pods were also provided. Orius
insidiosus adults of undetermined age and sex were collected from an alfalfa field at
the University of Arkansas experimental farm at Fayetteville. Before the experiment,
0. insidiosus were fed Helicoverpa zea (Boddie) eggs and green bean pods. The three
to five-week-old female G. punctipes and N. roseipennis and adult 0. insidiosus were
starved for 24h before the experiment.
Potted V2 soybean plants (CV:Bragg) were covered by cages and placed in an en-
vironmental chamber at 25 + 1 C, 60% RH and a photoperiod of 14:10 (L:D). Cages
were made by cutting off both ends of two-liter clear plastic soda bottles and gluing a
screen cloth at the top end to allow air movement. The soil at the base of the plants
was covered with a brown-paper disc to facilitate finding dead insects. The base of the
cage in contact with the soil was sealed by placing tape around the bottom of the cage
and the upper rim of the pot.
A completely randomized design with 10 replications was used. Treatments were
three predator species (G. punctipes, N. roseipennis, and 0. insidiosus) and six S. fes-
tinus developmental stages nymphall instars 1-5 and adult). Spissistilus festinus eggs
were not tested. One starved adult predator was placed with one S. festinus of a single
developmental stage (1 predator vs. 1 prey) on a potted soybean plant. Cages contain-
ing only plant and prey with no predator served as controls. Mortality was recorded
after 24 h. Any dead S. festinus were assumed to have been killed by the predator.
Mortality of S. festinus was not adjusted for the control because of low (<5%) mortality
in controls. Values of proportion dead were analyzed using a Chi-square test. Means
were compared by a two-sample binomial test (Ott 1984).


Significant differences in percentage mortality among the S. festinus developmen-
tal stages between G. punctipes and N. roseipennis were observed (Chi-square=9.52,
df=4, P<0.05). Orius insidiosus fed minimally (10% mortality) on the first nymphal
stage and did not feed on larger nymphal or adult S. festinus stages. Values for 0. in-
sidiosus were not included in the Chi-square test because of the low frequency of mor-
tality (<1) observed for all prey developmental stages. These results suggest that
Orius insidiosus is probably not an important predator of S. festinus in nature. The
highest S. festinus mortality (100%) was observed when first instars were exposed to
G. punctipes (Table 1). This mortality was not significantly different (P=0.05, two-
sample binomial test) from the mortality (90%) obtained when second and third
nymphal stages were exposed to G. punctipes and N. roseipennis predators, respec-
tively. Nabis roseipennis attacked all nymphal stages equally well (P=0.05, two-sam-

December, 1995

Medal et al.: Predation of Spissistilus festinus 563


Prey Developmental Stage

Predator N1 N2 N3 N4 N5 Adult

Geocoris 100 a' 90 ab 60 b 10 cd 0 d 0 d
Nabis 60 b 60 b 90 ab 50 be 50 bc 10 cd
Orius 10 0 0 0 0 0

'Values followed by the same letter do not differ at the 0.05 probability level using a two-sample binomial test
for equal proportions. Comparisons were carried out between Geocoris and Nabis predators and prey stages
within predator species indicated.
Values for Orius were not included in the Chi-square test because of the low frequency of mortality (<1) ob-

ple binomial test). Nabis roseipennis was the only predator that fed on adult S.
festinus (10% killed).
Geocoris punctipes caused the highest mortality of nymphal instars (first through
third), which are probably the most susceptible to attack (Table 1). Crocker & Whit-
comb (1980) found that the largest percentage (79%) of target prey captured under
natural conditions by Geocoris spp. are those that remain passive during physical con-
tact with the predator. Early and intermediate nymphal stages are probably not able
to adequately defend themselves and their strategy of remaining motionless for cer-
tain periods of time is not an effective behavior for avoiding predation. The lower pre-
dation on fourth and fifth instars by G. punctipes can be attributed to the more active
physical movements of the prey during predator-prey encounters and to their more
well-developed spines. Further biological studies on prey-predator interactions under
laboratory and field conditions will provide basic information for developing predic-
tive models on population dynamics of pests and natural enemies that can be used in
pest management programs. These studies indicate that because G. punctipes and N.
roseipennis fed on S. festinus nymphs in the laboratory they are potential predators
for this pest in nature.


We thank D. T. Johnson, W C. Yearian and S. Y. Young (Department of Entomology,
University of Arkansas) for reviewing the manuscript. This research was supported in
part by an Arkansas Soybean Promotion Board grant. Article published with the ap-
proval of the Director, Arkansas Agricultural Experiment Station, University of Ar-
kansas, Fayetteville, manuscript No. 95002.


BARRY, R. M. 1973. A note on the species composition of predators in Missouri soy-
beans. J. Georgia Entomol. Soc. 80: 284-286.
BELL, K. 0., AND W. H. WHITCOMB. 1963. Field studies on egg predators of the boll
worm, Heliothis zea. Florida Entomol. 47: 171-180.
CROCKER, K. 0., AND W. H. WHITCOMB. 1980. Feeding niches of the big-eyed bugs Geo-
coris bullatus, G. punctipes and G. uliginosus (Hemiptera: Lygaeidae). Environ.
Entomol. 9: 508-513.

Florida Entomologist 78(4)

HERTING, B., AND F. J. SIMMONDS. 1972. A catalogue of parasites and predators of ter-
restrial arthropods. Section A. Host or prey/enemy. Vol. II. Homoptera. CIBC.
210 p.
JORDAN, C. R. 1952. The biology and control of the threecornered alfalfa hopper, Spis-
sistilus festinus (Say). Ph.D. dissertation, Texas A&M University, College Sta-
LAWRENCE, R. K., AND T. F. WATSON. 1979. Predator-prey relationship of Geocoris
punctipes and Heliothis virescens. Environ. Entomol. 2: 245-248.
LOPEZ, J. D., R. L. RIDGWAY, AND R. E. PINNELL. 1976. Comparative efficacy of four in-
sect predators of the bollworm and tobacco budworm. Environ. Entomol. 5:
MUELLER, A. J., AND B. A. DUMAS. 1975. Effects of stem girdling by the threecornered
alfalfa hopper on soybean yields. J. Econ. Entomol. 4: 511-512.
OTT, L. 1984. An introduction to statistical methods and data analysis. Duxbury
Press. Boston. 775 p.
RICHMAN, D. B., R. C. HEMENWAY, AND W. H. WHITCOMB. 1980. Field cage evaluation
of predators of the soybean looper, Pseudoplusia includes (Lepidoptera: Noc-
tuidae). Environ. Entomol. 3: 315-317.
ROACH, S. H. 1980. Arthropod predators on cotton, corn, tobacco, and soybeans in
South Carolina. J. Georgia. Entomol. Soc. 15: 131-138.
SPARKS, A. N., AND D. J. BOETHEL. 1987. Late-season damage to soybean by threecor-
nered alfalfa hopper (Homoptera: Membracidae) adults and nymphs. J. Econ.
Entomol. 80: 471-477.
SPARKS, A. N., AND L. D. NEWSOM. 1984. Evaluation of the pest status of Spissistilus
festinus (Say) (Homoptera: Membracidae) on soybean in Louisiana. J. Econ.
Entomol. 77: 1553-1558.
SPURGEON, D. W. 1992. Threecornered alfalfa hopper (Homoptera: Membracidae) on
soybean: Insect-plant interactions. Ph.D. dissertation, University of Arkansas,

December, 1995

Osborne et al.: Ant Predation on Mites


'Central Florida Research and Education Center,
University of Florida, IFAS
2807 Binion Road
Apopka, FL 32703

'Tropical Research and Education Center,
University of Florida, IFAS
18905 SW 280 Street,
Homestead, FL 33031

3Department of Entomology
301 Funchess Hall
Auburn University, AL 36849


Natural infestations of Tapinoma melanocephalum, (Fab.) found in central Flor-
ida greenhouses were observed attacking Tetranychus urticae Koch in test evalua-
tions of two chemicals. Subsequent laboratory tests using isolated leaf discs,
greenhouse data, and whole plants demonstrated that T melanocephalum is a signif-
icant predator of T urticae.

Key Words: Biological control, ants, spider mites.


Durante la evaluaci6n de dos products quimicios en invernaderos de la region
central de la Florida fueron encontradas infestaciones naturales de Tapinoma mela-
nocephalum (Fab.) atacando Tetranychus urticae Koch.
Las pruebas de laboratorio realizadas posteriormente usando discos de hojas, da-
tos de invernadero y plants completes demostraron que T melanocephalum es un de-
predador significativo de T urticae.

The first report of movement of beneficial insects to control a pest insect was by
Forskal (1775) who reported that date palms in the Middle East were protected from
pests by ants. Annually, colonies of predatory ants were moved from the mountains
into infested areas. Groff & Howard (1924) stated that chinese citrus growers of
Kwangtung province would commonly use colonies of the cultured ant, Oecophylla
smaragdina Fab., as predators for citrus pests.
Ants found in greenhouses are generally treated as pests because they tend
aphids, mealybugs or soft scales or they are of quarantine importance (i.e., red im-
ported fire ants, Solenopsis invicta Buren). In almost all cases, we have traditionally
recommended the use of a pesticide to kill the pest insect being tended by the ants or,

Florida Entomologist 78(4)

specifically, to kill the ants. Ants are not thought of as biological control agents in
greenhouses, rather, they are considered disruptive to biological control of various ho-
mopterous pests.
One species of ant that can be found in greenhouses is Tapinoma melanocephalum
(Fab.). This species is considered a nuisance ant and occasionally an important pest in
houses. Field populations are found in south Florida but are limited to greenhouses
and buildings in the north (Nickerson & Bloomcamp 1988). This tropical species has
been so widely distributed by commerce that it is difficult to determine its origin, but
it is believed to be African or Oriental (Smith 1965). The workers are monomorphic
and approximately 1 mm long. The worker head and thorax are black, legs and abdo-
men and gastor are clear or light yellow. Colonies of this highly adaptable species con-
tain multiple queens polygynyy) and are often found in transient habitats, such as
plant debris, under potted plants, rotting wood planks and plant stems (Nickerson &
Bloomcamp 1988). Colonies may break into subunits that occupy different sites, but
individual ants will move between them (Oster & Wilson 1978).
Smith (1965) reported that T. melanocephalum fed on many different foods in the
house, but seemed to prefer sweets. In the greenhouse, she reported that they fed on
honeydew and on live or dead insects. Nickerson & Bloomcamp (1988) noted that a
colony of T. melanocephalum had established in the quarantine greenhouse of the
Florida Department of Agriculture in Gainesville, Florida, where it preyed on small
beetle larvae and lepidopterous larvae from the insect cultures in quarantine. In Ven-
ezuela, T. melanocephalum is the primary predator of eggs of Rhodnius prolixus Stal
(Gomez-Nunez 1971). The only report of this ant interacting with Arachnida was by
Shepard & Gibson (1972) who reported that a symbiotic relationship had developed in
Costa Rica between T. melanocephalum and jumping spiders (Araneae: Salticidae)
which were found inhabiting the nests. It is believed that the spiders provided some
protection to the ants from natural enemies and the nest served as a foundation for
web construction.
In this paper, we report chemical evaluations conducted in 1993 for control of
twospotted spider mites, Tetranychus urticae Koch, in the University of Florida-Apo-
pka greenhouses. The tests results led us to the conclusion that mite numbers de-
clined significantly on both treated and untreated plants because of predation by T.
melanocephalum. This paper describes this preliminary experiment and other tests to
define the interaction between T. urticae and T. melanocephalum.


Pesticide Trials

Salvia splendens cv. Red Hot Sally seedlings were potted into 15.2-cm round plas-
tic pots using Verlite Nursery Mix A (without superphosphate; Verlite Co. Tampa, FL)
amended with 4.4 kg Osmocote (19:6:12 slow release fertilizer; Grace-Sierra Chemical
Co., Milpitas, CA), 4.2 kg dolomite, and 0.9 kg Micromax per m3 (micronutrient source;
Grace-Sierra Chemical Co.). The pots were placed on raised benches in a glass green-
house. Forty plants were infested on September 16, 1993 with twospotted spider
mites (TSM) by placing pieces of infested Henderson Bush lima bean (Phaseolus li-
mensis Macfady) leaves on each plant. An initial count of all TSM stages on two
marked leaves per plant was conducted and five plants were assigned to each of five
treatments. Additional counts were conducted five days later. The five treatments
were: three with an experimental formulation containing neem oil supplied to us un-
der a non-disclosure agreement (UK-1%, UK-1.5%, UK-2%), the labeled rate of a stan-
dard acaricide (Pentac Aquaflow dienochlor) and an untreated control.

December, 1995

Osborne et al.: Ant Predation on Mites 567

Disc Trials

Henderson bush lima bean seeds were planted into 12.7-cm square black plastic
pots using the same soil described above. Five 20-mm diam discs were cut from ma-
ture leaves and placed on the bottoms of 10 cm2 plastic-petri dishes filled with moist
cotton. The pattern of placement was one disc at each corner of a square for the first
test; for the second and third tests one additional disc (total=5) was placed in the cen-
ter of this pattern. Five adult female TSM were placed on each disc. Two dishes were
prepared for each experiment. Half the dishes were individually isolated by placing
them in a saucer on an inverted 10-cm square plastic pot sitting in a second saucer
filled with soapy water. This served as an efficient method to prevent ants from ob-
taining access to mites on these discs. The remaining dishes were placed on a green-
house bench on which worker ants were observed foraging. The number of live TSM
(all stages) were counted on each leaf on days 1, 2, 3, and 6. This study was repeated
three times. All data from each test were pooled and the median number of mites per
disc for each treatment was compared using Mann-Whitney Rank Sum Test (SigmaS-
tat version 1.01; Jandel Scientific, San Rafael, CA, 1994).

Whole Plant Trials

Twenty Henderson bush lima bean seeds were planted into 12.7-cm square black
plastic pots using the same soil previously described. When the first true leaves were
fully expanded (two-leaf stage), each plant was infested with 10 adult female TSM per
leaf by hand with a camel-hair brush. Ten plants were individually isolated by placing
each pot in a saucer on an inverted 10-cm square plastic pot resting in a second saucer
filled with soapy water. This served as an efficient method to prevent ants from forag-
ing on these test plants. The remaining 10 plants were placed on a greenhouse bench
on which worker ants were observed foraging.
The numbers of live TSM (all stages) were counted on each leaf on day 7, 11, 14,
18, and 22. This study was repeated three times. All data from each test were pooled
and the median number of mites per leaf for each treatment was compared using
Mann-Whitney Rank Sum Test (SigmaStat version 1.01; Jandel Scientific, San
Rafael, CA, 1994).


Treatment Pre-count 7-Day Count 14-Day Count

UK-1.0%1 51.4 + 13.02 32.2 + 28.7 1.6 1.4
UK-1.5% 51.0 + 13.4 34.8 + 11.7 12.0 + 4.9
UK-2.0% 51.2 12.9 47.8 16.1 34.2 14.7
Dienochlor 51.0 12.6 15.0 6.9 0.0 + 0.0
Control 43.8 + 19.1 11.2 + 6.9 0.0 + 0.0

'This experiment consisted of five treatments, three with an experimental neem oil formulation of unknown
composition and supplied to us under a non-disclosure agreement (UK-1%, UK-1.5%, UK-2%), the labeled rate
of a acaricide standard (Pentac Aquaflow-dienochlor) and an untreated control.

Florida Entomologist 78(4)


Pesticide Trials
The results of the pesticide trials were contrary to those expected for a standard
pesticide efficacy study (Table 1), because mite numbers on all treatments declined
compared to the initial counts. In the three treatments where neem oil was applied, the
number of mites increased with the concentration applied. These results prompted us
to inspect all plants for the presence of predatory mites. No phytoseiid mites were
found on any treatments. We did notice a significant number of ants on many plants in
this study and on closer observation we found worker ants leaving plants with mites,
adult whiteflies, thrips and immature aphids in their mandibles. This observation
prompted us to conduct the disc and individual plant studies reported below.

Disc Trials
Results of the disc tests are presented in Fig. 1. Number of mites on each disc were
significantly different (P<0.01; Mann-Whitney Rank Sum Test) on days 1, 2, 3, and 7.

80 -1

0 1 2 3 4 5 6 7 8

Figure 1. A comparison of the number of mites (mean SE) per leaf disc when ants
were allowed to forage versus when they were excluded; plotted against time (day 1,
2, 3, and 7).

December, 1995

Osborne et al.: Ant Predation on Mites

Worker ants were observed walking across the wet cotton substrate used to maintain
the disc and prevent mites from escaping. They were never observed carrying eggs;
however, they were seen removing other stages of T. urticae from these discs. No ants
were observed on the isolated treatments. It appeared that the ants were able to move
across the moist cotton, find the mites, and remove them within 24 h from the time the
dishes were placed in the infested greenhouse.

Whole Plant Trials
Results of the individual plant tests are presented in Fig. 2. Numbers of mites on
each leaf were significantly different (P<0.01; Mann-Whitney Rank Sum Test) on days
7, 11, 14, 18, and 22. Ants were observed foraging on the exposed plants after 1 day of
exposure. Numbers of mites increased until day 11 after which the population de-
clined to a very low level. The age distribution of surviving mites changed throughout
the course of this study. Initially, the active stages were mainly adult females but
most of the mites counted after day 14 were nymphs or newly hatched larvae. We

350 -

300 -

250 -

S200 -
S150 -




15 20 25

Figure 2. A comparison of the number of mites (mean SE) per leaf when ants
were allowed to forage versus when they were excluded; plotted against time (day 7,
11, 14, 18, and 22).

/ \





570 Florida Entomologist 78(4) December, 1995

never observed a reduction in the numbers of mite eggs nor did we observe ants re-
moving eggs from the discs or whole plants.
T melanocephalum is usually considered to be a pest because of its impact on mite
colonies and its disruptive effect on experiments by feeding on the mites in controls.
Results of these studies show that T melanocephalum can be a significant predator of
twospotted spider mites; however, there are several important questions that must be
addressed before employing these ants as predators in a commercial setting. Exten-
sive studies should be conducted to learn what foods they will feed on under green-
house conditions. This would include extensive host range studies to determine if they
pose any threat to ornamental plants. Because we have observed T melanocephalum
feeding on a number of other arthropod prey, we also must determine if they have a
preference when given the option of feeding on a array of different prey. Secondly, we
would like to determine if T melanocephalum only feed on twospotted spider mites
when other food sources are scarce. We have noticed that they will feed on western
flower thrips and Ecinothrips americanus when no aphids or spider mites are present,
but when these hosts are in sufficient numbers the ants seem not to attack thrips.
The final step would be to determine their sensitivity to pesticides so that they
could be eradicated from a greenhouse if necessary, or to prevent their movement in
or on ornamental plants. T melanocephalum is considered a "fugitive" or opportunis-
tic ant (Hdlldobler & Wilson 1990) and its status as an urban pest cannot be ignored.
Because this ant can infest potted plant material, cardboard boxes, and other mate-
rial commonly used to ship plants, it may be necessary to develop management pro-
grams for control.


We wish to extend our appreciation to Drs. R. K Yokomi, G. L. Leibee and F. L.
Petitt for their critical review of this work. This is Florida Experiment Station Journal
Series No. R-04739.


FORSKAL, P. 1775. Descriptiones animalium, avium, amphibiorum, piscium, insecto-
rum, vermium: quae in itinere oriental observavit P. Forskal, post mortem auc-
toris edidit Carsten Niebuhr. Hauniae, Moeller (Pt. 3).
GOMEZ-NUNEZ, J. C. 1971. Tapinoma melanocephalum as an inhibitor of Rhodnius
prolixus populations. J. Med. Entomol. 8: 735-737.
GROFF, G. W., AND C. W. HOWARD. 1924. The cultured citrus of south China. Lingnan
Agr. Rev. 2: 108-14.
HOLLDOBLER, B., AND E. O. WILSON. 1990. The Ants. Belknap Press of Harvard Uni-
versity Press, Cambridge, Mass. 732 pp.
NICKERSON, J. C., AND C. L. BLOOMCAMP. 1988. Tapinoma melanocephalum (Fabri-
cius) (Hymenoptera: Formicidae). Entomol. Cir. No. 307. Florida Dept. Agric. &
Consumer Serv. Div. of Plant Industry. Gainesville, FL. 2 p.
OSTER, G. F., AND E. O. WILSON. 1978. Caste and Ecology in the Social Insects. Prin-
ceton Univ. Press, Princeton, New Jersey. 352 p.
SHEPARD, M., AND F. GIBSON. 1972. Spider-ant symbiosis: Cotinusa spp. (Araneida:
Salticidae) and Tapinoma melanocephalum (Hymenoptera: Formicidae). Cana-
dian Entomol. 104: 1951-1954.
SMITH, M. R. 1965. House-investing ants of the eastern United States; their recogni-
tion, biology, and economic importance. USDA-ARS Tech. Bull. 1326. 105 p.

Robacker: Attractant for Mexican Fruit Fly


Crop Quality and Fruit Insects Research
Agricultural Research Service, U.S. Department of Agriculture
2301 S. International Blvd.
Weslaco, TX 78596


A mixture of ammonium bicarbonate or ammonium carbonate, methylamine HC1
and putrescine (AMPu) was evaluated for attractiveness to gamma-irradiated Mexi-
can fruit flies, Anastrepha ludens (Loew), in a citrus orchard in 1-day tests. AMPu
(10:10:1 mixture of ammonium bicarbonate:methylamine HC1:putrescine) was tested
both in dilute aqueous solutions in the reservoir of McPhail traps and in more concen-
trated form in polypropylene tubes suspended in McPhail traps or fastened to yellow
sticky ball traps. The most attractive concentration of AMPu used in aqueous solution
captured only half as many flies as Torula yeast in McPhail traps. AMPu (6:10:1 mix-
ture of ammonium carbonate:methylamine HC1:putrescine) formulated into agar and
tested in tubes fastened to sticky ball traps captured as many male and female flies
as Torula yeast in McPhail traps.

