Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00054
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1996
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: VID00054
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

Hoddle & Van Driesche: Bemisia control with Encarsia formosa 1


Dept. of Entomology, University of Massachusetts, Amherst, MA, 01003-2410


Weekly releases of the parasitoid Encarsia formosa Gahan failed to control a low
density population (initially, 0.51 nymphs and pupae per plant) of the whitefly Bemi-
sia argentifolii Bellows & Perring on greenhouse grown poinsettia plants in Massa
chusetts when released at the rate of 4-7 adult females per plant. A lifetable
constructed for uncaged B. argentifolii in the presence of E. formosa indicated that
survivorship from the first/second instar to adult emergence was 14%. In contrast, in
a lifetable constructed for B. argentifolii on caged poinsettia from which E. formosa
was excluded, survivorship was 67%. Release of E formosa reduced the number of in
secticide applications on poinsettia by 75%, but the cost of using E. formosa (on a per
m2 basis) was 9.5 times that of insecticides alone.

Key Words: Augmentation, biological control, integrated pest management, green
house, whitefly, parasitoid.


Las liberaciones semanales de 4-7 hembras adults del parasitoide Encarsia for
mosa Gahan por plant no pudieron controlar una baja densidad poblacional (inicia
lente, 0.51 ninfas y pupas por plant) de la mosca blanca Bemisia argentifolii Bellows
& Perring en plants de flor de pascua en Massachusetts. Una tabla de vida cons
truida para B. argentifolii no mantenida enjaulas y en presencia de E. fomosa indic6
que la sobrevivencia desde el primero/segundo instar hasta la emergencia del adulto
fue del 14%. En contrast, en una tabla de vida construida para B. argentifolii sobre
plants de flor de pascua mantenidas enjaulas de las cuales E. formosa fue excluida,
la supervivencia fue del 67%. La liberaci6n de E. formosa redujo el numero de aplica
ciones de insecticide en las plants de flor de pascua en un 75%, pero el costo del uso
de E. formosa (por metro cuadrado) fue 9.5 veces el del insecticide solo.

The primary phytophagous pest affecting poinsettia (Euphorbia pulcherrima
Willd. ex Klotzsch) is the silverleaf whitefly, Bemisia argentifolii Bellows & Perring [=
the 'B' strain of Bemisia tabaci (Gennadius)] (Homoptera: Aleyrodidae) (Perring et al.
1993, Bellows etal. 1994). In Massachusetts, 200 greenhouses produce approximately
one million individually potted poinsettias annually, which have a wholesale value of
$5 million. Poinsettia is the highest ranked ornamental sold in the last quarter of the
year in this state [unpublished University of Massachusetts Integrated Pest Manage
ment Program Annual Report (1994)]. Because poinsettia is grown for its aesthetic
qualities, growers have an extremely low tolerance for the presence of whitefly
nymphs, adults, or honeydew Insecticides are often applied on a calendar schedule,
with applications being made every 3-5 days to reduce B. argentifolii populations to

Florida Entomologist 79(1)

acceptably low levels. Adverse effects of such intensive pesticide use against this pest
have been documented (Parrella et al. 1992; Heinz & Parrella 1994 a,b).
The University of Massachusetts Cooperative Extension System for floricultural
crops has initiated an integrated pest management program (IPM) to design a more
effective management program for B. argentifolii on poinsettia. In the first phase of
this program, scouts were employed to recommend pesticide application when suscep
tible whitefly stages exceeded tolerable densities. Targeted spraying in this manner
has reduced insecticide use by 18-50% with most poinsettia growers in the program
[unpublished University of Massachusetts Integrated Pest Management Program
Annual Report (1994)]. The second objective of the IPM program is to reduce insecti
cide use even further with biological control agents (in particular parasitic wasps) for
suppression of B. argentifolii.
One of the parasitoids that has been considered for use in the IPM program is En-
carsia formosa Gahan. Encarsia formosa is a commercially available parasitoid that
is used to control the greenhouse whitefly, Trialeurodes vaporariorum (Westwood), a
serious pest of greenhouse vegetable crops. Encarsia formosa is used