Key Words: Anastrepha ludens, trapping, lures, amines.


Fue evaluada la atractividad de una mezcla de bicarbonato de amonio o carbonate
de amonio, metilamina HC1 y putrescina (AMPu) sobre moscas mexicanas de la fruta
gamma-irradiadas,Anastrepha ludens (Loew), en un campo de citricos en pruebas de
un dia de duraci6n. La mezcla AMPu (mezcla de 10:10:1 de bicarbonato de amonio:me-
tilamina HCl:putrescina) fue probada en soluciones acuosas diluidas en el reservorio
de trampas de McPhail y en forma mas concentrada en tubos de polipropileno suspen-
didos en trampas de McPhail o atados a trampas pegajosas de bolas amarillas. La con-
centraci6n mas atractiva de AMPu, usada en soluci6n acuosa, capture solamente la
mitad de las moscas capturadas con levadura Torula en las trampas de McPhail. La
mezcla AMPu (mezcla de 6:10:1, de carbonate de amonio:metilamina HCl:putrescina)
fromulada en agar y probada en tubos atados a trampas pegajosas de bola capture
tantos machos y hembras como la levadura Torula en las trampas de McPhail.

Recently, I reported on development of a three-component attractant (AMPu) for
the Mexican fruit fly (Anastrepha ludens (Loew)) containing metabolites from biolog-
ical degradation of amino acids (Robacker & Warfield 1993). AMPu was more attrac-
tive than Torula yeast, the most commonly used bait for Mexican fruit fly, in flight
chamber tests in a greenhouse. This research was analogous to work by Wakabayashi
& Cunningham (1991) with the melon fly (Bactrocera cucurbitae (Coquillett)). More
recently, Heath et al. (1995) have reported on a similar attractant for the Mediterra-
nean fruit fly (Ceratitis capitata (Wiedemann)) and the Mexican fruit fly.
The purpose of the current research was to evaluate the initial attractiveness of
AMPu compared to Torula yeast in a citrus orchard. Various concentrations of AMPu

Florida Entomologist 78(4)

were tested in McPhail traps and yellow sticky ball traps. The ultimate goal of this re-
search is to develop an attractant that can be used with a dry trap to replace the
McPhail trap/Torula yeast trapping system.



The test flies originated from a culture from Morelos, Mexico, that had been main-
tained on laboratory diet for about 400 generations with no wild-fly introductions. A
recent study has shown that the mating behavior of this culture has changed only
slightly from that of wild flies (Robacker et al. 1991). Likewise, I presume that the re-
sponse of laboratory-reared flies to food-based attractants does not differ markedly
from that of wild flies.
Flies were irradiated with 7000-9200 rads (Cobalt 60 source) 1-2 days before adult
eclosion, to comply with quarantine laws for releasingA. ludens. Irradiated flies were
shown to be 20% less responsive than unirradiated flies to bacterial odor (Robacker &
Garcia 1993), the attractant on which development of AMPu was partially based.
Mixed-sex groups of 180-200 flies were kept in 473 ml cardboard cartons with
screen tops until used in tests. Laboratory conditions for holding flies were 22 2 C,
50 + 20% relative humidity and photophase from 0630 to 1930 hours provided by flu-
orescent lights. Flies were fed sucrose and water up until the time of release.

Citrus Orchard and Test Procedures

A mixed citrus orchard located near the laboratory in Weslaco, Texas, was used for
all experiments. The orchard contained several varieties of orange, lemon, grapefruit
and tangerine trees of varying ages. One row of ruby red grapefruit (Citrus paradisi)
and one row of Dancy tangerine (C. reticulata) were chosen for tests since they con-
tained relatively large (2-3 m height) fruit-bearing trees. Two blocks of seven consec-
utive trees were used in each row, for a total of four blocks in the orchard.
Flies were released into the test orchard when 2-10 days old during the late after-
noon of the day before a test. Robacker & Garcia (1993) showed that, within this age
range, fly age had little effect on attraction to bacterial odor, an attractant similar to
AMPu as discussed above. Approximately 2000 flies were distributed equally among
the 28 test trees in the four blocks. Attractants were tested either in McPhail traps
(Baker et al. 1944) or on sticky yellow ball traps (Robacker 1992) that were hung one
to a tree, north of center, at 1-2 m height. Traps were placed in the orchard during the
morning and removed for fly counts and cleaning on the following morning. Positions
of treatments within each block of trees were randomized for the first 1-day test of
most experiments. Positions of treatments in consecutive 1-day tests were not ran-
domized but were moved sequentially within each block.
Experiments 1 and 2 were conducted to determine the most attractive concentra-
tion of AMPu in aqueous solution in McPhail traps. The AMPu tested was an aqueous
solution of ammonium bicarbonate, methylamine HC1, and putrescine in the ratio
10:10:1 at pH 8.8 (adjusted with NaOH). The three chemicals were obtained from
Sigma Chemical Co. (St. Louis, MO) and were >98% pure.
Experiment 1 evaluated three concentrations of AMPu. AMPu 2000 was prepared
with the three chemicals at 2:2:0.2 mg/ml. AMPu 200 and AMPu 20 were 1:10 and
1:100 dilutions of AMPu 2000. Each AMPu trap contained 200 ml of one of these so-
lutions in the trap's liquid reservoir. Traps baited with three Torula yeast/borax "bait

December, 1995

Robacker: Attractant for Mexican Fruit Fly 573

pellets" (Sit-Khem Corp., Michigan City, IN) and 200 ml of tap water were included in
this and in Experiments 2 & 5 as attractiveness standards. Traps containing only 200
ml of water were used as blanks. All traps except Torula yeast contained amber color-
ing from a combination of red, yellow and green food colors (McCormick & Co., Inc.,
Baltimore, MD) to mimic the color of Torula yeast. These traps also contained 0.01%
Triton (Rohm and Haas Co., Philadelphia, PA) as a wetting agent. One each of the
three AMPu traps, one Torula yeast trap and one water-blank trap were included in
each of the four blocks of trees. Five 1-day tests were conducted for a total of 20 repli-
cations of each bait treatment. Baits were discarded after each 1-day test.
Experiment 2 compared AMPu 20 and AMPu 200 to three additional concentra-
tions: AMPu 400; AMPu 100; and AMPu 50. The additional concentrations were 1:5,
1:20 and 1:40 dilutions, respectively, of AMPu 2000. Experiment 2 was identical to Ex-
periment 1 except that each block contained five AMPu treatments instead of three.
Seven 1-day tests were conducted for a total of 28 replications of each bait treatment.
Experiment 3 was conducted to develop a formulation of AMPu that could be used
on a dry sticky trap. For this purpose, 1.9 ml polypropylene microcentrifuge tubes (A.
Daigger & Company, Inc., Wheeling, IL) were used to hold various AMPu prepara-
tions. The objective was to determine how much AMPu was needed in the tubes to
make them competitive with the best concentration used in McPhail traps.
AMPu was prepared in water at the concentration of 20:20:2 mg/ml at pH 8.8.
Amounts ranging from 1.0-1.6 ml of this solution or various dilutions of the solution
ranging down to 2:2:0.2 mg/ml were put into microcentrifuge tubes. Concentrations
greater than 20:20:2 mg/ml were not used because all indications at the time this ex-
periment was conducted were that greater concentrations would be less attractive.
Tubes were suspended inside McPhail traps just above the inner rim and traps were
filled with 200 ml of amber-colored water as described above. These traps were tested
against McPhail traps containing 200 ml of AMPu 200 in a flight chamber in a green-
house as described by Robacker & Warfield (1993). AMPu 200 was used as the stan-
dard because of its great attractiveness in previous flight chamber tests (Robacker &
Warfield 1993). Briefly, the procedure was as follows. For each test, a trap containing
AMPu 200 and a trap containing an AMPu tube were suspended side by side in the
upwind end of the screened chamber (2.0 m long x 0.7 m wide x 1.3 m high) with an
airflow of 0.1 to 1.0 m/sec. Traps were alternated between left and right sides of the
chamber every 15 min for a 1-h test. Approximately 200 sugar-fed, protein-starved
flies were released into the downwind end of the chamber at the beginning of each
test. Four to 15 replications were conducted for each of seven concentrations tested in
the microcentrifuge tubes.
Experiment 4 evaluated AMPu tubes vs AMPu 400 in McPhail traps in the field.
The purpose of this experiment was to determine if the optimum formulation of AMPu
in microcentrifuge tubes that was developed in the greenhouse flight chamber would
be equivalent to AMPu 400 in McPhail traps in the field. AMPu 400 was used as the
standard because this was the most attractive concentration in previous field tests
(Experiment 2). AMPu tubes contained 1.6 ml of AMPu at the concentration of 20:20:2
mg/ml at pH 8.8. Three traps of each type were placed alternately on six trees in each
of the four blocks of trees used in previous tests. Two 1-day tests were conducted for
a total of 24 replications of each bait treatment.
Experiment 5 evaluated AMPu tubes on ball traps vs Torula yeast in McPhail
traps in the field. Initially, AMPu microcentrifuge tubes were prepared containing 1.6
ml of AMPu at the concentration of 20:20:2 mg/ml at pH 8.8, the formulation that
proved equal in attractiveness to AMPu 200 in greenhouse tests (Experiment 3) and
AMPu 400 in field tests (Experiment 4). Subsequently, AMPu solutions were mixed

Florida Entomologist 78(4)

with agar to increase durability of the AMPu tubes exposed to weather on tops of
sticky traps. Concentrations of AMPu used in agar preparations were higher than
20:20:2 mg/ml because the initial tests with aqueous formulations indicated that
20:20:2 mg/ml was too low. Final AMPu concentrations in agar tubes ranged from
100:100:10 to 225:225:22.5 mg/ml of the three components at pH 8.5 to 8.8. Ammo-
nium carbonate (A.C.S. Reagent quality; Aldrich Chemical Co. Inc., Milwaukee, WI)
was substituted for ammonium bicarbonate in some preparations, keeping the molar
concentration of ammonia equivalent to previous ammonium bicarbonate prepara-
tions. Agar (Bacto Agar, Difco Laboratories, Detroit, MI) concentrations ranged be-
tween 1-3% in final preparations. The AMPu/agar tubes were prepared by mixing hot
agar solution with aqueous AMPu solution in microcentrifuge tubes.
AMPu tubes were fastened, with their caps open, to the tops of yellow ball traps
(13 cm diam) that were coated with Tangle-Trap (Tanglefoot Company, Grand Rapids,
MI). These traps were described previously in tests without olfactory lures (Robacker
1992). Ball traps with AMPu tubes, unbaited ball traps, and McPhail traps containing
Torula yeast as described above were compared in 1-day tests. Two each of the three
treatments were used in each of the four blocks in the citrus orchard for a total of eight
traps of each treatment per 1-day test. Fourteen 1-day tests were conducted using var-
ious combinations of AMPu concentrations and agar percentages.

Statistical Analyses

Analysis of variance (ANOVA) was conducted for both males and females for Ex-
periments 1, 2 and 5. Means separations were done by Fisher's protected least signif-
icant difference (LSD) method. t-Tests were used to determine if the ratio of females
to males captured by AMPu traps was different from the ratio of flies captured in
Torula yeast traps and to compare specific pairs of treatments in Experiments 3, 4
and 5.


Experiments 1 and 2: AMPu Dosage-Response

The results of Experiment 1 are shown in Fig. 1. AMPu treatments were generally
more attractive than water but significantly less attractive than Torula yeast to both
males (F = 18.7; df = 4,91; P < 0.0001) and females (F = 15.8; df = 4,91; P < 0.001).
AMPu 200 was significantly more attractive than AMPu 2000 but was not signifi-
cantly more attractive than AMPu 20. These results suggested that the most attrac-
tive concentration was probably between AMPu 20 and AMPu 2000.
The results of Experiment 2 are shown in Fig. 2. All AMPu treatments were sig-
nificantly more attractive than water to both males (F = 10.9; df = 6,189; P < 0.01) and
females (F = 12.0; df = 6,189; P < 0.01). Torula yeast was significantly more attractive
to both males and females than all AMPu concentrations except AMPu 400 for males.
Among AMPu treatments, AMPu 400 captured the most flies although it was not sig-
nificantly more attractive than most of the other AMPu treatments. In both Experi-
ments 1 and 2, the ratio of females to males captured by AMPu traps did not differ
significantly from the ratio of flies captured by Torula yeast traps.

Experiment 3: Optimum Concentration of AMPu in Microcentrifuge Tubes

Only the microcentrifuge tube containing 1.6 ml of AMPu at the concentration of
20:20:2 mg/ml was comparable in attractiveness to the McPhail trap containing

December, 1995

Robacker: Attractant for Mexican Fruit Fly

a a


Torula Yeast

[ males


b bc


AMPu 20 AMPu 200 AMPu


Figure 1. Mean captures ( SE; n = 20 each trap bait) of adultA. ludens in McPhail
traps containing water, Torula yeast or three concentrations of AMPu (Experiment 1).
Within each sex, bars with the same letter are not significantly different from each
other by Fisher's protected LSD (P < 0.05).

AMPu 200. Mean captures by traps containing these AMPu tubes (n = 12) were 23.8
2.2 (SE) flies. Mean captures by traps containing AMPu 200 (n = 12) were 21.6 + 1.7

Experiment 4: Field Evaluation of AMPu Tubes vs AMPu 400 in McPhail Traps

Traps containing AMPu tubes at 20:20:2 mg/ml of the three components were
equal in attractiveness to traps containing AMPu 400. Mean captures by traps con-
taining AMPu tubes (n = 24) were 19.4 3.1 (SE) flies. Mean captures by traps con-
taining AMPu 400 (n = 24) were 17.2 2.2 flies.

Experiment 5. Field Evaluation of AMPu Tubes on Ball Traps vs Torula Yeast in
McPhail Traps

Ball traps with AMPu tubes containing aqueous AMPu at 20:20:2 mg/ml were
much less attractive than McPhail traps containing Torula yeast. Mean captures on
balls with AMPu tubes (n = 16) were 12.9 + 1.7 (SE) flies compared with 28.9 4.4 flies
captured by Torula yeast traps (n = 16). The difference was highly significant (t = 3.3;
df = 30; P < 0.01). Ball traps with AMPu tubes were much more attractive than un-
baited balls (n = 16) which captured only 1.5 0.4 flies per trap.
Ball traps with AMPu tubes containing AMPu in agar were competitive with
McPhail traps containing Torula yeast. No consistent differences were observed for
the various AMPu and agar concentrations, so they were combined for analysis. Table
1 shows captures of flies by the traps, summed over all AMPu and agar concentra-
tions. Captures by unbaited ball traps were significantly lower than captures by the

Florida Entomologist 78(4)


Trap/Bait Males' Females' Males + Females'

Ball 0.21 + 0.05 a 0.30 + 0.06 a 0.51 + 0.08 a
Ball/AMPu Tube 1.83 0.33 b 3.15 0.34 b 4.97 0.61 b
McPhail/Torula Yeast 2.16 0.25 b 3.21 0.37 b 5.37 0.56 b

'Means in a column followed by the same letter are not significantly different by Fisher's protected LSD.

other two traps for males (F = 22.0; df = 2,314; P < 0.0001), females (F = 43.2; df =
2,314; P < 0.0001), and males + females (F = 39.6; df = 2,314; P < 0.0001). Captures of
males, females and total flies by AMPu traps and Torula yeast traps were not signifi-
cantly different at the 5% level.


Wakabayashi & Cunningham (1991) developed a chemically defined attractant for
the melon fly (B. cucurbitae) for use in McPhail traps. Their attractant was an aque-
ous preparation of ammonium bicarbonate, linolenic acid, putrescine and pyrrolidine
in which concentrations of ammonium bicarbonate and putrescine were similar to
those in AMPu 400. Wakabayashi & Cunningham's attractant was more attractive

6.0 males
d U females


4< 3.0- bc bc
> b
_j 2.0-T
: b
4: rLIfib

a a


Torula Yeast AMPu 20 AMPu 50 AMPu 100 AMPu 200 AMPu 400

Figure 2. Mean captures ( SE; n = 28 each trap bait) of adultA. ludens in McPhail
traps containing water, Torula yeast or five concentrations of AMPu (Experiment 2).
Within each sex, bars with the same letter are not significantly different from each
other by Fisher's protected LSD (P < 0.05).

December, 1995

Robacker: Attractant for Mexican Fruit Fly 577

than NuLure in 1-day orchard tests. They did not evaluate their attractant in tests
lasting longer than 1 day, nor did they attempt to develop a formulation that could be
used in a dry trap. However, their results were very important in showing that chem-
ically defined "food bait" attractants are at least competitive with standard proteina-
ceous lures.
Heath et al. (1995) developed an attractant for Mediterranean and Mexican fruit
flies for use with a newly designed dry trap. Their attractant contained ammonium ac-
etate and putrescine in a formulation that released ammonia and acetic acid at stable
rates for at least a month. Although putrescine emissions were not measured by
Heath et al. (1995), their data indicate that the lures remained attractive to both spe-
cies of flies for 6 weeks in the field in Guatemala. The lure was as attractive as Torula
yeast to feral Mediterranean fruit flies, but despite its similarity to AMPu, it was less
than 20% as attractive as Torula yeast to feral Mexican fruit flies. The explanation for
the low attractiveness of their lures to Mexican fruit flies compared with the AMPu
lures used here is unknown. It could be related to chemical differences between AMPu
and their lure, laboratory flies vs feral flies, race differences between the Mexican and
Guatemalan flies, or a combination of the three factors.
As in Wakabayashi & Cunningham (1991), all experiments conducted here with
AMPu were 1-day tests. It would have been desirable to conduct tests for 1 week, the
standard period that McPhail traps with Torula yeast are used. However, liquid for-
mulations used in this work were not sufficiently attractive to warrant longer tests
(Figs. 1 and 2) and the longevity of the agar preparations is unknown but likely less
than 1 week. The initial attractiveness of the AMPu/agar preparations on sticky traps
was sufficient to warrant longer tests with sticky traps if a longer-lasting AMPu for-
mulation that can be used with the traps can be developed. The goal will be to develop
a formulation that emits the three components at constant rates for 2 weeks or more.
This should be possible in light of the success of Heath et al. (1995). It certainly will
be necessary to develop such a formulation before AMPu lures will be practical for
trapping programs by regulatory agencies.

I thank Pravait Kaochoung (Office of Atomic Energy for Peace, Bangkhen,
Bangkok, Thailand), Maura Rodriguez and Marco Gomez for technical assistance;
Sammy Ingle for insects; and A. W. Guenthner (USDA-APHIS, Mission, Texas) for ir-
radiation of pupae. I thank Dr. Peter Landolt and Dr. Nancy Epsky for reviewing the
manuscript. Use of a product brand in this work does not constitute an endorsement
by the USDA.


BAKER, A. C., W. E. STONE, C. C. PLUMMER, AND M. MCPHAIL. 1944. A review of stud-
ies on the Mexican fruitfly and related Mexican species. U. S. Department of
Agriculture Misc. Publ. 531.
MEYER 1995. Development of a "dry" plastic trap with food-based synthetic at-
tractant for the Mediterranean and Mexican fruit flies (Diptera: Tephritidae).
J. Econ. Entomol. (In press).
ROBACKER, D. C. 1992. Effects of shape and size of colored traps on attractiveness to
irradiated, laboratory-strain Mexican fruit flies (Diptera: Tephritidae). Florida
Entomol. 75: 230-241.
Mating behavior and male mating success in wildAnastrepha ludens (Diptera:
Tephritidae) on a field-caged host tree. J. Insect Behav. 4: 471-487.

578 Florida Entomologist 78(4) December, 1995

ROBACKER, D. C., AND J. A. GARCIA. 1993. Effects of age, time of day, feeding history,
and gamma irradiation on attraction of Mexican fruit flies (Diptera: Tephriti-
dae) to bacterial odor in laboratory experiments. Environ. Entomol. 22: 1367-
ROBACKER, D. C., AND W. C. WARFIELD. 1993. Attraction of both sexes of Mexican fruit
fly,Anastrepha ludens, to a mixture of ammonia, methylamine, and putrescine.
J. Chem. Ecol. 19: 2999-3016.
WAKABAYASHI, N., AND R. T. CUNNINGHAM. 1991. Four-component synthetic food bait
for attracting both sexes of the melon fly (Diptera: Tephritidae). J. Econ. Ento-
mol. 84: 1672-1676.


578 Florida Entomologist 78(4) December, 1995


Department of Entomology
University of California
Riverside, CA 92521


Superparasitism and intrinsic larval competition by the solitary larval-pupal par-
asitoid Archytas marmoratus (Townsend) were studied in vivo. In superparasitized
hosts, when two parasitoids entered a host pupa, only one parasitoid completed devel-
opment. The surviving A. marmoratus maggot eliminated the conspecific competitor
through physiological suppression during the second stadium of the supernumerary
maggot. Supernumerary parasitoids never survived beyond the second instar, regard-
less of the time interval between initial parasitism and subsequent superparasitism.
Physical combat was not evident because the parasitoid eliminated did not show signs
of physical injuries. Scramble competition for host resources was not a probable mech-
anism of elimination because puparial weights and adult eclosion rates from super-
parasitized host pupae, and those from singly parasitized pupae, were not signif-
icantly different.

Key Words: Parasitoid competition, intraspecific interactions.


El parasistismo y la competencia larval intrinseca del parasitoide larvo-pupalAr-
chytas marmoratus (Townsend) fueron estudiados en vivo. En hospedantes superpa-
rasitados, cuando dos parasitoides entraron en una pupa hospedante, solamente uno
complete el desarrolo. La larva sobreviviente de A. marmoratus elimin6 el competitor
conespecifico mediante supresi6n fisiol6gica durante el segundo estadio de la larva su-
pernumeraria. Los parasitoides supernumerarios nunca sobrevivieron mas alla del
segundo instar, independientemente del intervalo entire el parasistismo inicial y el su-
perparasitismo subsecuente. El combat fisico no fue evidence debido a que el parasi-
toide eliminado no mostr6 sefiales de dafo fisico. La competencia por recursos de

Reitz: Larval Competition in A. marmoratus 579

hospedante no fue un mecanismo probable de eliminaci6n debido a que los pesos pu-
pales y las tasas de eclosi6n de adults de las pupas superparasitadas y de las para-
sitadas por una sola larva no fueron significativamente diferentes.

Archytas marmoratus (Townsend) (Diptera: Tachinidae) is a solitary larval-pupal
parasitoid of numerous species of Noctuidae (Lepidoptera). Included in its host range
are many important pest species in the genera Helicoverpa, Heliothis, Pseudaletia,
and Spodoptera (Arnaud 1978, Ravlin & Stehr 1984).
A. marmoratus has a complex life history that allows it to parasitize a wide range
of host instars. Females do not oviposit directly on hosts; instead they deposit numer-
ous eggs in the vicinity of potential host larvae. The eggs soon hatch into planidia-type
larvae (Wood 1987). Parasitism occurs when a host contacts a planidium which then
burrows between the host cuticle and epidermis where it resides (Bratti et al. 1993).
The first instar of A. marmoratus begins feeding on the host larva, but it does not molt
until after the host pupates. The first instar must reenter the host following each lar-
val-larval molt of the host. After the host undergoes its larval-pupal molt, the first-in-
star parasitoid penetrates the hemocoel under the host wing pad, where it induces the
formation of a respiratory tunnel. A. marmoratus development within the host pupa
is rapid. The maggot molts to the second instar 1 2 days following host pupation; the
second and third stadia last 2 4 days each, with pupariation occurring within the
host remains (Vickery 1929).
Because female A. marmoratus deposit multiple eggs at one time (Vickery 1929),
and more than one female may oviposit in the same location, considerable potential
for superparasitism exists. Despite this potential, only one A. marmoratus maggot
completes development in a host (Vickery 1929, Hughes 1975). Among the possible
mechanisms for the elimination of supernumerary parasitoids are physical combat,
scramble competition for host resources, or physiological suppression (Salt 1961,
Fisher 1971, Vinson & Iwantsch 1980). In this study, I examine aspects of superpara-
sitism and intrinsic larval competition, including parasitism rates, elimination of su-
pernumerary parasitoids, and effects on parasitoid development and emergence.


All tests were conducted at the Istituto di Entomologia "Guido Grande", Univer-
sita degli Studi di Bologna, Bologna, Italy. A. marmoratus adults were reared in plexi-
glass cages (40 x 40 x 40 cm) in an environmental chamber maintained at 27 2 C,
60 10% R.H. and 14:10 (L:D) photoperiod (fluorescent light). To obtainA. marmora-
tus planidia for parasitization, pieces of pleated filter paper were placed on the bottom
of cages as oviposition substrate the day prior to parasitism. Thus, all planidia were
less than 24 h old at the time of parasitization. Larvae of the factitious host, Galleria
mellonella L. (Lepidoptera: Galleriidae), were reared on artificial medium (Cam-
padelli 1973) in plastic containers (23 x 11 x 8 cm) held at 30 2 C, 60 10% R.H.,
and 0:24 (L:D) photoperiod.
Penultimate-instar larvae of G. mellonella in apolysis were isolated in containers
with fresh diet. The following morning, these groups were reexamined for newly-
molted ultimate-instars. Each G. mellonella larva was infested by gripping it behind
the head capsule with a soft forceps and transferringA. marmoratus planidia to the

580 Florida Entomologist 78(4) December, 1995

larval thorax with a fine camel hair brush. Each larva was held until the planidia bur-
rowed into the cuticle. Then it was placed in a new plastic container with fresh diet.
Six parasitism treatments were used, with three groups of G. mellonella larvae be-
ing superparasitized and three groups being singly parasitized. The three superpar-
asitized groups were: (A) Newly molted (day 1) ultimate-instar G. mellonella larvae
parasitized with two A. marmoratus planidia (Superparasitized Day 1), (B) Day 1
larvae parasitized once and superparasitized two days later (Superparasitized Day
1, 3), and (C) Day 1 larvae parasitized and superparasitized four days later (Super-
parasitized Day 1, 5). Only larvae with visible planidia were superparasitized. The
three corresponding singly parasitized control groups were parasitized on day 1 (Day
1 Control), day 3 (Day 3 Control), or day 5 (Day 5 Control) of the ultimate stadium, re-
G. mellonella were weighed upon pupation and individually isolated untilA. mar-
moratus pupariation, or G. mellonella eclosion or death. A. marmoratus puparia were
weighed one day after pupariation and held individually until eclosion. Host remains
were dissected to determine the number and status ofA. marmoratus maggots. To ver-
ify that all parasitoid remains had been recovered during the initial inspection, the
host remains were macerated in 10% KOH and reexamined. No additional A. marm-
oratus were recovered by this procedure. All parasitoid remains (bodies or exuviae)
were identified to larval instar (Ravlin & Stehr 1984), and bodies were carefully in-
spected for signs of physical injury. The size and degree of sclerotization of the ceph-
alopharyngeal skeletons of maggots from superparasitized hosts were compared at
the stage at which the first maggot died. TheA. marmoratus in each superparasitized
host were classified according the developmental stage to which they survived. If nei-
ther parasitoid survived longer than the other, the competitive interaction was con-
sidered a "tie". True winners of competitive interactions were those that actually
survived to adult eclosion.
Chi square tests were used to examine differences in parasitism rates and parasi-
toid survival among the different treatments. To determine if parasitoid size and de-
velopment were affected by parasitism treatment, data were analyzed by analysis of
covariance (ANCOVA) with host pupal weight serving as a covariate. Pairwise com-
parisons between superparasitized treatments and their corresponding controls were
made by least squares means t-tests.


Because first-instar parasitoids had to reenter the host successfully after it pu-
pated, not all G. mellonella pupae contained A. marmoratus (Table 1). However, par-
asitoid entries into host pupae were independent events because the proportion of
superparasitized hosts across superparasitism treatments (19.8%) was approxi-
mately equal to the square of the proportion of the parasitized pupae in all control
treatments (44.5%, X2 = 0.001, df = 1, P > 0.9). The first planidium that entered the
host pupa did not exclude the entry of the other planidium because superparasitism
rates were not significantly less than the values expected had entries been indepen-
dent events. Even though approximately 20% of host pupae were superparasitized
(Table 1), a maximum of one A. marmoratus survived per host.
Based on the condition of the parasitoid remains in host pupae, supernumerary
parasitoids were almost always eliminated during their second stadium (test for dif-
ferences amongA. marmoratus instars, X2 = 133, df = 2, P < 0.001, Table 2); this dif-
ference was consistent across superparasitism treatments (test for differences among
superparasitism treatments, X2 = 3.4, df = 2, P = 0.18, Table 2). Three of the A. mar-

Reitz: Larval Competition in A. marmoratus


Number ofA. marmoratus Maggots per
G. mellonella Pupa

Total Number of
Treatment' 1 Maggot 2 Maggots Pupae Tested

Day 1, 1 93 (36%) 47 (18%) 261

Day 1 Control 48 (41%) -118
Superparasitized 87 (44%)
Day 1, 3 40(20%) 196

Day 3 Control 56 (62%) -91
Superparasitized 47 (44%)
Day 1, 5 25(23%) 108

Day 5 Control 44 (36%) 123
Total 375 112 897

'See text for details of each treatment.

moratus maggots that died as first instars (17%, n = 18) had failed to penetrate the
host cuticle following host pupation. The remainder that did enter their host pupa
grew (when compared with newly hatched planidia), but did not become successfully
established in the host and molt. Only four (22%) of these first-instars showed signs
of physical injury, such as melanized wounds. Few of those maggots that died as sec-
ond instars had signs of physical injury (5%, n = 94), but they typically had smaller
and/or less sclerotized cephalopharyngeal skeletons than those that survived to a
later stage (75%, n = 94, X2 = 17.7, df = 2, P < 0.001). In two superparasitized pupae
(2%, n = 112), the supernumeraryA. marmoratus died before completing their final
larval ecdysis.
Adult eclosion of A. marmoratus did not differ across treatments (X2 = 10.1, df =
5, P > 0.07, Table 2), or when considering superparasitism versus single parasitism
treatments (X2 = 3.6, df = 1, P = 0.06, Table 2). Overall, development times ofA. mar-
moratus (from host pupation to A. marmoratus adult eclosion) from singly and super-
parasitized host pupae were not significantly different (overall x SE = 13.4 0.2
days, F = 0.1, df = 1, 128, P = 0.80). In pairwise comparisons between superparasitized
groups and their respective control groups, the only significant difference was be-
tween the Day 1, 3 superparasitized and Day 3 control groups; the controls emerged
one day earlier (12.7 0.3 days) than the parasitoids from the superparasitized group
(13.9 0.4 days) (t = 3.1, P = 0.0024).
The size ofA. marmoratus puparia increased significantly with host weight (pu-
paria from superparasitized hosts: y = 0.3 + 0.46x, r' = 0.82; puparia from singly par-
asitized hosts: y = 1.3 + 0.42x, r' = 0.73). However, puparia from singly parasitized
hosts were not significantly heavier (71.1 1.9 mg) than those from superparasitized
hosts (68.2 2.5 mg, F = 1.7, df = 1, 136, P = 0.19, test for homogeneity of intercepts).
Host weight alone accounted for over 74% of the variation inA. marmoratus weights.

Florida Entomologist 78(4)

December, 1995

a a

















1 ^


a- ~-
C) ) C C
- cf cf ~

to tl

00 '-I o0 0 Cl Cl
,q Cl Cl tO CO 0 Cl
0 N0 0 to


Q Cl t 0 Ll 0 CO
cT S CO 0 Q0 0 0

O N e 0 I 0 Cl

0 0 0 -I N
f- f- ^f

-S r j
+^ mcu +jcu o C
cc Ni cc +-i cc +-
rt ^ r ^ rt

i-i rt -i t ii r

Reitz: Larval Competition in A. marmoratus 583


A variety of mechanisms for the elimination of supernumerary larvae exist among
the Tachinidae. Physical combat has been observed among first instars ofMarquartia
chalconota Meigen (Mellini & Baronio 1971). Anoxia is responsible for elimination of
older supernumerary maggots of Lixophaga diatraeae (Townsend) (King et al. 1976).
Superparasitism results in reduced body size for several potentially gregarious ta-
chinids (Pschorn-Walcher 1971, Ziser et al. 1977, Reitz 1994). This variation in para-
sitoid size is attributed to resource depletion in superparasitized hosts. However,
these species have significantly different life histories fromA. marmoratus. Because
of the high potential of superparasitism occurring in the field, and the relationship be-
tween parasitoid and host size (A. marmoratus puparia are > 50% of the weight of
host pupae), A. marmoratus would be expected to have evolved an effective mecha-
nism for eliminating conspecific competitors.
Given the consistent stage at which supernumerary maggots of A. marmoratus
are eliminated and the lack of demonstrable physical injuries to these "losing" mag-
gots, physiological suppression of conspecific competitors cannot be excluded as a
mechanism of intraspecific competition. If competition were based solely on physical
attacks, all "losing" maggots should show signs of injury (Mellini & Baronio 1971,
Mellini 1990). Also, some encounters, especially in hosts superparasitized on the same
day, could be expected to be resolved when both parasitoids were third instars.
In fact, if direct physical combat was responsible, it should occur most often among
third instars. Unlike many solitary hymenopteran parasitoids that have free-roaming
larvae adapted for fighting (Vinson 1985, Kfir & van Hamburg 1988, McBrien &
Mackauer 1990), first and second instars ofA. marmoratus reside in respiratory tun-
nels (Mellini 1990) that form along the wing pad margins of host pupae. Only third in-
stars become mobile (Hughes 1975, Bratti et al. 1992). However, no host pupae
contained two third instars. While second and third instars ofA. marmoratus possess
sickle-shaped mandibles that could inflict serious damage, any observed damage
could have occurred after a competitor had already died from other causes. In addi-
tion, second instars are considerably smaller than G. mellonella pupae; therefore, un-
less parasitoid respiratory attachments are in close proximity, second instars would
not encounter one another, further increasing the occurrence of encounters among
third instars.
The possibility that supernumerary maggots are eliminated through scramble
competition for host resources is also not supported by the present data. If scramble
competition was operating, greater variation in the stage at which competitors are
eliminated would be expected. In particular, larger hosts should more frequently sup-
port multiple third instars. Additionally, if scramble competition was responsible for
elimination of supernumerary maggots, the size ofA. marmoratus should vary with
the number of maggots present in a host pupa. However, the relationship betweenA.
marmoratus weights and host pupal weights did not vary with respect to whether a
host was singly or superparasitized. Also, adult eclosion rates did not differ signifi-
cantly as might be expected as a result of scramble competition.
A possible scenario for physiological suppression of supernumerary A. marmora-
tus is that older maggots (i.e., those parasitizing the host larva first) are more devel-
opmentally advanced, and initially have faster development rates in the host pupa,
thus molting to the second and third instar sooner than subsequent maggots (Bratti
et al. 1992, 1993). Because the host pupa dies by the time anA. marmoratus maggot
molts to the third instar (Allen 1926), the maggot that molts to its final instar first
could make the host environment unsuitable for younger maggots to continue their
development. Therefore, maggots reaching their final instar first could suppress com-

Florida Entomologist 78(4)

petitors by production of proteolytic enzymes (Mellini 1990) or degradation of the
host. Further in vitro studies of intrinsic competition should elucidate the specific
means of physiological suppression used byA. marmoratus.


A. Bratti, M. Mariani and the Staff of the Istituto di Entomologia "Guido Grande",
University degli Studi di Bologna provided valuable technical assistance. I appreciate
the assistance of P. H. Adler, L. Correlli-Grappadelli, K. A. Luhring, and J. T. Trumble
in making this study possible, and the comments of two anonymous reviewers which
have improved this manuscript. This study was made possible with financial support
from a W. C. Nettles Memorial Grant, a Sigma Xi Grant-in-Aid of Research, and a
Florida Entomological Society Scholarship.


ALLEN, W. H. 1926. Life history of the variegated cutworm tachini fly, Archytas ana-
lis. J. Agric. Res. 32: 417-435.
ARNAUD, P. H., JR. 1978. A host-parasite catalog of North American Tachinidae
(Diptera). United States Dept. Agric. SEA Misc. Publ. No. 1319, 860 pp.
BRATTI, A., G. GARDENGHI, AND G. MIGLIOLI. 1993. Behavior and growth rate of Ar-
chytas marmoratus (Town.) (Diptera: Tachinidae) planidia in larvae of Galleria
mellonella L. (Lepidoptera Galleriidae). Boll. Ist. Entomol. Univ. Bologna 47:
BRATTI, A., W. C. NETTLES, JR., AND P. FANTI. 1992. Influence ofHelicoverpa zea (Lep-
idoptera: Noctuidae) age during the last instar on rates of parasitization by the
larval-pupal parasitoid, Archytas marmoratus (Diptera: Tachinidae). Environ.
Entomol. 21: 1196-1201.
CAMPADELLI, G. 1973. Allevemento di Galleria mellonella L. con dieta semiartificiale.
Boll. Ist. Entomol. Univ. Bologna 32: 2-25.
FISHER, R. C. 1971. Aspects of the physiology of endoparasitic Hymenoptera. Biol.
Rev. Cambridge Philos. Soc. 46: 243-278.
HUGHES, P. S. 1975. The biology ofArchytas marmoratus (Townsend). Ann. Entomol.
Soc. America 68: 759-767.
KFIR, R., AND H. VAN HAMBURG. 1988. Interspecific competition between Telenomus
ullyetti (Hymenoptera: Scelionidae) and Trichogrammatoidea lutea (Hy-
menoptera: Trichogrammatidae) parasitizing eggs of the cotton bollworm He-
liothis armiger in the laboratory. Environ. Entomol. 17: 664-670.
KING, E. G., L. R. MILES, AND D. F. MARTIN. 1976. Some effects of superparasitism by
Lixophaga diatraeae of sugarcane borer larvae in the laboratory. Entomol. Exp.
Appl. 20: 261-269.
MCBRIEN, H., AND M. MACKAUER 1990. Heterospecific larval competition and host
discrimination in two species of aphid parasitoids:Aphidius ervi and Aphidius
smith. Entomol. Exp. Appl. 56: 145-153.
MELLINI, E. 1990. Sinossi di biologia dei Ditteri Larvevoridae. Boll. Ist. Entomol.
Univ. Bologna. 45: 1-38.
MELLINI, E., AND P. BARONIO. 1971. Superparasitismo sperimentale e competizioni
larvali del parassitoide solitario Marquartia chalconota Meig. Boll. Ist. Ento-
mol. Univ. Bologna 30: 133-152.
PSCHORN-WALCHER, H. 1971. Experiments on inter-specific competition between
three species of tachinids attacking the sugar cane moth borer, Diatraea sac-
charalis (F). Entomophaga 16: 125-131.
RAVLIN, F. W., AND F. W. STEHR. 1984. Revision of the genus Archytas (Diptera: Ta-
chinidae) for America north of Mexico. Entomol. Soc. America Misc. Pub. 58, 58

December, 1995

Scheffrahn & Roisin: Antillean Parvitermes 585

REITZ, S. R. 1994. Reproductive biology of Eucelatoria bryani and Eucelatoria ruben-
tis (Diptera: Tachinidae), larval parasitoids of Helicoverpa zea (Lepidoptera:
Noctuidae). Ph. D. Thesis, Clemson University.
SALT, G. 1961. Competition among insect parasitoids. Symp. Soc. Exp. Biol. 15: 96-
VICKERY, R. A. 1929. Studies on the fall army worm in the Gulf coast district of Texas.
United States Dept. Agriculture Tech. Bull. 138, 64 pp.
VINSON, S. B. 1985. The behavior of parasitoids, pp. 417-469 in G. A. Kerkut and L. I.
Gilbert [eds.], Comprehensive insect physiology, biochemistry and pharmacol-
ogy, vol. 9. New York, Pergamon Press.
VINSON, S. B., AND G. F. IWANTSCH. 1980. Host suitability for insect parasitoids.
Annu. Rev. Entomol. 25: 397-419.
WOOD, D. M. 1987. Tachinidae, pp. 1193-1269 in J. F. McAlpine [ed.], Manual of Ne-
arctic Diptera. Ottawa, Agriculture Canada.
ZISER, S. W., J. A. WOJTOWICZ, AND W. C. NETTLES, JR. 1977. The effects of the num-
ber of maggots per host on length of development, puparial weight, and adult
emergence of Eucelatoria sp. Ann. Entomol. Soc. America 70: 733-736.


Scheffrahn & Roisin: Antillean Parvitermes


'Fort Lauderdale Research and Education Center
University of Florida, Institute of Food and Agricultural Sciences
3205 College Avenue, Ft. Lauderdale, FL 33314

'Research Associate, National Fund for Scientific
Research, Belgium; Universit6 Libre de Bruxelles, CP 160/12
Laboratoire de Biologie Animale et Cellulaire Avenue F.D. Roosevelt 50, B-1050
Brussels, Belgium

Three Parvitermes species are described from recent Antillean collections. All ex-
hibit a dimorphic soldier caste characterized by rare major soldiers, reported from
Parvitermes for the first time, and minor soldiers with constricted heads. Workers
share similarities in digestive tube morphology and enteric valve armature. Soldiers
and workers of a new species, Parvitermes collinsae, are described and the descrip-
tions of Parvitermes wolcotti (Snyder, 1924) from Puerto Rico and P. pallidiceps
(Banks, 1919) from Hispaniola are revised.
Key Words: Taxonomy, new species, castes, West Indies.

Tres species de Parvitermes son descritas de colecciones antillanas recientes. To-
das exhiben una casta dim6rfica de soldados caracterizada por soldados mayores, re-

Florida Entomologist 78(4)

portados de Parvitermes por primera vez, y soldados menores con cabezas
constringidas. Las obreras tienen similitudes en la morfologia del tubo digestive y en
la armadura de la valvuala ent6rica. Son descritos los soldados y obreras de la nueva
especie, Parvitermes collinsae, y son revisadas las descripciones de Parvitermes wol-
cotti (Snyder, 1924) de Puero Rico y P. pallidiceps (Banks, 1919) de La Espaiola.

The genus Parvitermes was established by Emerson (in Snyder 1949) for an assem-
blage of small nasute termite species from the Neotropical Region. The type species,
P. brooks (Snyder), and five of the other seven species currently placed inParvitermes
are known only from the West Indies (Araujo 1977, Scheffrahn & Krecek 1993). A re-
cent morphological study of the worker digestive tube, mandibles, and enteric valve
armature indicates that Parvitermes is endemic to the Greater Antilles and that this
and other Antillean nasute genera, as well as the mainland species of Parvitermes, re-
quire a reevaluation of their generic status (Roisin et al. unpubl.).
An ongoing survey of the termites of the West Indies has uncovered a number of
new species, species records (Scheffrahn et al. 1994), and undescribed castes. In-
cluded among these findings are a new Parvitermes from the Dominican Republic and
the discovery of a major soldier caste in P. pallidiceps (Banks) and P. wolcotti (Snyder).
In addition to a major soldier, minor soldiers of all three species exhibit a well-marked
head constriction. We include herein the description of the soldier and worker castes
of Parvitermes collinsae n. sp. and a redescription of P. pallidiceps from Hispaniola
and P. wolcotti from Puerto Rico to supplement the earlier descriptions by Banks
(1919) and Snyder (1924), respectively.


Measurements of specimens preserved in 85:15 ethanol:water were made with a
calibrated ocular micrometer and follow those defined by Sands (1965) and Roonwal
(1970). Terms used to describe soldier morphology and color follow those of Sands
(1965). Left mandible index of workers equals the distance between the apical and
first marginal tooth divided by distance between first and third marginals (Emerson
1960). In Parvitermes, the second marginal tooth of the left worker mandible is re-
duced to a long, rather straight, cutting edge. Head height was measured without
postmentum to the highest point of the head capsule. Scanning electron micrographs
of soldiers dehydrated by the method of Nation (1983) were made with a Hitachi S-
4000 field emission microscope at 10kV.
The digestive tube of large workers was observed after removal of the abdominal
wall and fat tissues under a dissecting microscope. The digestive tube ofP. pallidiceps
was drawn with the aid of a camera lucida at 50X and described using the terminology
of Noirot & Noirot-Timothee (1969) and Kovoor (1969). Enteric valves were longitudi-
nally split with a fine scalpel. Mandibles and split enteric valves were dehydrated and
mounted in toto for microphotography.
The holotype minor soldier and morphotype major soldier and worker of P. collin-
sae will be deposited in the collection of the National Museum of Natural History,
Washington, D.C. Paratype specimens will be deposited in the Florida State Collec-
tion of Arthropods, Fla. Dept. Agric. Cons. Serv., Division of Plant Industries, Gaines-
ville, Florida, at the institutions of the first and second authors (in Ft. Lauderdale and
Brussels, respectively), and at the Universidad de Santo Domingo, Dominican Repub-

December, 1995

Scheffrahn & Roisin: Antillean Parvitermes

Figures 1-4. Scanning electron micrographs of dorsal and lateral views of Parvit-
ermes collinsae n. sp. minor (1, 2) and major (3, 4) soldiers.

Parvitermes collinsae Scheffrahn and Roisin, new species

IMAGO. Unknown.
MINOR SOLDIER (Figs. 1-2, Table 1). Head capsule, postmentum gulaa), and first
antennal article orange, nasus chestnut brown; antennal articles three and beyond se-


Measurement in mm (n = 46) Range Mean SD Holotype

Head length with nasus 0.95 1.16 1.06 0.056 1.06
Head length without nasus 0.58 0.71 0.66 0.036 0.66
Head width, maximum 0.45 0.64 0.55 0.056 0.55
Head width, anterior lobe 0.43 0.54 0.49 0.029 0.49
Nasus width at base 0.11 0.14 0.13 + 0.0094 0.13
Nasus width at middle 0.059 0.081 0.069 0.0056 0.063
Head height, maximum 0.33 0.44 0.38 0.035 0.38
Pronotum width 0.33 0.39 0.36 0.014 0.35
Pronotum length, maximum 0.14 + 0.19 0.17 + 0.012 0.18
Hind tibia length 0.59 0.83 0.73 0.059 0.74
Total length 2.06 3.10 2.53 0.25 2.93

Florida Entomologist 78(4)


Measurement in mm (n = 4) Range Mean SD Morphotype

Head length with nasus 1.09 1.13 1.12 + 0.020 1.13
Head length without nasus 0.69 0.74 0.72 0.024 0.74
Head width, maximum 0.60 0.62 0.61 + 0.0088 0.62
Head width, anterior lobe 0.58 0.60 0.59 0.0094 0.60
Nasus width at base 0.18 0.19 0.18 + 0.0063 0.18
Nasus width at middle 0.088 0 0.088
Head height, maximum 0.43 0.46 0.44 0.020 0.43
Pronotum width 0.41 0.43 0.42 0.0088 0.43
Pronotum length, maximum 0.18 0.19 0.18 + 0.0060 0.19
Hind tibia length 0.84 0.90 0.87 0.026 0.88
Total length 2.60 3.16 2.84 + 0.26 3.16

pia brown; thoracic nota and abdominal tergites pale brown; and abdominal sternites,
coxae, and femora very pale brown.
Head capsule in dorsal view slightly constricted near middle, maximum width of
anterior lobe usually less than that of posterior lobe; in lateral view, anterior and pos-
terior lobes raised equally and rounded in front of and behind constriction. Head cap-
sule covered with dozens of short and medium length setae tilted at various angles.
Antennae with 13 articles; second less than or as long as third, third usually
longer than fourth, fourth shorter or as long as fifth. Mandibles usually with distinct
points. Nasus very weakly conical, projecting almost straight forward, i.e., parallel to
dorsal plane of head.
MAJOR SOLDIER (Figs. 3-4, Table 2). Head capsule, postmentum, and first anten-
nal article ferruginous orange, nasus chestnut brown; distal antennal articles sepia
brown; thoracic nota and abdominal tergites pale brown; and abdominal sternites,
coxae, and femora very pale brown.
Head capsule in dorsal view roundly constricted near middle, maximum width of
anterior lobe greater than that of posterior lobe; in lateral view, anterior and posterior
lobes raised equally and rounded in front of and behind shallow constriction. Head
capsule covered with dozens of randomly tilted medium and long setae.


Measurement in mm (n = 36) Range Mean SD Morphotype

Head width, maximum 0.85 -1.11 0.96 0.070 0.96
Head length to postclypeus
anteclypeus suture 0.78 -1.03 0.90 + 0.059 0.90
Postclypeus width 0.40 0.54 0.47 0.032 0.48
Postclypeus length 0.15 0.24 0.20 + 0.019 0.20
Hind tibia length 0.74 0.94 0.82 0.056 0.83
Total length 2.85 -4.15 3.51 + 0.33 3.95

December, 1995

Scheffrahn & Roisin: Antillean Parvitermes

Figures 5-8. Scanning electron micrographs of dorsal and lateral views of Parvit-
ermes pallidiceps (Banks) minor (5, 6) and major (7, 8) soldiers.

Antennae with 14 articles; second shorter than third, third longer than fourth, and
fourth equal to fifth. Mandibles with points. Nasus stout, slightly conical; tilting
about 15 above dorsal plane of head.
WORKER (Figs. 13,17; Table 3). Head capsule pale brown except for unpigmented
epicranial suture and genae; nota and tergites very pale brown; antennal articles
brown; no pigmentation on remainder of body. Dozens of variable setae covering head.
Antennae with 14 or 15 articles; relative lengths somewhat variable but usually if
14, second longer than third, third equal or longer than fourth, and fourth shorter
than fifth; if 15, second equal to or longer than third, third shorter than fourth, and
fourth as long as fifth.
Mandibles with posterior edge of apical tooth slightly shorter or equal in length to
anterior edge of first marginal tooth, left mandible index 0.24-0.35; right molar plate
abruptly notched near base, 5-6 ridges total, 4 well-developed ridges distal to notch
(Fig. 13).
Digestive tube consisting of moderate-sized crop, well-armed gizzard, and me-
senteron followed by very long mixed segment (as in P. pallidiceps, Fig. 15). Mesen-
teric part of mixed segment on internal/dorsal side, attached to remainder of
mesenteron by thin peduncle. Malpighian tubules inserted on external/ventral side of
mesentero-proctodeal junction. P1 extremely elongated, forming a long loop on ven-
tral side of paunch. Enteric valve latero-dorsal, comprising diffuse areas with many
small spines, followed by a ring of six spiny swellings, three major ones bearing 12-20
long, curved spines (Fig. 17), alternating with minor ones with small spines. P3 to rec-
tum without remarkable features.

Florida Entomologist 78(4)

Figures 9-12. Scanning electron micrographs of dorsal and lateral views of Parvi-
termes wolcotti (Snyder) minor (9, 10) and major (11, 12) soldiers.

Comparisons. Although allopatric in distribution, minor soldiers and workers of P.
collinsae are closest to P. wolcotti. Minor soldiers of P. wolcotti can be distinguished
by their much darker, more brown pigmentation, and proportionally thicker nasus.
Minor soldiers ofP. collinsae from Pedernales Province are proportionally larger than
those from Barahona and Azua Provinces.
Major soldiers can be distinguished by the shape of their head capsule which lacks
constriction in P. wolcotti, while that of P. collinsae is constricted and has an anterior
lobe wider than the posterior lobe.
Workers of P. wolcotti have no exoskeletal pigmentation and have antennae with
only 14 articles, whereas P. collinsae worker bodies are dorsally pigmented and have
a nearly equal proportion of individuals with 14 and 15 antennal articles. The molar
plate of the right mandible of P. wolcotti workers has 5 well-developed ridges distal to
the basal notch, whereas P. collinsae has four such ridges (Figs. 13-14).
Soldiers and workers of P. collinsae can be readily separated from those of the par-
tially sympatric P. pallidiceps on the basis of simple characters. Minor soldiers of the
former have 13 antennal articles and distinct mandibular points while the latter have
12 articles and points are either lacking or vestigial. Likewise, major soldiers of P. col-
linsae have 14 antennal articles and points while P. pallidiceps major soldiers have
14-15 articles but no points. Large workers of P. collinsae have 14-15 antennal articles
and head width ranging between 0.85-1.11 mm while large workers of P. pallidiceps
have only 15 articles and greater head widths of 1.05-1.36 mm.
Material Examined. Specimens from 25 foraging groups of P. collinsae n. sp. were
measured from eleven localities in the Dominican Republic. HOLOTYPE minor sol-
dier and MORPHOTYPE worker: Barahona Province, 2 km W Fondo Negro (18'25'N,

December, 1995

Scheffrahn & Roisin: Antillean Parvitermes 591

Figures 13-14. Scanning electron micrographs of P. collinsae (13) and P. wolcotti
(14) right molar plates of worker mandibles.

Florida Entomologist 78(4)


Measurement in mm (n = 48) Range Mean SD

Head length with nasus 1.03 -1.26 1.14 0.051
Head length without nasus 0.65 0.80 0.71 0.030
Head width, maximum 0.51 0.67 0.57 0.033
Head width, anterior lobe 0.44 0.55 0.47 0.023
Nasus width at base 0.094 0.13 0.11 0.0090
Nasus width at middle 0.059 0.075 0.065 0.0040
Head height, maximum 0.36 0.51 0.42 0.033
Pronotum width 0.33 0.40 0.35 0.016
Pronotum length, maximum 0.15 0.20 0.17 0.010
Hind tibia length 0.74 0.95 0.83 0.047
Total length 2.23 3.26 2.82 0.21

71 05'W), 30-III-1993, J. Chase and J. de la Rosa Guzman, nest series reference no.
DR801 and DR802. MORPHOTYPE major soldier: Azua Province, Barahona/San
Juan highway intersection (18 28'N, 70 52'W), 3-V-1992, J. Chase, J. de la Rosa Guz-
man, and R. Scheffrahn, nest series reference no. DR350. PARATYPES: Azua Prov-
ince, 9 km E Quita Coraza (18 28'N, 71 58'W), 3-V-1992, Caracoles (18 25'N,
70 37'W), 27-III-1993; Baoruco Province, Vijia de Mena (18 23'N, 71 11'W), 21-VI-
1991: Barahona Province, La Canoa 1 km W Vincente Noble (18 23'N, 7111'W), 27-
II-1992, 2 km N Canoa (18 22'N, 71 09'W) 3-V-1992; Pedernales Province, 12 km W
Oviedo (1752'N, 71 29'W), 29-III-1993, 15 km E Cabo Rojo (1756'N, 71 32'W), 29-III-
1993, 3 km E Pedernales (18 00'N, 71 41'W), 29-III-1993, Pedernales (18 03'N,
71 44'W), 28-III-1993, all collected variously by J. Chase, J. de la Rosa Guzman, J.
Mangold, and R. Scheffrahn.
Etymology. This species is named in honor of Dr. Margaret S. Collins, isopterist at
the Smithsonian Institution and retired Professor of Zoology, Howard University, for
her lifelong dedication to the study of termites.

Parvitermes pallidiceps (Banks, 1919)

IMAGO. Unknown.
MINOR SOLDIER (Figs. 5-6, Table 4). Head capsule and first several antennal arti-
cles orange-yellow, nasus darker, grading to chestnut brown in distal region and be-
coming lighter at tip to give banded appearance, postmentum lighter than head
capsule; distal antennal articles pale brown; thoracic nota and abdominal tergites
very pale brown; abdominal sternites, coxae, femurs and tibiae paler than tergites.
Head capsule in dorsal view slightly constricted near middle, maximum width of
anterior lobe always less than that of posterior lobe; in lateral view, posterior lobe
raised and in line with tip of nasus, anterior lobe raised slightly between constriction
and base of nasus. Head capsule scattered with medium and long setae.
Antennae with 12 articles; the second shorter than the third, and the third equal
to the fourth. Mandibles without points or only one point vestigial. Nasus slender, cy-
lindrical except near base, tilting about 20 above plane of head.

December, 1995

Scheffrahn & Roisin: Antillean Parvitermes 593


Measurement in mm (n = 11) Range Mean + SD

Head length with nasus 1.29 -1.41 1.35 0.035
Head length without nasus 0.88 0.96 0.91 + 0.026
Head width, maximum 0.76 0.88 0.81 + 0.040
Head width, anterior lobe 0.70 0.79 0.73 0.028
Nasus width at base 0.15 0.24 0.19 0.024
Nasus width at middle 0.081 0.12 0.10 + 0.011
Head height, maximum 0.50 0.63 0.55 0.036
Pronotum width 0.49 0.51 0.50 + 0.0098
Pronotum length, maximum 0.24 0.28 0.26 0.012
Hind tibia length 1.18 1.24 1.21 + 0.019
Total length 2.73 3.63 3.15 0.23

MAJOR SOLDIER (Figs. 7-8, Table 5). Head capsule and first several antennal ar-
ticles ferruginous orange, nasus darker, grading to dark chestnut brown in distal re-
gion and becoming lighter at tip to give banded appearance, postmentum lighter than
head capsule; distal antennal articles brown; thoracic nota and abdominal tergites or-
ange-yellow; abdominal sternites, coxae, femurs and tibiae paler than tergites.



Figure 15. Dorsal (D), right (R), ventral (V), and left (L) configurations of the di-
gestive tube in situ of P. pallidiceps worker. CP, crop; M, mesenteron (stippled) includ-
ing mesenteric part of MS, mixed segment; 0, oesophagus; P1, first proctodeal
segment; P2, enteric valve; P3, paunch; P4, colon; R, rectum; T, Malpighian tubules
(darkly stippled). Scale bar = 0.5 mm.

Florida Entomologist 78(4)


Measurement in mm (n = 31) Range Mean SD

Head width, maximum 1.05 1.36 1.21 + 0.064
Head length to postclypeus anteclypeus
suture 0.96 1.26 1.13 0.079
Postclypeus width 0.48 0.60 0.55 0.029
Postclypeus length 0.19 0.25 0.22 0.017
Hind tibia length 0.95 -1.26 1.14 0.063
Total length 3.25 4.85 4.03 0.45

Head capsule in dorsal view roundly constricted near middle, maximum width of
anterior lobe always less than that of posterior lobe; in lateral view, posterior lobe
raised and in line with base of nasus, anterior lobe raised beyond constriction and con-
tinues to rise with upward tilt of nasus. Head capsule scattered with dozens of me-
dium and long, randomly tilted setae.
Antennae with 14-15 articles; the second equal or longer than the third, the third
shorter than the fourth, and the forth equal to the fifth. Mandibles without points. Na-
sus conical, especially near base, tilting about 25 above plane of head.
WORKER (Figs. 15-16, Table 6). Head capsule pale brown except for unpigmented
epicranial suture and genae; nota and tergites very pale brown; antennal articles
light brown; no pigmentation on remainder of body. Dozens of variable setae covering
Antennae with 15 articles; second longer than third, third shorter than fourth, and
fourth equal to fifth.
Mandibles similar to P. collinsae with the exception that molar plate of right man-
dible has 6-7 ridges total, including 5 well-developed ridges distal to basal notch. Di-
gestive tube (Fig. 15) and enteric valve armature also similar to P. collinsae.


Measurement in mm (n = 33) Range Mean SD

Head length with nasus 0.93 1.16 1.02 0.052
Head length without nasus 0.58 0.73 0.64 0.035
Head width, maximum 0.51 0.65 0.57 0.036
Head width, anterior lobe 0.45 0.60 0.50 + 0.030
Nasus width at base 0.11 0.15 0.13 0.010
Nasus width at middle 0.069 0.094 0.079 0.0061
Head height, maximum 0.34 0.43 0.37 0.022
Pronotum width 0.33 0.40 0.36 0.018
Pronotum length, maximum 0.14 0.18 0.16 0.010
Hind tibia length 0.64 0.81 0.71 0.038
Total length 2.00 3.30 2.61 0.30

December, 1995

Scheffrahn & Roisin: Antillean Parvitermes 595

Material Examined. Specimens compared favorably with paratype minor soldiers
from type colony, Diquini, Haiti, XI-1912, W.M. Mann. Specimens from 48 foraging
groups of P. pallidiceps were measured from 12 localities in the Dominican Republic:
Peravia Province, 2 km E Las Calderas (18 13'N, 70 29'W), 3-V-1992, J. Chase, J. de
la Rosa Guzman, and R. Scheffrahn; Azua Province, 12 km N Yayas de Viajama
(18 39'N, 70 56'W), 3-V-1992, J. Chase, J. de la Rosa Guzman, and R. Scheffrahn;
Azua Province, 9 km E Quita Coraza (18 28'N, 70 58'W), 3-V-1992, J. Chase, J. de la
Rosa Guzman, and R. Scheffrahn; Azua Province, Barahona/San Juan highway inter-
section (18 28'N, 70 52'W), 3-V-1992; Barahona Province, La Cienega (18 02'N,
71 08'W), 4-VIII-1992, 2 km W Fondo Negro (18 25'N, 71 05'W), 4-VIII-1992, 30-III-
1993; Distrito Nacional, Santo Domingo (18 29'N, 69 54'W), 4-V-1992; La Romana
Province, La Romana (18 27'N, 68 58'W), 12-VI-1992; La Vega Province, El Piflo de La
Vega (19 09'N, 70 29'W), 8-VI-1992; Peravia Province, 8 km W Bani (18 19'N,
70 24'W), 26-II-1992; Samana Province, 15 km S Las Galeras (19 12'W, 69 13'N), 6-
VI-1992; San Pedro de Macoris Province, Juan Dolio (18 25'N, 69 25'W), 4-VIII-1992,
all collected variously by J. Chase, J. de la Rosa Guzman, J. Mangold, and R. Schef-

Parvitermes wolcotti (Snyder, 1924)

IMAGO. Not described; suspected collections from light traps.
MINOR SOLDIER (Figs. 9-10, Table 7). Head capsule brown, nasus darker, grading
to very dark sepia brown in distal region, slightly lighter at tip; antennal articles
slightly lighter than head capsule; margins of thoracic nota and abdominal tergites
pale brown; and abdominal sternites, legs, and interior areas of nota very pale yellow-
Head capsule in dorsal view slightly constricted near middle, maximum width of
anterior lobe usually less than that of posterior lobe; in lateral view, posterior and an-
terior lobes equally raised and rounded in front of and behind constriction. Head cap-
sule covered with dozens of short and medium length setae tilted at various angles.
Antennae with 13 articles; second as long as third, third longer than fourth, fourth
shorter than fifth. Mandibles with points. Nasus very weakly conical, projecting al-
most straight forward.
MAJOR SOLDIER (Figs. 11-12, Table 8). Head capsule brown, nasus darker, grad-
ing to very dark sepia brown in distal region, slightly lighter at tip; antennal articles
slightly lighter than head capsule; thoracic nota and abdominal tergites pale brown;
and abdominal sternites, legs, and interior areas of nota very pale yellow-brown.
Head capsule in dorsal view ellipsoidal, not constricted near middle; in lateral
view, head indented near middle raised and rounded in front of and behind indenta-
tion. Head capsule covered with dozens of short and medium length setae randomly
Antennae with 14 articles; second as long as third, third shorter than fourth,
fourth shorter than fifth. Mandibles with points. Nasus conical, projecting almost
straight forward.
WORKER (Fig. 14, Table 9). Head capsule yellow brown except for unpigmented ep-
icranial suture and genae; nota and tergites light yellow or unpigmented; antenna
articles pale yellow brown; no pigmentation on remainder of body. Dozens of variable
setae covering head.
Antennae with 14 articles; second longer than third, third equal to or longer than
fourth, fourth shorter than fifth..

Florida Entomologist 78(4)


Measurement in mm (n = 1)

Head length with nasus 1.11
Head length without nasus 0.76
Head width, maximum 0.65
Head width, anterior lobe
Nasus width at base 0.15
Nasus width at middle 0.088
Head height, maximum 0.38
Pronotum width 0.41
Pronotum length, maximum 0.21
Hind tibia length 0.80
Total length 2.66

'Dehydrated and sputter-coated with gold for SEM photography.

Mandibles with posterior edge of apical tooth slightly shorter than anterior edge
of first marginal tooth, left mandible index 0.45; right molar plate abruptly indented
(notched) near base, 6-7 ridges total, 5 well-developed ridges distal to indentation
(Fig. 14). Digestive tube also similar to P. collinsae.
Material Examined. Available type material of P. wolcotti was badly damaged and
unusable, however, Snyder's (1924) description of the minor soldier was sufficient for
identification. Specimens from 18 foraging groups of P. wolcotti were measured from
6 localities in Puerto Rico: Guanica State Forest Reserve. (17'58'N, 66'52'W), 13-I-
1993, S. Jones; NE Salinas, Rancho Guama Rd @ 706/713 (18'01'N, 66'13'W), 15-V-
1992, S. Jones; Bafios de Coamo (1802'N, 66'22'W), 30-V-1993, J. Chase, J. de la Rosa
Guzman, J. Mangold, and R. Scheffrahn; Bafios de Coamo, 29-V-1993; highway 1 & 52
intersection, 9 km NE Salinas (18'01'N, 66'15'W), 29-V-1993; 5 km W Coamo on high-
way 14 (18'03'N, 66'29'W), 30-V-1993; Yauco (1802'N, 66'49'W), 30-V-1993, Guanica
Reserve, 30-V-1993, J. Chase, J. de la Rosa Guzman, J. Mangold, and R. Scheffrahn;
Guanica Reserve, 15-1-1993, 6-V-1992, 24-25-V-1993, S. Jones.


Measurement in mm (n = 28) Range Mean SD

Head width, maximum 0.88 1.08 0.95 0.042
Head length to postclypeus anteclypeus
suture 0.80 0.98 0.88 0.045
Postclypeus width 0.41 0.50 0.45 0.020
Postclypeus length 0.19 0.24 0.21 0.012
Hind tibia length 0.71 0.84 0.77 0.030
Total length 2.95 4.00 3.59 0.29

December, 1995

Scheffrahn & Roisin: Antillean Parvitermes 597


i h

F ~LI! :~Y~~P~

Florida Entomologist 78(4)

?C, i

sr f


Figure 17. Longitudinal section of enteric valve of P. collinsae worker, showing two
of three major spine-bearing swellings.

Originally described from minor soldiers collected in Haiti (Banks 1919), Parviter-
mes pallidiceps is widely distributed throughout the Dominican Republic, whereas P.

December, 1995

-i ,

Scheffrahn & Roisin: Antillean Parvitermes 599

collinsae is restricted to the more arid southwestern Dominican Provinces. Arid local-
ities in southeastern Haiti will likely also yield P. collinsae. In the Dominican Repub-
lic, both species occur sympatrically in the southwestern Province of Azua and
adjacent areas of Barahona Province. Parvitermes wolcotti is known from the south-
western and southern dry forest lowlands of Puerto Rico (Jones et al. 1995) and the
Virgin Islands (Scheffrahn et al. 1994 and unpublished).
Parvitermes pallidiceps and P. collinsae foragers have been collected in and under
dried ruminant dung on soil, and in or on various cellulosic materials. Parvitermes
pallidiceps is reported to attack sugarcane seed in Haiti (Araujo 1970). Although the
nest center with royal pair has not been observed for either species, early instars,
workers, soldiers, and brachypterous nymphs of both species have been collected from
similar nesting structures. Their subterranean excavations consist of a diffuse system
of chambers interconnected by narrow tunnels and are usually found under large
rocks, stones, or heavy surface debris. Nest chambers of P. pallidiceps often contain
bits of dried grass which suggests they forage in the open at night as this species has
not been collected under protective sheeting. In contrast, P. collinsae and P. wolcotti
will cover dried twigs and grass stems with thin soil sheeting when foraging above
ground. Parvitermes wolcotti has been collected from beneath dung pats, soil sheeting,
and foraging tubes covering wood, or from buried cardboard traps. Snyder's (1924)
types for P. wolcotti were collected from beneath soil sheeting covering rotten wood.
Although minor soldiers are common, major soldiers are extremely rare for all
three species, constituting much less than 1% of total foraging populations thus far
collected. Unexpectedly, major soldiers have been collected more often with small for-
aging groups containing fewer than 100 individuals than with large groups consisting
of more than 1,000 members. Their role in colony defense remains obscure.
Brachypterous nymphs have been collected in foraging groups in May and August
for P. pallidiceps and in October for P. collinsae suggesting dispersal flights in late
summer or fall. Although not collected with foragers, alates believed to be P. wolcotti
were collected at night in October 1992 from light traps on Guana Island, British Vir-
gin Is. during rain (J. Krecek, unpubl.). No additional information is available on the
swarming habits of these three Parvitermes species.

We are indebted to J. A. Chase, J. de la Rosa G., and J. R. Mangold for their relent-
less dedication to collecting termites throughout the West Indies and elsewhere; S.C.
Jones for her diligent collecting efforts in Puerto Rico; D. S. Williams of the ICBR Elec-
tron Microscope Core Facility at the University of Florida, Gainesville, for assisting
with scanning electron microscopy; Dr. Sule Oygur of the American Museum of Nat-
ural History for the loan of P. pallidiceps paratypes and J. Krecek, J. Tsai, and N.-Y.
Su for critically reviewing and improving this contribution no. R-04475 of the Univer-
sity of Florida Experiment Station Series.


ARAUJO, R. L. 1970. Termites of the Neotropical Region. Chapter 12 in K.Krishna and
F. M. Weesner [eds.], Biology of Termites, Vol. 2. Academic Press, New York.
ARAUJO, R. L. 1977. Catalogo dos Isoptera do Novo Mundo. Acad. Brasileira de Cien-
cias, Rio de Janeiro, RJ. 92 pp.
BANKS, N. 1919. Antillean Isoptera. Bull. Mus. Comp. Zool. 62: 474-489 + 2 plates.
EMERSON, A. E. 1960. New genera of termites related to Subulitermes from the Ori-
ental, Malagasy, and Australian Regions (Isoptera, Termitidae, Nasutitermiti-
nae). American Mus. Nov. 1986: 1-28.

Florida Entomologist 78(4)

December, 1995

FRAHN. 1995. Survey of the termites of Puerto Rico. Florida Entomol. (submit-
KOVOOR, J. 1969. Anatomie comparee du tube digestif des termites II. Sous-Famille
des Nasutitermitinae. Insectes Soc. 16: 195-233.
NATION, J. A. 1983. A new method using hexamethyldisilazane for the preparation of
soft insect tissue for scanning electron microscopy. Stain Technol. 55: 347-352.
NOIROT, C., AND C. NOIROT-TIMOTHEE. 1969. The digestive system. Chapter 3 in K.
Krishna and F. M. Weesner [eds.], Biology of Termites, Vol. 1. Academic Press,
New York.
ROONWAL, M. L. 1970. Measurements of termites (Isoptera) for taxonomic purposes.
J. Zool. Soc. India 21: 9-66.
SANDS, W. A. 1965. A revision of the termite subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Bull. British Mus. Nat. Hist., Entomol.
Suppl. 4: 1-172.
SCHEFFRAHN, R. H., AND J. KRECEK. 1993. Parvitermes subtilis, a new subterranean
termite (Isoptera: Termitidae) from Cuba and the Dominican Republic. Florida
Entomol. 76: 603-607.
1994. Termites (Isoptera: Kalotermitidae, Rhinotermitidae, Termitidae) of the
West Indies. Sociobiology 24: 213-238.
SNYDER, T. E. 1924. Description of a new termite from Puerto Rico. Proc. Entomol.
Soc. Washington 26: 131-132.
SNYDER, T. E. 1949. Catalog of the termites (Isoptera) of the world. Smithson. Misc.
Collect. 112:1-490.


Florida Entomologist 78(4)

December, 1995


Department of Entomology and Nematology
P.O. Box 110620, University of Florida
Gainesville, FL 32611-0629


Rearing techniques for Ageniaspis citricola and Cirrospilus quadristriatus and
their host, the citrus leafminer, Phyllocnistis citrella, are discussed as related to a
classical biological control program. Methods for rearing the three trophic levels (cit-
rus plants, leafminers, and parasitoids) are described in detail. An average of 1,155
adult citrus leafminers were produced in a cage filled with 60-72 young citrus trees.
Between April and October 1994, a total of 15,230A. citricola were produced, with an
average of 750 adults per cage. Between late July and October 1994, a total of 1,328
C. quadristriatus were produced, with an average of 144 adults per cage. The two par-
asitoids were released into leafminer-infested citrus groves throughout Florida in a
classical biological control project.

Smith & Hoy: Rearing CLM and Two Parasitoids

Key Words: Citrus leafminer, Phyllocnistis citrella, Ageniaspis citricola, Cirrospilus
quadristriatus, biological control, rearing methods.


Se discuten las t6cnicas para la cria del minador de los citricos Phyllocnistis citre-
lla y dos de sus parasitoides,Ageniaspis citricola y Cirrospilus quadristriatus, en re-
laci6n a un program de control biol6gico clasico. Son descritos en detalle los m6todos
para criar los tres niveles tr6ficos plantss de citricos, minadores y parasitoides). Se
produjeron un promedio de 1,155 adults del minador en unajaula provista con 60-72
plants j6venes. De abril a octubre de 1994 se produjeron un total de 15,230 A. citri-
cola, con un promedio de 750 por jaula. De fines de julio a octubre de 1994 se produ-
jeron un total de 1,328 C. quadristriatus, con un promedio de 144 adults por jaula.
Como parte de un proyecto de control biol6gico clasico, se liberaron los dos parasitoi-
des en huertas de citricos infestadas por el minador, en various sitios del estado de Flo-

The citrus leafminer (CLM), Phyllocnistis citrella Stainton, (Lepidoptera: Gracil-
lariidae) is an important pest of citrus. It affects production and can augment the se-
verity of citrus canker (Pseudomonas citri Hasses) on damaged plants (Sohi & Sandhu
1968). The CLM originates from southeast Asia (Heppner 1993). Perhaps due to nat-
ural dispersal patterns and shipment of infested citrus, it has become established in
other citrus production areas throughout the world.
The adult CLM is a minute moth, 2.5 mm long with folded wings and with a 4.5
mm wingspan. Adults emerge from their pupal chambers early in the morning (Beat-
tie & Smith 1993). Mating, which lasts an average of 22 minutes (Pandey & Pandey
1964), occurs at dusk and early evening, 9 to 12 hours after emergence. Egg-laying be-
gins 1 to 8 days later (Badawy 1967; Ba-Angood 1977; Beattie & Smith 1993). A single
female can lay up to 20 eggs per night, for a total of more than 50 in her 5- to 20-day
lifetime (Beattie & Smith 1993). The translucent oval eggs are typically laid near the
midrib of young leaves on the under side of the leaf (Badawy 1967; Ba-Angood 1977;
Beattie & Smith 1993). Egg eclosion occurs 1 to 10 days later (Pandey & Pandey 1964;
Beattie & Smith 1993) with the young larva immediately burrowing under the waxy
cuticle of the leaf (Sohi & Verma 1965). The larva feeds on the cells of the epidermis,
creating silvery, serpentine mines along the leaf (Sohi & Verma 1965) often causing
the leaf to curl (Heppner 1993). During warm weather, the small larva progresses
through three feeding instars in 5 to 6 days and enters a fourth, non-feeding stage
(prepupa) for one day before forming a pupal chamber by folding over a leaf edge. The
pupal stage can last 6 to 22 days, depending on the time of year (Pandey & Pandey
1964). A generation is thus completed in 14 to 17 days during warm summer months
(Beattie & Smith 1993), but can be as long as 52 days in winter (Pandey & Pandey
The CLM was first recorded from Florida in May 1993 and has since dispersed
throughout the state (Heppner 1993). It is now also recorded in Alabama, Louisiana,
and Texas. The use of pesticides to control CLM is inefficient due to several reasons:
price requirements for multiple applications, CLM larval stages are protected within
their mines from topical applications, and the CLM may develop resistance to pesti-
cides. Research to develop an integrated pest management program in citrus groves
was initiated shortly after the CLM arrived. Classical biological control was identified
as a high priority component of the IPM program. As part of the classical biological

Florida Entomologist 78(4)

control project for CLM in Florida, two parasitic wasps were imported from Australia:
Ageniaspis citricola Logvinovskaya (Hymenoptera: Encyrtidae) and Cirrospilus
quadristriatus Subba Rao & Ramamani (Hymenoptera: Eulophidae) in April 1994
(Hoy & Nguyen 1994a, b, c). These parasitoids are native to Asia and reported to be
host specific to the CLM (Beattie 1992).
Synchronized rearing methods were developed in order to produce large numbers
of the CLM and its parasitoids so inoculative releases could be made in Florida. The
benefits of synchronized rearing for each species include: 1) the developmental stage
and age of the colony is known; 2) the purity of the colony is easier to maintain; and
3) the likelihood of introducing pests and diseases into the colony is reduced. The pur-
pose of this paper is to describe rearing methods for: 1) producing large numbers of cit-
rus trees in flush suitable for rearing CLM in the greenhouse; 2) the CLM; and 3) the
parasitoids A. citricola and C. quadristriatus.



Citrus was either grown from seed or obtained as seedlings or grafted trees from
local nurseries. Rough lemon (Citrus jambhiri Lushington) seeds were planted in a
1:1 mixture of potting soil and vermiculite in a cavity seedling tray (Hummert Inter-
national, Earth City, MO) containing 96 2.5 x 2.5 x 7.5 cm cavities. Three 7- to 10-cm
tall seedlings were transferred to each 3.8 liter plastic pot or 3.9 liter black plastic
nursery bag (Poly-Cel, Hummert International, Earth City, MO). Two weeks after be-
ing transplanted, the seedlings were fertilized with a long-acting, slow-release fertil-
izer (19-10-10 plus iron, Once, Grace-Sierra, Horticultural Products Company,
Milpitas, CA). Fertilizer was reapplied after 6 months. Seedlings were ready to use as
hosts for the CLM when the stems were approximately 5 mm in diam and 30 to 50 cm
tall. The time required to achieve this size depended upon the time of year the seeds
were planted, ranging from 7 months if planted in fall/winter to 4 months if planted
in spring/summer.
Grapefruit (Citrus x paradisi Macf) and sour orange grafted on trifoliate orange
(Citrus trifoliata (L.) Raf.) rootstock also are suitable as hosts for CLM. Both produce
a large amount of flush from the nodes after pruning (1/4 to 1/2 of each branch re-
moved) and leaf stripping. Grafted trees were obtained from nurseries when they
were approximately 60 cm tall. Grafted trees were treated to reduce pest infestations
by pruning and spraying them with oil (15 ml 97% petroleum oil, Ortho Volck Oil
Spray, Chevron Chemical Company, CA in 3.8 liter water). The pruned plants pro-
duced sufficient flush for use after approximately 2 weeks at 30 C and 80% relative
humidity. Other citrus varieties, including rough lemon, lime, trifoliate orange, and
single, also were donated as seedlings and reared as above. All proved suitable for
rearing the CLM.
Citrus plants were maintained either in a greenhouse or a shadehouse. Seedlings
were grown in a 6.6 x 9.1 m greenhouse covered with a shadecloth that provided 35%
shade. The average temperature in this greenhouse was 30 C (temperatures occasion-
ally reached a maximum of 37.8 C and a minimum of 23.3 C) and the average relative
humidity was 80% (90-100% for approximately 14 h per day, as low as 50% for a short
time during the heat of the afternoon). Trees, including those recently pruned, do-
nated material, and other extras, were also housed in a 6.1 x 24.4 m shadehouse con-
structed of 50% shadecloth. Recently-pruned plants were placed in the shadehouse
within 61 x 61 x 61 cm mesh cages (BioQuip, Gardena, CA) until they had produced

December, 1995

Smith & Hoy: Rearing CLM and Two Parasitoids 603

new flush with no CLM. All trees were watered when needed, usually 2 to 3 times per
Pest Problems. A variety of pests had to be managed in the greenhouse where trees
were reared, including citrus whitefly (Dialeurodes citri Ashmead), broad mite
(Polyphagotarsonemus latus Banks), and citrus red mite (Panonychus citri McGre-
gor). Trees in the greenhouse and shadehouse were monitored at least weekly for
pests. Broad mites were controlled by lightly hand dusting only the new flush with
sulfur (90% sulfur, Southern Agricultural Insecticide, Inc., Hendersonville, NC) semi-
weekly or when needed. Citrus whiteflies and citrus red mite were controlled by
spraying with 5% insecticidal soap solution (Safer, Inc. Eden Prairie, MN) semi-
weekly. The sulfur and insecticidal soap were applied on alternate weeks.
Citrus mealybug (Planococcus citri Risso) was an occasional pest in the shade-
house. Mealybugs were physically removed from plants when detected. Some trees
also became infested with scale insects, primarily Caribbean black scale (Saissetia ne-
glecta DeLotta) and cottony cushion scale (Icerya purchase Mask.). When scales of any
type were discovered, the adult scales were removed by hand and the plant was
sprayed with insecticidal soap, or the plant was discarded if the infestation was se-

Citrus Leafminer

CLM-infested foliage was initially obtained from citrus groves around Lake Alfred,
Florida in February 1994. Infested foliage was also obtained occasionally from trees
in Gainesville to supplement the colony. Occasionally, mines were found with dead
larvae (<5%), but no bacterial, fungal, or viral diseases were observed in CLM larvae
or pupae although detailed observations were not made.
Initially, isolation of adult moths from the infested foliage was difficult. Several
standard methods produced few adults or were very labor-intensive. Placing infested
leaves on a water-soaked cotton pad in glass petri dishes was attempted; moth emer-
gence rates were high, but collection was slow and it was a space- and labor-consum-
ing procedure. Infested foliage was placed in several dark containers of different sizes
fitted with one or more glass emergence tubes at the top, streaked with honey. The
moths preferred to rest on the leaves and did not fly toward the light unless disturbed.
The few that flew into the tubes often did not stay for a long period of time but re-
turned to the foliage to rest.
The most efficient emergence method tested involved placing infested leaves with
pupal chambers in clear plastic bags. Approximately 120 leaves were placed in each
30.4 x 25.2 cm bag. If only a few leaves were placed in a bag, a pad of moistened cotton
was added to prevent the leaves from drying out. The bags were then inflated by blow-
ing into them, and the end of the bag was twisted and secured. The bags were placed
under a fluorescent light in the laboratory. A high rate (averaging 81%) of adult CLM
emerged, usually early in the morning. Adult moths could be removed from the bags
every other day in late morning or early afternoon using a vacuum aspirator. The vac-
uum pump aspirated the moths through 5 mm plastic tubing into a 29.6-ml plastic
cup. Mouth aspiration is unsafe because the moth scales are allergenic. The bags were
wiped with paper towels after each aspiration to reduce condensation, although the
CLM adults did not seem to be adversely affected by free moisture.
After the moths were aspirated into a cup, they were fed honey by adding a honey-
soaked tissue (Kimwipe, Kimberly-Clark, Roswell, GA) or by streaking thin lines of
honey on the lid or sides of the cup. Adult CLM were allowed to feed for 1 to 2 h before
being placed in a large cage to mate and to oviposit on potted citrus trees in flush.

Florida Entomologist 78(4)

Citrus trees suitable for oviposition by CLM were those with young flush 1 to 2 cm
long. We placed 20 to 24 pots into a 76.2 x 114.3 x 91.4 cm screened cage. Honey was
streaked in fine lines on two 5 x 8 cm clear plastic sheets which were taped to the in-
side top of the cage frame as a food source for CLM adults. We added 175 to 250 adult
moths (sex ratio unknown) to each cage by placing the opened plastic cups on the floor
of the cage.
Fresh honey was streaked on the plastic sheets after 2 days. The trees were wa-
tered as needed, typically every three days. Cages were maintained in a 2.8 x 6 m
greenhouse covered with shadecloth which provided 35% shade. The greenhouse for
rearing CLM averaged 30 C (but occasionally reached a maximum of 37.8 C for a
brief duration and a minimum of 23 C) and 80% relative humidity (with a minimum
of 50% for approximately 1 h during the middle of the day). Because CLM adults sur-
vive best in >85% RH (J. Villanueva-Jim6nez, unpublished), we attempted to main-
tain high humidity (around 80-90%) by flooding the floor of the greenhouse once or
twice a day and/or by running portable humidifiers.
Young mines (1-3 mm long) were observed on the foliage after 4 to 6 days. Plants
were then used to rear the 2 parasitoid colonies or for maintaining the CLM colony
(Fig. 1). If the infested foliage was used for colony maintenance, leaves with pupal
chambers, which developed 9 to 12 days after adding adults to the cage, were removed
from the trees and placed in plastic bags in the laboratory for adult emergence as de-
scribed above. Adults emerged in bags 4 to 21 days after the first pupal chambers
To estimate productivity of the rearing, 3 large cages were selected at random be-
tween June and July and the number of CLM produced from 20 to 24 pots (containing
57 to 70 trees) per cage was recorded.
Pest Problems in the Citrus Greenhouse. To manage ants in the greenhouse, sticky
barriers (The Tanglefoot Company, Grands Rapid, MI) were applied to the legs of the
greenhouse benches. Additionally, commercial ant baits (Combat Insect Control Sys-
tems, Oakland, CA) were placed on the benches and in the cages. One species, Tapi-
noma melanocephalum, was especially difficult to control because they were not
controlled with commercial baits. To reduce infestation by T. melanocephalum, plants
were thoroughly watered in an attempt to flush any ant colonies from the pots before
they were placed in cages with the CLM and again before the CLM-infested trees were
transferred to cages containing the parasitoids.

Ageniaspis citricola

A. citricola is an endoparasitoid, parasitizing eggs and early instar larvae (Logvi-
novskaya 1983; Hoy & Nguyen 1994b; O. R. Edwards, personal communication) and
producing 1 to 10 individuals per single host. Both males and females are found, con-
trary to previous reports of thelytoky (Evans 1995). Unmated females produce only
male progeny, suggesting that this species is arrhenotokous (0. Edwards, personal
communication). When first instar CLM larvae were visible on foliage, typically 4 to
6 days after CLM adults were introduced, the trees were ready to be used forA. citri-
cola colonies. At this stage, foliage will have both eggs and mines containing first in-
star larvae. Prior to placing the trees in cages withA. citricola, they were thoroughly
watered to reduce ant densities.
Parasitoid cages were maintained in the same greenhouse where citrus plants
were housed. Fifteen to twenty pots, usually containing three infested citrus trees
each, were added to each cage and 50 to 75 A. citricola adults were then introduced.
The sex ratio of the introduced adults was unknown, although the average sex ratio

December, 1995

Smith & Hoy: Rearing CLM and Two Parasitoids

Figure 1. A flow diagram indicating the steps and approximate time involved in
rearing citrus, the CLM and the two parasitoids. See text for more detail.

was 1 male:1.8 females based on 204 individuals sexed from random samples from 7
different cages between October and November. Honey was streaked on several pieces
of plastic suspended from the top of the cage to provide food for parasitoid adults.
Honey was reapplied after 48 hours and the plants were watered as needed.
Between April and October, new A. citricola adults emerged 16 to 18 days after
adults were introduced into the cages. The parasitoids were collected by pruning foli-
age containing CLM pupal chambers from the trees after 15 to 17 days. The exact time
was judged by opening a few pupal chambers to determine if the majority of A. citri-
cola pupae were dark brown. The leaves were then placed in inflated plastic bags with

Florida Entomologist 78(4)

paper towels, and the bags were checked daily to collect wasps and to eliminate con-
densation on the bags. Adult parasitoids were collected once a day via mouth aspira-
tion into a 50 ml vial containing tissues in the bottom to provide a soft surface. A
honey-soaked strip of tissue was placed in the vial to provide food for the adult para-
If the parasitoids were to be released into citrus groves, the vials were placed in a
growth chamber at 19C until they were shipped. Parasitoids were delivered by auto-
mobile or shipped via overnight mail in styrofoam containers with blue ice packs to
maintain temperatures at approximately 17'C. If the parasitoids were used to main-
tain the colony, they were allowed to feed and then were introduced into a new cage
with trees infested with CLM eggs and early instar larvae. AdultA. citricola only live
for 2 to 5 days, so they must be used for colonies or shipped to growers as soon as they
are collected.
Problems Encountered. We encountered several problems in rearing A. citricola.
Initial attempts to rear the parasitoids in a rearing room or in a shaded alcove failed
to produce progeny, possibly because light intensity was low. Cages were then held
within a greenhouse with a relative humidity below 60%. In an attempt to increase
relative humidity, plastic sheeting was placed over the cages and a humidifier was
placed under the greenhouse bench. These cages also did not produce wasps, possibly
due to the high temperatures (>38'C) that were reached under the plastic.
Adult A. citricola are small (<2 mm) and difficult to collect from the cages because
they tend to rest on the foliage and do not go to the top of the cage. High rates of adult
A. citricola emergence were achieved by placing leaves with parasitized CLM pupae
into inflated plastic bags, in a manner similar to that used to obtain adult CLM emer-
gence. Adult parasitoids could easily be aspirated from the bags. One problem with
plastic bags is the amount of condensation that develops on the inside.A. citricola are
easily trapped and die in free moisture, so the bags must be wiped dry at least once a

Cirrospilus quadristriatus

C. quadristriatus is an ectoparasitoid of late instar larvae of the CLM (Beattie
1992; Hoy & Nguyen 1994c) producing a single individual per host. Both males and
females are produced. Under our conditions, foliage inoculated with CLM reached the
suitable host stage in 7 to 10 days. Trees with third and fourth instar larvae were wa-
tered and transferred into a new cage. Because adults live for almost 2 weeks, a mixed
age class of trees was added to the cages (same size as used with the CLM). One third
of the cage was filled with trees that had been infested with CLM for 7 to 10 days. We
then added 50 to 75 adult wasps (unknown sex ratio, extremely difficult to sex) to the
cage. After 2 days, another third of the cage was filled with new trees that were 7 to
10 days old. The last third was filled after another 2 days. Honey was streaked every
other day on plastic sheets suspended in the cages and the plants were watered when
needed. These cages were held in the citrus greenhouse.
Adult C. quadristriatus began emerging 11 to 13 days after they were introduced
into the cages. Adults were aspirated from the cage every afternoon, when the wasps
were most active. C. quadristriatus are easy to locate in cages due to their large size,
orange color, and because they typically rest on the top of the leaves. After allowing
the wasps to emerge in the cage for approximately one week, leaves with intact pupal
chambers were pruned off each plant and placed in plastic bags to allow additional C.
quadristriatus adults to emerge. This procedure was adopted to allow early wasps to
emerge while allowing later larvae to continue to develop. Adult parasitoids were fed
with a honey-soaked piece of tissue or by streaking thin lines of honey in the vial.

December, 1995

Smith & Hoy: Rearing CLM and Two Parasitoids 607

Adult C. quadristriatus can be held longer before being supplied to growers or used
in colony maintenance because they live for approximately 2 weeks. The adults were
placed in a growth chamber held at 19C until they were used or shipped in the same
manner as A. citricola. If held for a longer period of time, the adults were supplied
with fresh honey every 48 hours.


Citrus Leafminer

A mean of 1426 168 ( SD) intact pupal chambers were produced in each cage.
The average number of leaves infested per tree was 10.2 7.9, with a range from 0 to
35 leaves. The average number of CLM pupae produced per leaf was 2.8 1.8. The
maximum number of pupal chambers on one leaf was 13. Approximately the same
number of pupal chambers were located on the lower surface of the leaf (1.8 1.1) as
compared to the upper surface (1.1 + 1.0).

Ageniaspis citricola

Between April and October, the productivity of cages (n=21) used to rearA. citri-
cola was evaluated. An average of 750 adults (+ 410) was produced from 18 to 24 pots
containing an average number of 60 trees in each cage. The maximum number of
adults from a single cage was 1491, while the minimum was 109. One to tenA. citri-
cola develop from a single CLM pupa; the average number ofA. citricola individuals
emerging per CLM pupal chamber in our greenhouse cages was 2.8 1.1.

Cirrospilus quadristriatus

Since their release from quarantine in late July, an average of 144 (+ 25.8) parasi-
toids have been reared from each of 9 cages, each containing approximately 60 trees.
The maximum number of adults obtained from a single cage was 171 while the mini-
mum was 101. Reasons for low rate of production of these parasitoids are unknown.
We do not know how many eggs are laid by each female or the preferred relative hu-
midity. Also, as already stated, only one C. quadristriatus is produced per host.


The methods described provide parasitoids for inoculative releases, but do not al-
low large scale augmentative releases. Rearing is time consuming and expensive be-
cause 3 trophic levels must be maintained. A total of 15,230 A. citricola were reared
between April and October 1994, and 1328 C. quadristriatus were reared between late
July and October 1994. This required 225.7 m2 in greenhouse and shadehouse space,
approximately 2,500 citrus trees, and one full-time employee devoted solely to this
project (plus some hours performed by other employees). The potential for augmenta-
tive releases would be improved if an artificial diet was available either for the CLM
or the parasitoids.


We thank Ru Nguyen for his advice on rearing and his efforts in rearing the para-
sitoids in the quarantine facilities. We thank Owain Edwards for advice and assis-

608 Florida Entomologist 78(4) December, 1995

tance and Juan Villanueva-Jim6nez for the spanish translation of the abstract.
Ayyamperumal Jeyaprakash, Denise Johanowicz, Mark Pomerinke, Jim Presnail,
Shawn Rogers, and three anonymous reviewers provided advice on the manuscript.
This work was funded in part by the Citrus Production Research Advisory Council.
This is journal series number R-04482.


BA-ANGOOD, S. A. S. 1977. A contribution to the biology and occurrence of the citrus
leafminer, Phyllocnistis citrella Staint., in the Sudan. Zeit. Angew. Entomol. 83:
BADAWY, A. 1967. The morphology and biology of Phyllocnistis citrella Staint., a citrus
leaf-miner in the Sudan. Bull. Soc. Entomol. Egypte LI: 95-103.
BEATTIE, G. A. C. 1992. Biological control of citrus leaf miner introduction and re-
lease of natural enemies. Final Report Project C/0031, NSW Agriculture.
BEATTIE, G. A. C., AND D. SMITH. 1993. Citrus leafminer. Agfact H2.AE.4, second edi-
tion. NSW Agriculture.
EVANs, G. A. 1995. Discovery of the male ofAgeniaspis citricola (Hymenoptera: En-
crytidae) parasitoid of the citrus leafminer Phyllocnistis citrella (Lepidoptera:
Gracillaridae). Florida Entomol. 78: 134-136.
HEPPNER, J. B. 1993. Citrus leafminer (CLM) Phyllocnistis citrella Stainton. Florida
State Collection ofArthropods, DPI/FDACS.
HOY, M. A., AND R. NGUYEN. 1994a. Classical biological control of the citrus leafminer
in Florida. Citrus Industry, April: 22, 25.
HOY, M. A., AND R. NGUYEN. 1994b. Classical biological control of the citrus leafminer
in Florda: a progress report. Citrus Industry, June: 61-62.
HOY, M. A., AND R. NGUYEN. 1994c. Classical biological control of the CLM: release of
Cirrospilus quadistriatus. Citrus Industry, November: 14.
LOGVINOVSKAYA, T. V. 1983. A new species of Ageniaspis Dahlbom 1857 (Hy-
menoptera, Encryrtidae) from Vietnam. Entomol. Rev. 62: 150-152.
PANDEY, N. D., AND Y. D. PANDEY. 1964. Bionomics of Phyllocnistis citrella Stt. (Lep-
idoptera: Gracillariidae). Indian J. Entomol. 26: 417-422.
SOHI, G. A. S., AND M. S. SANDHU. 1968. Relationship between Citrus leaf-miner
(Phyllocnistis citrella Stainton) injury and citrus canker (Xanthomonas citri
(Hasse) Dowson) incidence on Citrus leaves. J. Res. Punjab Agric. Univ. 5: 66-
SOHI, G. A. S., AND G. C. VERMA. 1965. Feeding habits of Phyllocnistis citrella Stain-
ton in relation to the anatomical structure of the leaf. Indian J. Entomol. 27:

Scientific Notes


USDA-APHIS-PPQ-IFA, 3505-25th Avenue, Gulfport, MS 39501

There are a number of published records of predation on the newly mated queens
of the red imported fire ant, Solenopsis invicta Buren, including attacks on founding
queens by numerous insectivorous predators (Edwards et al. 1974, Glancey 1981, Lu-
cas & Brockman 1981, Nichols & Sites 1991). Whitcomb et al. (1973) listed 22 verte-
brate and invertebrate predators. Among these, only one spider Lycosa timuga
Wallace was observed feeding on founding queens. Nyfeller et al. (1988) collected the
remains of 34 S. invicta queens from the webs of 100 Latrodectus mactans (F.) spiders
in cotton fields. Red imported fire ant queens made up over 15% of the total prey of L.
mactans. In the same study, 16 additional spiders from eight families were listed as
predators of red imported fire ants. Of these, only two (Neoantistea sp. and Phidippus
audax) were observed with S. invicta queens as prey.
Beginning at 1015 hours on 23 August 1990, a single black and yellow garden spi-
der, Argiope aurantia Lucas, was observed capturing alate red imported fire ant fe-
males in flight. The spider, a penultimate female, had established her vertical web
approximately 1.0 m (at its center) above and 1.0 m to the northeast of an active S. in-
victa mound [category 9 on the Lofgren & Williams (1982) scale]. The spiraled area of
the web covered an area of approximately 0.8 m. Prevailing winds placed the spider's
web directly downwind of the mound. Alate females were seen emerging from the col-
ony, ascending adjacent plants (primarily wild grasses and a wild sparkelberry bush
Vaccinium arboreum Marsh.) before launching themselves into the air. During a pe-
riod of 30 min, 275 alate females were observed leaving the mound by flight. During
this same period, the garden spider snared 37 ants. For the first twelve min of this ac-
tivity, the spider reacted to the arrival of the alate females by rushing to the point of
web contact and quickly subduing the prey by swathing it with silk. Twenty-four of
the alates were handled in this manner. Nine of these were first envenomated by the
spider before swathing. The remaining 15 were either bitten after swathing or, as was
the case for the last 5 taken, merely swathed without a bite. The final 13 alates to con-
tact the web were handled differently from the first twenty-four. The spider reacted by
jerking the web to dislodge the prey (n=5). Failing to dislodge the prey, the spider
would cautiously approach the captured queen and carefully sever the lines to which
the ant was attached to the web allowing it to drop from the web to the ground strata
(n=8). Ants that were shaken from the web climbed an adjacent plant and again took
flight (one of which was again caught in the web). Those alates severed from the web
were unable to regain flight and, upon closer examination, were found to have web-
bing attached to their wings and body.
Research into biological control of imported fire ants has concentrated on patho-
gens and parasitoids almost to the exclusion of generalist predators (Allen 1980, Jou-
venaz 1983, Jouvenaz et al. 1980, Williams 1980, Williams & Whitcomb 1974).
Although only a single observance, a 15% reduction by a spider in alate females re-
ported here may represent a more widespread and frequent phenomena. Further
study of the potential of such generalist predators as biocontrol agents of S. invicta
should be encouraged.

610 Florida Entomologist 78(4) December, 1995


The serendipitous placement of an Argiope aurantia web in the direct flight path
of emerging alate fire ants succeeded in disrupting the mating flight of approximately
15% of the observed queens, either through direct predation or indirectly by hamper-
ing their ability to fly (capture and release).


ALLEN, G. E., AND J. D. KNELL. 1980. Pathogens associated with the Solenopsis sae-
vissima complex in South America. Proc. Tall Timbers Conf. Ecol. Anim. Con-
trol Habitat Manag. 7: 87-94.
EDWARDS, G. B., J. F. CARROLL, AND W. H. WHITCOMB. 1974. Stoidis aurata, a spider
predator of ants. Florida Entomol. 57: 337-346.
GLANCEY, B. M. 1981. Two additional dragonfly predators of queens of the red im-
ported fire ant, Solenopsis invicta Buren. Florida Entomol. 64: 194-195.
JOUVENAZ, D. P. 1983. Natural enemies of fire ants. Florida Entomol. 66: 111-121.
JOUVENAZ, D. P., W. A. BANKS, AND J. D. ATWOOD. 1980. Incidence of pathogens in fire
ants, Solenopsis spp., in Brazil. Florida Entomol. 63: 345-346.
LOFGREN, C. S., AND D. F. WILLIAMS. 1982. Avermectin Bla: a highly potent inhibitor
of reproduction by queens of the red imported fire ant (Hymenoptera: Formi-
cidae). J. Econ. Entomol. 75: 798-803.
LUCAS, J. R., AND H. J. BROCKMAN. 1981. Predatory interactions between ants and
antlions. J. Kansas Entomol. Soc. 54: 228-232.
NICHOLS, B. J., AND R. W. SITES. 1991. Ant predators of founder queens of Solenopsis
invicta (Hymenoptera: Formicidae) in Central Texas. Environ. Entomol. 20:
NYFELLER, M., D. A. DEAN, AND W. L. STERLING. 1988. The southern black widow spi-
der, Latrodectus mactans (Araneae: Theridiidae), as a predator of the red im-
ported fire ant, Solenopsis invicta (Hymenoptera: Formicidae), in Texas cotton
fields. J. Appl. Entomol. 106: 52-57.
WHITCOMB, W. H., A. BHATKAR, AND J. C. NICKERSON. 1973. Predators of Solenopsis
invicta queens prior to successful colony establishment. Environ. Entomol. 2:
WILLIAMS, R. N. 1980. Insect natural enemies of fire ants in South America. Proc. Tall
Timbers Conf. Ecol. Anim. Control Habitat Manag. 5: 123-124.
WILLIAMS, R. N., AND W. H. WHITCOMB. 1974. Parasites of fire ants in South America.
Proc. Tall Timbers Conf. Anim. Control Habitat Manag. 5:49-59.

Scientific Notes


'Department of Zoology, Southern Illinois University, Carbondale, IL 62901

'Department of Entomology, National Museum of Natural History,
Washington, D.C. 20560

3Wilbur R. Enns Entomology Museum, Department of Entomology,
University of Missouri, Columbia, MO 65211

Diaditus tejanus was described in 1980 by Giacchi, who listed it from Mexico (Te-
hauantepec, Oaxaca; Chuminopolis, Yucatan) and Texas (Alvin, Brownsville, Mer-
cedes). This species represented only the second of the genus reported from America
north of Mexico. D. pictipes, described by Champion in 1898, is known from Guate-
mala, Mexico, and Texas (Giacchi 1982). Nothing else has been published on the biol-
ogy of either species.
Recently, while examining material from the entomology collections at the Univer-
sities of Missouri and Mississippi, we discovered 25 specimens of D. tejanus that had
been collected in Florida and Mississippi. These collection records represent an east-
ward range extension from Alvin to Highlands Co., Florida (see below), of approxi-
mately 1,300 km, and a northeastern extension from Yucatan to Highlands Co. of
approximately 1,100 km. These distribution records, along with those of Giacchi
(1980, 1982), suggest that the range of this species may be restricted by coastal fac-
tors, because all records are from Florida to Mexico along the Gulf Coast and across
the isthmus of Mexico at Tehuantepec.
The label information and repositories for these specimens are given below; collec-
tion abbreviations include: University of Missouri-Columbia, UMC; University of
Mississippi, UMISS; Southern Illinois University at Carbondale, SIUC; and National
Museum of Natural History, NMNH:
MISSISSIPPI: Hancock Co., Pt. Clear Island, 12-VIII-1986, Paul K. Lago (Coll.) (5
6 6), Sam Testa (Coll.) (4 6 6); same label information other than date, 15-VIII-1986,
Paul K. Lago (Coll.) (6 6 6), Sam Testa (Coll.) (3 6 6); Hancock Co., 1.5 mi SW lake-
shore, 12-VIII-1986, Sam Testa (1 6) (all specimens, UMISS).
FLORIDA: Highlands Co., Highlands Hammock St. Pk., black light, 30-III-1979,
E. Riley & D. LeDoux (Coll.) (1 6, UMC); Gadsden Co., 15 mi SW Quincy, 21-VIII-
1992, R. Beiriger (Coll.) (1 6, 1 9, NMNH; 1 6, SIUC; 2 6 6, UMC).
We thank Paul K. Lago, Department of Biology, University of Mississippi, Univer-
sity, for the loan of the Mississippi specimens mentioned above.


The geographic range of Diaditus tejanus Giacchi, based on 25 specimens, is ex-
tended eastward in the United States from Texas through Mississippi to Florida.


CHAMPION, G. C. 1897-1901. Insecta: Rhynchota (Hemiptera-Heteroptera). Vol. II, in
Goodwin and Salvin [eds.]. Biologica Centrali-Americana. London. 1898: 33-

612 Florida Entomologist 78(4) December, 1995

GIACCHI, J. C. 1980. Una nueva especie para el genero Diaditus Stal, 1859 (Stenopo-
dainae-Reduviidae). Revista Sociedad Entomologia Argentina 39 (1-2): 1-4.
GIACCHI, J. C. 1982. Revision de los stenopodainos americanos. V. El genero Diaditus
Stal, 1859 (Heteroptera, Reduviidae). Physis (Buenos Aires), Secc. C., 41(100):

Scientific Notes


Dept. of Ecology and Evolution
State University of New York
Stony Brook, NY 11794-5245

In the course of keeping Orchelimum katydids (Tettigoniidae: Conocephalinae)
collected as nymphs and adults from the southeastern U.S. and Washington, D.C., I
have obtained a small number of puparia of a parasitic tachinid fly, as follows: two pu-
paria from a single adult male O. agile from Gainesville, FL; one puparium from an
early instar female 0. pulchellum from Appling, Columbia Co., GA; one puparium
from an adult male 0. silvaticum and one puparium from an adult male O.pulchellum
from Montgomery, AL; one puparium from an adult male 0. pulchellum from Gautier,
MS; one puparium from an adult male 0. nigripes from Fairview-Riverside State
Park, LA; and one puparium from an early instar female 0. pulchellum from Wash-
ington, D.C. The host exhibits a characteristic syndrome in response to the parasite:
the katydid (nymph or adult) becomes sluggish and a distinct bulge develops in the
abdomen. Within a few days, a puparium appears outside of the katydid and the ka-
tydid dies.
Three of the pupae were reared in the laboratory (from Louisiana 0. nigripes, Mis-
sissippi 0. pulchellum, and Alabama 0. pulchellum) and a representative adult (from
the Louisiana 0. nigripes) was identified as Ormia lineifrons. The identified fly has
been deposited in the collection of the U.S. National Museum. Since an adult fly was
actually reared and identified only from a single 0. nigripes, the association of Ormia
lineifrons with the other parasitized Orchelimum species (0. pulchellum, 0. silvati-
cum, and 0. agile) must remain tentative, although all the puparia and reared adults
appeared to be the same. The only hosts previously known for this fly are Neocono-
cephalus katydids, especially N. triops (T. J. Walker 1994 in litt., Burk 1982). Accord-
ing to Burk (1982), 0. lineifrons is attracted to tape recordings of the calling song of
N. triops; Walker reports (in litt.) that each year 0. lineifrons is attracted in small
numbers to mole cricket (Scapteriscus vicinus) sound trapping stations, although it
does not parasitize mole crickets. (A Brazilian Ormia species is currently being used
as a biocontrol agent for this introduced cricket in Florida.)
While some acoustically orienting tachinids apparently depend almost entirely on
host calling song to locate hosts (e.g., Lakes-Harlan & Heller 1992), this reliance on
calling song to locate individual hosts is not always so complete (e.g., Walker & Win-
eriter 1991). Given that two of seven parasitized Orchelimum were nymphs (and
hence silent), 0. lineifrons apparently does not depend strongly (perhaps does not de-
pend at all) on song to find individual Orchelimum hosts, although it could be that fe-
male flies are attracted to the general area around calling males. In the absence of
playback experiments using Orchelimum songs and gravid 0. lineifrons females, it is
impossible to assess precisely the role of calling song in 0. lineifrons host searching
behavior. It may be that 0. lineifrons orients acoustically to Neoconocephalus (Burk
1982), the primary host, while other katydids are attacked opportunistically. Ob-
served rates of tachinid parasitism of Orchelimum are very low.
Tachinid parasitism of Orchelimum katydids has been reported previously only
very briefly and generally by Feaver (1983), who observed in her Michigan study pop-

Florida Entomologist 78(4)

ulation two 0. nigripes individuals that had been parasitized by an undetermined ta-
chinid. In this note I report the first specific identification of a tachinid parasite of
I thank Dr. N. Woodley of the Systematic Entomology Laboratory of the U.S.D.A.
for identifying the fly and Dr. T. J. Walker for providing information on flies, katydids,
and crickets. I thank D. J. Funk, M. C. Keese, T. J. Walker, and an anonymous reviewer
for their valuable comments on this manuscript. My Orchelimum work has been sup-
ported by the National Science Foundation, the Washington Biologists' Field Club, the
Explorers' Club, the Theodore Roosevelt Fund of the American Museum of Natural
History, the Sigma Xi Scientific Society, and the Florida Entomological Society.


Ormia lineifrons was identified as a parasite of Orchelimum nigripes (and, tenta-
tively, of several other species of Orchelimum katydids), inhabiting both nymphs and
adults. These observations document both a new host genus for 0. lineifrons and the
first specific identification of a tachinid from Orchelimum.


BURK, T. 1982. Evolutionary significance of predation on sexually signalling males.
Florida Entomol. 65: 90-104.
FEAVER, M. 1983. Pair formation in the katydid Orchelimum nigripes (Orthoptera:
Tettigoniidae), pp. 205-239 in D. T. Gwynne and G. K. Morris [eds.], Ortho-
pteran Mating Systems: Sexual Competition in a Diverse Group of Insects.
Westview Press, Boulder, CO.
LAKES-HARLAN, R., AND K.-G. HELLER 1992. Ultrasound-sensitive ears in a parasi-
toid fly. Naturwissenschaften 79: 224-226.
WALKER, T. J., AND S. A. WINERITER 1991. Hosts of a phonotactic parasitoid and lev-
els of parasitism (Diptera: Tachinidae: Ormia ochracea). Florida Entomol. 74:

December, 1995

Scientific Notes 617


'USDA-ARS, Sugarcane Research Unit
P.O. Box 470, Houma, LA 70361-0470

'Department of Entomology
Louisiana State University
Life Science Bldg., Baton Rouge, LA 70803-1710

3USDA-ARS, Sugarcane Production Research
Star Rte. Box 8, Canal Point, FL 33438

The sugarcane delphacid, Perkinsiella saccharicida Kirkaldy, an insect pest of sug-
arcane, was first discovered in Louisiana on October 19, 1994, in a sugarcane field ap-
proximately 58 km southeast of Lafayette. Identification was provided by F. W. Mead,
Florida Department of Agriculture, Division of Plant Industry, P.O. Box 1269, Gaines-
ville, FL 32614.
The sugarcane delphacid is a recent introduction to North America having been
first reported in Florida in 1982 (Sosa 1985), Georgia in 1983 (Nguyen 1984), Texas in
1989 and Mexico in 1991 (Meagher et al. 1991). The sugarcane delphacid is probably
native to Papua, New Guinea, but with the movement of sugarcane, it is widespread
in Java, Taiwan, southern China, Malaysia, and eastern Australia. It is also estab-
lished in the Hawaiian Islands, Mauritius, Reunion, Madagascar, and South Africa
(Fennah 1969). In the Western Hemisphere, Risco (1969) reported the sugarcane del-
phacid in Ecuador in 1966 and in Peru in 1967. Feeding by nymphs and adult ovipo-
sition cause some plant damage (Allsopp & Bull 1990), but of principal concern is the
insect's ability to vector 1. .. sp., the causal agent of Fiji disease (Francki & Griv-
ell 1972).
After the initial discovery of the pest in Louisiana, the sugarcane producing par-
ishes (counties) of Louisiana were sampled during November 1994 to determine sug-
arcane delphacid population densities and geographic distribution within the state.
One to four fields were sampled per parish in each of 20 sugarcane producing par-
ishes. These 20 parishes comprised about 159,296 ha of cultivated sugarcane. Fields
of harvestable cane (about 3.8 m tall), uniformly spaced within each parish (about 8
km apart), were selected for survey. Visual counts of nymphs and adults were made
on 4 sugarcane stalks at 10 sites about 3 m apart along one or two field edges. Addi-
tionally, one leaf (3-5 down from the whorl) was examined for oviposition.
The sugarcane delphacid was found in 22 of the 60 fields sampled and in 13 of the
20 parishes sampled. The sugarcane in these 13 parishes comprises 79% (about
125,821 ha) of the total sugarcane cultivated in Louisiana. Numbers of individuals
(adults + nymphs) per locations ranged from 0 to 12. The highest density was 0.3 per
stalk in a field in St. Mary Parish. When the sugarcane delphacid was initially found
in Florida, densities ranged from 0.5 per stalk to 35.7 per stalk (Sosa 1985). No ovipo-
sition was detected; oviposition was obviously occurring but apparently at such low
levels that it was not detected.
Although the sugarcane delphacid was found in very low numbers, our survey de-
tected an infestation gradient with the highest populations found in the coastal par-
ishes and decreasing density in the inland parishes (Fig. 1). Immatures were, in
general, found in parishes with high adult numbers.

Florida Entomologist 78(4)

Figure 1. Distribution of the sugarcane delphacid in Louisiana sugarcane. Group-
ings in legend are based on the mean total number of adults and nymphs per location
within a parish.

Densities of the sugarcane delphacid remained low in Louisiana sugarcane fields
through the winter. Sweep net samples taken from 100 fields in late May and early
June of 1995 did not detect any adults or nymphs. Because of potential of the sugar-
cane delphacid to become an economic pest, we will continue monitoring field densi-
ties and geographic distribution of this insect in Louisiana.
Thanks are extended to Lance Rodriguez, Louisiana State University Agricultural
Center, Baton Rouge, LA and Griffin Bell, USDA-ARS, Canal Point, FL for technical
support. Voucher specimens were deposited in the Louisiana State University Insect
Collection and the Florida Department of Agriculture, Division of Plant Industry,
Gainesville, FL.


The sugarcane delphacid, Perkinsiella saccharicida Kirkaldy, was discovered in
Louisiana, 19 October 1994. This insect is a new record for Louisiana and was found
in 13 of 20 sugarcane producing parishes surveyed.


ALLSOPP, P. G., AND R. M. BULL. 1990. Sampling distributions and sequential sam-
pling plans for Perkinsiella saccharicida Kirkaldy (Hemiptera: Delphacidae)
and Tytthus (Hemiptera: Miridae) on sugarcane. J. Econ. Entomol. 83: 2284-

December, 1995

Scientific Notes 619

FENNAH, R. G. 1969. Damage of sugarcane by fulgoridea and related insets in relation
to the metabolic state of the host plant, pp. 367-389 in J. R. Williams, J. R. Met-
calfe, R. W. Mungomery and R. Mathes [eds.] Pests of sugar cane Elsevier Pub.
Co., Amsterdam.
FRANCKI, R. I. B., AND C. J. GRIVELL. 1972. Occurrence of similar particles in Fiji dis-
ease virus-infected sugar cane and insect vector cells. Virol. 48: 305-307.
BREENE. 1991. Documentation of two potential insect pests of south Texas sug-
arcane. Southwestern Entomol. 16: 365-366.
NGUYEN, RU, O. SOSA, JR., AND F. W. MEAD. 1984. Sugarcane delphacid, Perkinsiella
saccharicida Kirkaldy 1903 (Homoptera: Delphacidae). Florida Dept. Agric. &
Consumer. Serv. Entomology Circular 265.
RISCO, S. H. 1969. Notas adicionales sobre el "saltahoja" de la cania de azucar Perkin-
siella saccharicida K. Revista Peruana de Entomol. 9: 181-187.
SOSA, JR., 0. 1985. The sugarcane delphacid, Perkinsiella saccharicida (Homoptera:
Delphacidae), a sugarcane pest new to North America detected in Florida. Flor-
ida Entomol. 68: 357-360.

Scientific Notes


Entomology & Nematology Department, University of Florida,
Gainesville, FL 32611-0620, USA

'Current address: Crofton, Baldhoon Road, Laxey,
Isle of Man IM4 7NA, United Kingdom

Larra is a largely tropical genus of digger wasps (Sphecidae) with atypical behav-
ior. Typical sphecid females sting and paralyze other arthropods which then are taken
to cells where they serve as food for larvae. Larra females attack and sting mole crick-
ets (Gryllotalpidae), which suffer paralysis for only a few minutes. The Larra females
oviposit on the mole crickets that they have paralyzed, and the neonate larvae develop
as external parasitoids on active hosts (Bohart & Menke 1976). The only known hosts
of Larra are mole crickets.
Larra analis F. is the only species native to coastal southeastern USA, and its host
is Neocurtilla hexadactyla (Perty), the only mole cricket native to this region. Three
immigrant species of mole crickets of the genus Scapteriscus arrived in the southeast-
ern USA about 1900. Tens of thousands of these Scapteriscus mole crickets have been
examined by personnel of the University of Florida's mole cricket research program
since 1978, but none was found with an egg or larva of L. analis. This is strong evi-
dence that L. analis does not attack Scapteriscus spp. in nature.
Some mole cricket species are pests of agriculture and horticulture. Notable exam-
ples are Gryllotalpa orientalis Burmeister in Hawaii, Scapteriscus didactylus (La-
treille) in Puerto Rico and some other West Indian Islands, and Scapteriscus vicinus
Scudder, S. abbreviatus Scudder, and S. borellii Giglio-Tos in the southern USA

620 Florida Entomologist 78(4) December, 1995

(Frank 1994). These five species are immigrants in these areas and have been subject
to classical biological control, i.e., the attempted introduction of natural enemies from
their homelands into the areas where they are immigrants.
The earliest attempts at classical biological control of mole crickets used species of
Larra. Larra amplipennis (F. Smith) introduced in 1921 from the Philippines and L.
bicolor F. introduced in 1924 from Brazil did not become established in Hawaii, but L.
polita (F. Smith) subspecies luzonensis Rohwer, introduced in 1925 from the Philip-
pines, did become established (Williams 1928, Bohart & Menke 1976). Larra bicolor,
introduced from Belem, Para, Brazil, became established in Puerto Rico by 1941 (Wol-
cott 1938, 1941). We have discovered no assessment of the effectiveness of these wasps
in suppressing mole cricket populations in Hawaii and Puerto Rico.
An attempt was made in the 1940s to introduce L. bicolor into Florida, summa-
rized by Frank (1990). Plantings were made of Spermacoce verticillata L. and Hyptis
atrorubens Poit. (nectar-bearing plants favored by L. bicolor) at Gainesville, and flow-
ers of the latter attracted L. analis. Unfortunately, L. bicolor adults and larvae
shipped from Brazil were dead on arrival in Florida and the attempt was not repeated
in that decade.
A renewed effort was made to import L. bicolor into Florida beginning in 1979 un-
der the leadership of R. I. Sailer, as part of the University of Florida's mole cricket re-
search program. J. L. Castner, H. G. Fowler, W. G. Hudson, and J. R. Reinert
(University of Florida) and E. Abreu (University of Puerto Rico) participated. Puerto
Rican populations were surveyed and five sites for release were identified in Florida;
these sites were prepared with plantings of Spermacoce verticillata. In 1981, scores of
female wasps were released at Ft. Lauderdale, Gainesville, and Tampa, in 1982 at
Bradenton and Ft. Lauderdale, and in 1983 near Lakeland. However, a population of
L. bicolor became established only at Ft. Lauderdale (Sailer 1985, Frank 1990).
By late 1984, the Ft. Lauderdale L. bicolor population occupied two sites: a golf
course, and the University of Florida's Agricultural Research Station, about 1 km dis-
tant. Attempts to expand the population to additional sites were unsuccessful (Cast-
ner 1988a). Mole crickets were trapped to assess the proportion infected with eggs or
larvae of L. bicolor. Only S. abbreviatus were found infected; however, the combined
number of trapped S. borellii and S. vicinus accounted for only 15% of the total (n =
677) collected in pitfall traps (Castner 1988a). Only 1% of mole crickets examined
were infected with eggs or larvae (Castner 1988a). Establishment ofL. bicolor only at
Ft. Lauderdale and not at the four more northerly sites in Florida suggested that this
wasp, of tropical origin, could not withstand colder or longer winters farther north in
Florida. This theory was supported by poor survival of wasp pupae overwintered out-
doors experimentally at Gainesville (Castner 1988a).
F. D. Bennett joined the mole cricket research program in 1985. He shared J. L.
Castner's view that "Puerto Rican" L. bicolor had not become established in northern
Florida because of the equatorial origin of this biotype at low altitude in Belem. He
thought that a Larra stock from southern South America, or from high altitude else-
where in South America, might be better adapted to survive in northern Florida. Al-
though Larra females of other species had been observed to attack Scapteriscus mole
crickets in South America (i. e., Uruguay and southern Brazil), only small numbers of
these other Larra had been observed, and their identity was tenuous due to incom-
plete systematic treatment (Frank 1990). Bennett began studies on Larra with C. J.
Pruett at Santa Cruz de la Sierra, Bolivia, where species attributed to L. bicolor and
L. braunsii Kohl (and perhaps a third species) occurred (Frank 1990). By this time, A.
S. Menke was revising Neotropical Larra. His identifications convinced Bennett and
Pruett that they were dealing with L. bicolor and L. praedatrix (Strand), on which

Scientific Notes

they published behavioral notes (Bennett & Pruett 1991, Pruett & Bennett 1991).
Live Larra were brought to Florida for release in Alachua County in 1988-1989. It is
likely that specimens of three species, L. bicolor, L. praedatrix, and L. godmani Cam-
eron (senior synonym of L. braunsii Kohl), were imported and released, but in un-
known proportions. Female L. praedatrix and L. bicolor are not distinguishable with
certainty (Menke 1992).
The release sites were as follows: Micanopy, about 10 km south of Gainesville (Oc-
tober 1988) where 175 2 wasps, 35 6 wasps, and 80 mole crickets bearing Larra eggs
were released; the North Florida Regional Medical Center, Gainesville (March and
May 1989) where 86 2 wasps, 24 6 wasps, and 29 mole crickets bearing Larra eggs
were released; and the University of Florida Honey Plant, Gainesville (June 1989)
where 13 2 wasps and 23 6 wasps were released. The Honey Plant site was the same
location where L. bicolor had been released by R. I. Sailer in 1981. This site contained
the original plot of Spermacoce verticillata planted by Sailer.
Spermacoce verticillata plants were brought from Miami to Gainesville in 1987 by
J. H. Frank and maintained in pots. These plants were established in 1991 at the
southeast corner of the new Entomology/Nematology building of the University, about
2 km northwest of the Honey Plant. In 1992, A. S. Menke published his major revision
of Neotropical Larra, allowing reliable identification for the first time. No Larra were
observed in Gainesville before the retirement ofF. D. Bennett in July 1993 and his de-
parture from Florida.
In October 1993, Entomology/Nematology Dept. technician J. A. Gillmore reported
observing a wasp attack a mole cricket outside the departmental building. Larra
adults were then observed feeding on flowers of Spermacoce verticillata in the plot at
the southeast corner of the building. Dissection and microscopic examination revealed
that these were L. bicolor (not L. analis or L. godmani or L. praedatrix). Microscopic
examination further revealed a dense punctation of the vertex of the head typical of
specimens from Santa Cruz, Bolivia, and not a sparse punctation typical of specimens
from Puerto Rico (Menke 1992). Larra adults with the same morphological character-
istics were found at the Honey Plant release site. We concluded that the wasps found
were progeny of those released by F. D. Bennett in 1988-1989, originating from Bo-
Wasps were seen at the Spermacoce plot at the Entomology/Nematology building
almost daily through the autumn of 1993. The last observation was on 9 December
1993, whereafter freezing temperatures occurred and the foliage was killed by frost.
The plants regenerated in spring 1994, and the first Larra bicolor was seen on 9 May
1994. Wasps were seen occasionally in the plot during subsequent weeks to Septem-
ber 1994, but at a lower density than in the autumn of 1993. Peak wasp abundance at
Ft. Lauderdale occurs in autumn (Castner 1988a), therefore, relatively low numbers
in spring and summer are not surprising. In August 1994, a Larra bicolor female was
observed by P. G. Koehler 2 km northeast of the Entomology/Nematology building at
the University of Florida track and field complex. Adult wasps have been observed at
the Entomology/Nematology building from October 1993 to September 1994. The pop-
ulation has spread a distance of at least 4 km, which is greater than ever observed at
Ft. Lauderdale.
Life histories and behavior of Larra analis and of Larra bicolor are described by
Smith (1935) and Castner (1988b), respectively. Under laboratory conditions Larra bi-
color will sometimes attack Neocurtilla hexadactyla but is thwarted by the defensive
secretion of this mole cricket (Castner 1984) or the inability of the larvae to develop
on the host (Pruett & Bennett 1991). Larra analis has not been found to infect Scap-
teriscus mole crickets under field conditions. The native wasp is specialized to the na-

Florida Entomologist 78(4)

tive mole cricket, and the introduced wasp is specialized to immigrant (pest) mole
crickets of the genus Scapteriscus.
Larra bicolor joins Steinernema scapterisci Nguyen & Smart (Rhabditida: Stein-
ernematidae) and Ormia depleta (Wiedemann) (Diptera: Tachinidae) as South Amer-
ican biological control agents established in Alachua County, Florida as natural
enemies of Scapteriscus mole crickets.
We are grateful to L. Nong for help in handling and rearing Larra wasps, inducing
them to parasitize mole crickets in the laboratory, and releasing them in 1988-1989.
We also thank C. J. H. Pruett and his colleagues at CIMCA, Santa Cruz, Bolivia, for
help in collecting Larra, and to D. H. Habeck and T. J. Walker for critical comments
on an earlier version of this text. This is Florida Agricultural Experiment Station
journal series no. R-04082.


Larra bicolor, a biological control agent of Scapteriscus mole crickets, has estab-
lished a population in Gainesville, northern Florida. This population results from
specimens collected in Santa Cruz de la Sierra, Bolivia, and released by F. D. Bennett
in 1989. Previously, a Florida population of this wasp had been established only at Ft.
Lauderdale in southern Florida; it resulted from releases made in the early 1980s by
R. I. Sailer. The proximal origin of the Ft. Lauderdale population of L. bicolor is Pu-
erto Rico, but its initial home of origin is Belem, Para, Brazil.


BENNETT, F. D., AND C. J. PRUETT. 1991. Observations on copulation in Florida and
on the behavior of male and female wasps of the genus Larra in Santa Cruz, Bo-
livia. Sphecos 21: 16-17.
BOHART, R. M., AND A. S. MENKE. 1976. Sphecid wasps of the world. A generic revi-
sion. Univ. California Press; Berkeley.
CASTNER, J. L. 1984. Suitability of Scapteriscus spp. mole crickets as hosts of Larra
bicolor (Hymenoptera: Sphecidae). Entomophaga 29: 323-329.
CASTNER, J. L. 1988a. Evaluation of Larra bicolor as a biological control agent of mole
crickets. PhD dissertation, Univ. Florida.
CASTNER, J. L. 1988b. Biology of the mole cricket parasitoid Larra bicolor (Hy-
menoptera: Sphecidae), pp. 423-432 in V. K. Gupta [ed.], Advances in Parasitic
Hymenoptera Research. Brill; Leiden.
FRANK, J. H. 1990. Mole crickets and other arthropod pests of turf and pastures, pp.
131-139 in D. H. Habeck, F. D. Bennett, and J. H. Frank [eds.], Classical Bio-
logical Control in the Southern United States. Southern Coop. Series Bull. 355:
1-viii, 1-197.
FRANK, J. H. 1994. Inoculative biological control of mole crickets, pp. 467-475 in A. R.
Leslie [ed.], Integrated Pest Management for Turf and Ornamentals. Lewis
Publishers; Boca Raton.
MENKE, A. S. 1992. Mole cricket hunters of the genus Larra in the New World (Hy-
menoptera: Sphecidae, Larrinae). J. Hym. Res. 1: 175-234.
PRUETT, C. J., AND F. D. BENNETT. 1991. Behavior of two species of Larra in Santa
Cruz, Bolivia. Sphecos 21: 15-16.
SAILER, R. I. 1985. Natural enemies, pp. 23-32 in T. J. Walker [ed.], Mole crickets in
Florida. Florida Agric. Exp. Stn. Bull. 846 (1984), 54 p.
SMITH, C. E. 1935. Larra analis Fabricius, a parasite of the mole cricket Gryllotalpa
hexadactyla Perty. Proc. Entomol. Soc. Washington 37: 65-82.
WILLIAMS, F. X. 1928. Studies in tropical wasps their hosts and associates (with de-
scriptions of new species). Hawaii. Sug. Plrs' Assoc. Exp. Stn. Entomol. Ser.
Bull. 19: 1-179.

December, 1995

Scientific Notes 623

WOLCOTT, G. N. 1938. The introduction into Puerto Rico of Larra americana Saus-
sure, a specific parasite of the changea" or Puerto Rican mole cricket, Scap-
teriscus vicinus [sic]. J. Agric. Univ. Puerto Rico 22: 193-218.
WOLCOTT, G. N. 1941. The establishment in Puerto Rico of Larra americana Saus-
sure. J. Econ. Entomol. 34: 53-56.

Scientific Notes


Ft. Lauderdale Research and Education Center
University of Florida, Institute of Food & Agricultural Sciences
3205 College Avenue, Ft. Lauderdale, FL 33314

On 3 January 1995, we were asked to identify termites collected on 30 December
1994 from a house in Miami. The sample, a termite-infested board and attached nest
material (carton) contained thousands of workers and hundreds of soldiers and young
brachypterous nymphs. We were struck by their small size and recognized the soldiers
as Heterotermes which are characterized by their slender and straight mandibles in
contrast to the relatively thick, curved mandibles of the native Reticulitermes spp.
[Fig. 1; also see Mathews (1977) for character diagnosis of Heterotermes]. The speci-
mens more or less fit the brief description ofH. convexinotatus (Snyder 1924), a widely
reported northern Neotropical species (Araujo 1977). However, the taxonomic status
of Heterotermes in this region is vague and in need of revision, therefore, we have not
as yet assigned a species name to this find.
On 12 January 1995, we inspected the infested property, a small, older single-fam-
ily house located 0.6 km east of Interstate Highway 95 and 0.3 km north of Interstate
Highway 195 in the "Little Haiti" district of Miami. The Heterotermes infestation was
centered in a room addition which was under construction on the north side of the
house. The addition consisted of concrete-block walls opening without a ceiling to a
covered wood-truss roof. We observed extensive drywood termite damage to the roof
rafters of the original house. The floor area of the addition was unfinished, consisting
of fill sand on bare soil. The sand completely or partially buried numerous plywood
and solid lumber scraps. Additional scraps were scattered or stacked near the former
north wall of the original house. Although conditions were dry, nearly all wood scraps
were under some degree of Heterotermes attack. Foraging tubes criss-crossed the sur-
faces of some of the wood. Fist-size pieces of carton were attached to the older dam-
aged wood. No foraging tubes were observed on any of the structure itself. The
condition of the infestation and large numbers of brachypterous nymphs indicated
that it had been active for years and had likely undergone one or more annual repro-
ductive dispersal cycles.
Our suspicion that additional colonies were established in the neighborhood were
confirmed on 30 March 1995 when several soldiers and workers were collected by a
pest control operator from a warehouse located about 300 m ENE from the original
site. Unlike the original site, the warehouse showed many signs of above-ground for-

Florida Entomologist 78(4)

0 0

i u

Figure 1. Soldiers of Coptotermes formosanus Shiraki (left), Heterotermes sp. (cen-
ter), and Reticulitermes flavipes (Kollar) (right), all known from Dade County, Florida.

aging activity and damage to the structure itself. As in the original site, however, none
of the activity was associated with wet conditions or moisture sources. Foraging tubes,
originating from the foundation slab, extended several meters up concrete walls. The
small size of the Heterotermes foragers may have contributed to their ability to pene-
trate narrow cracks, fissures, or joints in the building's foundation. Destruction of
wood was observed in framing, molding, and doors. Indoor workings of Heterotermes
were spread over a 50 m distance suggesting an infestation by either one massive or
several smaller colonies. The only damage observed outside the building was to a
wooden exterior door. As of this writing, dispersal flights have not been observed, nor
alates collected from either site. Based on our observations of Heterotermes in the
West Indies, fully formed alates do not appear in colonies until the beginning of the
rainy season (May-June).
This second infestation is important because it confirms that Heterotermes is well-
established and thriving in a relatively large urban location and demonstrates the de-
structive potential of this species. The infestation also underscores the significance of
Heterotermes' small size and ability to forage under dry conditions allowing it to oc-
cupy a niche currently not exploited by other subterranean termite species in Florida.
Our long-term expectations are that Heterotermes will expand its distribution well be-
yond current bounds and rival the pest status of preexisting subterranean termite
species in Dade and adjacent counties of subtropical Florida.
This is the first known successful exotic introduction of Heterotermes into the
United States. These termites almost certainly originated from the West Indies or the
Neotropical mainland. The only U.S. species, the native H. aureus (Snyder), is found

December, 1995

Scientific Notes

bn a -t4 1 0
iE il b ^ a - i
no C |
g | |
o a -a P, cc A
I sg o | g i i s
- ^ C - ^ A |

*i ^ I I ^ |t4 -C ^ ^-^ 1-
-C C Cn

Sl bl |4 -i C)
v a *-bf- c

-C bn A
I ~' i lj i^lir^l
L r 'V cc C2

1-4 C~l C6 ) L6 10 1^ 00 C) C
A- A-) c^^~- c ~ )
y Q *o P 'B i ? o '
C bl bfl ' di b^l C
C)C h -fl fl'

0 bn

m a ba n bn 'a ~C
g, -'4,
^ ^ a, sli g
ccU .' < h~ V t flca~.! Stf l'
So o 'i e t to e N ~ c

k C) i C) bnj bf g

P4C C) .
It 'V | |. 1
'V bfl +n

^ t o^ S^^ S Sr
C) ~Lc c~
U ~ .- E a, bfo 'ca~a c

pq "o-O '
^~ | | I. i, *l ^l

pq c lc 6 L 6 11 0 *i -
a4n a
m EbnEgMi;m*Y
^ I~~~a .if" i Ii F1 2e^
s ~ ~ ~ b a~c M a < aa a^. as ,CT
C) r C ) g c
'V C i i c C
en Ma il he;C Q- 3
^ ~~ ^ " *
Fl a; .-g g F r
Io Ifif ~ ill

) C) -s t i) C
1 g a gC6 L6 *a
< ~ ~~ ~~ 'VCa B "iao^
c~ caA

Z/J -i C +C +C cc.3 -j .--

H ~ ~ ~ ~ ~ ~ c *^ CCj*- c u ^u^
& a, t, to g m N.K '
s 'V" ab |-4 '

b S -- 3 -S

w r- ci co i< edcrio
i t

F~ a
F C'Vh
C~ C) P C
cfl=r ccfl+ c&E
ill ~ ~ to~Fu

Florida Entomologist 78(4)

in the Sonoran and Colorado deserts of Arizona, southern California, and adjacent ar-
eas of northern Mexico (Snyder 1954). Soldiers and workers of H. aureus are propor-
tionally larger than those from the Miami discovery and are not conspecific. Like
Reticulitermes, Heterotermes spp. are generally serious structural pests where they
are found. In Table 1 are listed general similarities and differences between New
World species of these two genera.
From a worldwide perspective, Heterotermes Froggatt is primarily a tropical genus
(Emerson 1971) with the exception of several species in southern Australia (Hill 1942)
and the aforementioned H. aureus. Eight species are described from throughout the
Neotropical region, including three from the West Indies (Araujo 1977). Recently, ex-
tensive collecting in the West Indies has cast some doubt on the validity of one or two
of the Heterotermes species which were described from there (Scheffrahn et al. 1994).
Although occurring in drier habitats in the northern Neotropics, Heterotermes spp.
are found in a wide range of pantropical habitats. Emerson (1971) suggests that indi-
vidual species of Heterotermes are confined to their respective climatic zones by the
limits of soil moisture and temperature.
It is unclear why West Indian Heterotermes spp. have not previously become estab-
lished in southern Florida or the Florida Keys, or why Reticulitermes spp. do not occur
in the West Indies or, at least, on nearby offshore islands of the Bahamas. Since cli-
matic differences between these land groups are minimal, we suspect that allopatry
has been maintained because these genera are poor candidates for introduction across
ocean barriers by natural means. Introduction of Heteotermes to new localities by hu-
man activity is rare. Gay (1967) reports only two known cases of established human-
aided introductions of Heterotermes; H. perfidus (Silvestri) from unknown origin to St.
Helena about 1840, and H. philippinensis (Light) from the Philippines to Madagascar
and Mauritius early in the 1900s. Emerson (1971) speculates that H. convexinotatus
was introduced to the Galapagos Islands by man.
In light of their broad distribution and structure-infesting potential, Heterotermes,
along with Coptotermes, represent the two principle rhinotermitid subterranean pest
genera of the tropical world. In addition to Heterotermes, three other pestiferous ter-
mite species have been introduced into Florida. These include: Coptotermes formosa-
nus Shiraki (Fig. 1), the Formosan subterranean termite, which has saturated or
appeared in numerous urban and suburban sites throughout the state; and two kalo-
termitid species, Cryptotermes brevis (Walker), the West-Indian powderpost drywood
termite, a widely distributed species; and Incisitermes minor (Hagen), the western
drywood termite, which is occasionally encountered in Florida (Scheffrahn & Su
We acknowledge the alert attention of Norm Sage and Burt Silver (Inspection
Group, Pompano Beach) who first collected Heterotermes in Miami, Tom Lostetter (Ac-
tive Pest Control, North Miami) who collected Heterotermes at the second site, and
Mike Petrozzino and Frank Valdes (Florida State Bureau of Entomology, Miami), who
inspected both sites. We thank Jan Krecek, Phil Busey, and Robin Giblin-Davis for
their critical reviews. Florida Agricultural Experiment Station Journal Series No. R-


Two well-established structural infestations of the subterranean termite genus
Heterotermes Froggatt were discovered in Miami, Florida, in 1995. This is the first
record of an exotic Heterotermes sp. in the United States and constitutes the fourth ex-
otic termite species living in Florida.

December, 1995

Scientific Notes 627


ARAUJO, R. L. 1977. Catalogo dos Isoptera do Novo Mundo. Acad. Brasileira de Cien-
cias, Rio de Janeiro, RJ. 92 pp.
EMERSON, A. E. 1971. Tertiary fossil species of the Rhinotermitidae (Isoptera), phy-
logeny of genera, and reciprocal phylogeny of associated Flagellata (Protozoa)
and the Staphylinidae (Coleoptera). Bull. American Mus. Nat. Hist. 146 (3):
GAY, F. J. 1967. A world review of introduced species of termites. CSIRO Melbourne,
Australia, Bull. 286:1-88.
HILL, G. F. 1942. Termites (Isoptera) of the Australian Region. CSIRO Melbourne,
Australia. 479 Pp.
MATHEWS, A. G. A. 1977. Studies on termites from the Mato Grosso State, Brazil. 267
pp. Academia Bras. de Ciencias, Rio de Janeiro.
SCHEFFRAHN, R. H., AND N.-Y. Su. 1994. Keys to soldier and winged adult termites
(Isoptera) of Florida. Florida Entomol. 77: 460-474.
1994. Termites (Isoptera: Kalotermitidae, Rhinotermitidae, Termitidae) of the
West Indies. Sociobiology 24: 213-238.
SNYDER, T. E. 1924. Descriptions of new species and hitherto unknown castes of ter-
mites from America and Hawaii. Proc. U.S. Natl. Mus. 64: 1-40.
SNYDER, T. E. 1954. Order Isoptera. The termites of the United States and Canada.
Natl. Pest Control Assn., New York. 64 Pp.


Scientific Notes


Rutgers Research Center, 121 Northville Road, Bridgeton, NJ 08302

'Atlantic County Extension Office, 1200 W. Harding Highway,
Mays Landing, NJ 08330-1533

The carrot weevil, Listronotus oregonensis (LeConte), is an important pest of car-
rots, parsley and celery in the northeastern United States (Simonet & Davenport
1981) and can also be a pest of parsnips (Ryser 1975). Adults overwinter in or near
fields where carrots or celery were grown the previous year, emerging in late April to
early May. They feed directly on the roots, and females oviposit from the beginning of
May until late June in carrot and parsley roots. Larvae tunnel extensively throughout
the upper third of the roots. Carrot weevils may damage up to 40% of the crop in un-
treated fields (Boivin 1985). Pepper (1942) reported two full broods with a partial
third in northern New Jersey, and three full broods with a partial fourth in southern
New Jersey.
Until recently, growers generally used in-furrow granular insecticides or multiple
soil-directed sprays of insecticides, such as azinphosmethyl, for control of carrot wee-
vils. These materials killed the adults and larvae before they tunnelled into the roots.
However, azinphosmethyl and parathion registrations on carrots were discontinued,
and the management of carrot weevils is now obtained by repeated foliar applications
of the pyrethroid esfenvalerate directed at the overwintered adults before they ovi-
posit. However, the pyrethroids have a short residual period when exposed to the en-

628 Florida Entomologist 78(4) December, 1995

10.2cm (4")-



here I

h r '0.7cm slots spaced
51 0.7cm apart (1/4")
5.1cm diam.
hole (2")

Figure 1. Schematic diagram of holes and cuts necessary to make two modified
Boivin carrot weevil traps from one piece of pressure-treated post. English units are
in( ).

vironment, and an accurate population monitoring system is necessary to determine
adult activity. In New Jersey, it is recommended that growers begin insecticide sprays
when adult activity begins (Anonymous 1995). A monitoring system for the weevils,
then, must provide reliable and timely captures of adults.
The traps currently used to monitor the adult carrot weevil activity include Ma-
sonR jars fitted with a funnel and baited with carrots or carrot baby food (Ryser 1975),
raw carrots placed in the soil to observe oviposition scars (Stevenson 1981), and a trap
consisting of wooden plates with a carrot as bait (Boivin 1985). Of these, the latter was
more effective as a monitoring device for adult carrot weevils than either the MasonR
jar type or the carrot in the soil (Boivin 1985). The Boivin trap, however, consisted of
many small wooden plates, a wooden top and bottom, two long bolts with nuts, and
several dozen washers. Even after assembly, the trap consisted of three major parts:
the body, a top and a bottom plate.
In this note, we describe a modified Boivin trap constructed from a block of wood.
The design is inexpensive, easy to construct, durable, and can be made from a single
piece of pressure-treated post. In field comparisons, this trap was as effective as stan-
dard Boivin traps in catching carrot weevil adults in a parsley field in New Jersey.
Trap Construction. The materials needed to construct the trap are listed below.
All measurements are reported in both metric and english units because the lumber
supply companies of the United States sell their materials based on english units:
1 Pressure-treated pine post, 10.2 x 15.3 cm (4 x 6 in)
1 5.1 cm diam drill (2 in)
1 Table saw, 25.4 cm diam (10-in saw)
Cut the pressure-treated post into 22.9-cm (9 in) long sections. Drill a 5.1-cm diam
hole through the center of each post section (Fig. 1). With the table saw, cut the post
section lengthwise through the center to yield two 10.2 x 7.6 cm (4 x 3 in) pieces (Fig.
1). Then cut 0.7 cm (1/4 in) grooves 4 mm apart and 5 cm deep from the face of the post
that was cut down the center (the face with the semicircular cut) for the length of the
block (Fig. 2). Although not necessary for trap function, a bottom plate of 0.7 mm thick

Scientific Notes

Figure 2. Completed trap, with a bottom plate.

plywood can be made for each trap to allow the trap to sit flat in the soil (Fig. 2) and
to provide a surface to bang the trap against to shake free the weevils hiding within
the slats.
Trap Effectiveness. Traps were compared for trapping effectiveness in a parsley
field in Buena, NJ from 31 May through 30 June 1994. A total of three Boivin (1985)
traps and two modified Boivin traps were randomly placed approximately 3 m apart
within the edge rows of the parsley field. The edge rows bordered a 1.5-m wide row of
privet (Ligustrum ovalifolium L.), commonly used as a windbreak in vegetable fields
throughout NJ. The weevils overwinter in hedgerows, such as privet (Ryser 1975),
and migrate into the field from these hedgerows. A fresh, whole carrot was placed in
each trap when set out. The traps were emptied twice weekly by banging the trap on
a wood board to shake loose the weevils. Adult carrot weevils were counted, collected,
and removed from the field, and the old carrot was replaced with a fresh carrot in each
trap. Traps in the field were maintained until 30 June 1995.
The standard Boivin trap caught an average of 6.3 weevils per trap, and the mod-
ified Boivin trap caught an average of 11.0 weevils per trap (Table 1) during the 4 wk

NJ 1994.

31 May- Total per
Trap Type 7 Jun 8-13 Jun 14-21 Jun 22-28 Jun Trap

Boivin 0.9 2.5 0.3 2.6 6.3
Modified Boivin 1.0 4.5 0.0 5.5 11.0

630 Florida Entomologist 78(4) December, 1995

We thank D. Collins for monitoring and identifying the weevils from the traps.
This work was funded by the Campbell's Soup Company, the New Jersey Agricultural
Experiment Station publication No. D-08130-10-95, and by the United States Hatch


An inexpensive and simple method is described for constructing a modified Boivin
(1985) carrot weevil trap. It is constructed from a single piece of pressure-treated post
22.9 cm in length. The trap is as effective as the Boivin trap in catching adult carrot


ANONYMOUS. 1994. Commercial vegetable production recommendations for New Jer-
sey. Rutgers Coop. Ext. Bull. EO-001J. 144 pp.
BOIVIN, G. 1985. Evaluation of monitoring techniques for the carrot weevil, Listrono-
tus oregonensis (Coleoptera: curculionidae). Canadian Entomol. 117: 927-933.
PEPPER, B. B. 1942. The carrot weevil, Listronotus latiusculus (Bohe.) in New Jersey
and its control. New Jersey Agric. Expt. Sta. Bull. 693. 20 pp.
RYSER, B. W. 1975. Investigations regarding the biology and control of the carrot wee-
vil, Listronotus oregonesis (LeConte) in New Jersey. M.Sc. thesis, Rutgers
Univ., New Brunswick, NJ.
SIMONET, D. E., AND B. L. DAVENPORT. 1981. Temperature requirements for develop-
ment and oviposition of the carrot weevil. Ann. Entomol. Soc. America. 74: 312-
STEVENSON, A. B. 1981. Carrot insects. Min. Agric. Food, Ontario, Fact Sheet 81-007.
Agdex 258/605. 4 pp.

Book Review


GOLDSMITH, M. R. AND A. S. WILKINS. (eds.) Molecular Model Systems in the Lep-
idoptera. Cambridge University Press, New York, xii + 542 p. ISBN 0-521-40249-2.
Hardback. $125.00.

Yes, molecular genetics research is conducted on insects other than Drosophila
melanogaster! The stated aim of this book is to provide readers with a review of mo-
lecular research in Lepidoptera and to convince readers that Lepidoptera can serve as
important model systems. The chapters cover topics as diverse as silkworm genetics;
transposable elements of Lepidoptera; lepidopteran molecular phylogeny, embryogen-
esis, and development; chorion gene regulation and evolution; silk protein gene regu-
lation in the silk gland; hormone action on the central nervous system; the molecular
genetics of moth olfaction and the immune response; and, the use of baculoviruses for
insect pest control. This book contains a wonderfully rich body of fundamental infor-
mation and provides an entire to an extensive literature. As such, it is a welcome ad-
dition to the growing list of books providing information on insect molecular genetics
and molecular biology.
The editors and authors are experts and have produced a well-written and illus-
trated volume of value to entomologists looking for molecular arthropod models other
than the ubiquitous fruit fly, Drosophila melanogaster. Ninety-three pages of refer-
ences provide access to much of the relevant literature. This book is an excellent re-
view of lepidopteran molecular genetics and could serve as supplementary reading in
courses on insect molecular genetics. It was not intended, and is not suitable, for an
introductory text on insect molecular genetics. This book reminds us that D. melano-
gaster is a very specialized insect, and a full understanding of insect genetics and evo-
lution requires comparative studies using other species.
The Lepidoptera contain species of great economic importance and esthetic value,
as well as providing species sufficiently large to be particularly amenable to physio-
logical, behavioral, genetic, and ecological studies. Using Bombyx mori, Ephestia,
Manduca sexta, Antheraea pernyi, and Hyalophora cecropia, fundamental advances
have been made in insect genetics, endocrinology, and biochemistry. The 16 chapters
provide: a history of Lepidoptera as model systems (J. H. Willis, A. S. Wilkins and M.
R. Goldsmith), an overview of silkworm genetics (M. R. Goldsmith), a review of mobile
elements of Lepidoptera (T. H. Eickbush), a review of phylogeny and comparative de-
velopment (J. C. Regier, T. Friedlander, R. F. Leclerc, C. Mitter and B. M. Wiegmann),
a summary of embryogenesis and experimental embryology (L. M. Nagy), a discussion
of homeotic genes in Bombyx development (K. Ueno, T. Nagata and Y. Suzuki), an
overview of structure, function, and regulation of chorion genes (F. C. Kafatos, G.
Tzertzinis, N. A. Spoerel and H. T. Nguyen), molecular models of chorion gene evolu-
tion (T. H. Eickbush and J. A. Izzo), a review of silk protein gene regulation and ho-
meobox genes in silk gland development (C. Hui and Y. Suzuki), an analysis of the
control of transcription of B. mori RNA polymerase III (K. U. Sprague), a review of
hormonal regulation of gene expression during development (L. M. Riddiford), a re-
view of the impact of hormones on the central nervous system (J. W. Truman), an over-
view of the molecular genetics of moth olfaction (R. G. Vogt), an analysis of the
molecular biology of the immune response (A. B. Mulnix and P. E. Dunn), a discussion
of engineered baculoviruses as tools for understanding development and physiology
and as potential agents for pest control (K. Iatrou), and an epilogue containing a sum-
mary of the unresolved issues and prospects for Lepidoptera as model systems (A. S.
Wilkins and M. R. Goldsmith).

Florida Entomologist 78(4)

Wilkins and Goldsmith point out that the Lepidoptera have several nearly unique
features and phenomena that make them novel and intrinsically interesting (e. g.,
elaborate wing patterns; silk production with its specialized translational apparatus;
and pheromone production and response, with the possibility of integrating behavior,
molecular, neurological and evolutionary aspects of moth pheromone utilization).
Furthermore, certain lepidopteran species are exceptional models for more general
phenomena, especially hormonal changes associated with metamorphosis and the
construction of the insect chorion. Hemolymph and cuticular proteins, and endocrinol-
ogy can be studied easily in Lepidoptera because the model insects are relatively large
and comparatively slow in their developmental rate, thereby facilitating experimen-
The Lepidoptera are also important in comparative analyses of important devel-
opmental processes. B. mori has been shown to have an unusual development which
is distinctly different from the classic long germ band mode of development found in
D. melanogaster. The homeotic genes and segmentation of B. mori are organized and
expressed differently, and their chorion genes are regulated differently. Thus, the Lep-
idoptera are valuable in comparative studies of insect genome structure and evolu-
tion. Novel families of proteins are involved in immune responses in the Lepidoptera,
and these may even have relevance to understanding vertebrate immune systems.
The editors conclude that the limitations to further progress in molecular biology of
the Lepidoptera include the lack of conventional genetics and molecular maps, as well
as a reliable and efficient genetic transformation system for genetic manipulation.
This book should stimulate a new generation of molecular entomologists to con-
sider the virtues and limitations of lepidopteran species as model systems for analysis
of the molecular biology and genetics of insects.

Marjorie A. Hoy
Department of Entomology and Nematology
University of Florida, Gainesville 32611

December, 1995



L. S. Bauer--Resistance:A Threat to the Insecticidal Crystal Proteins of Bacillus thu-

Change first sentence in conclusions (p. 434) to read: The genetic capacity of insect
populations to evolve resistance to Bt 5-endotoxins is now well documented in many
species within three different insect orders.

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

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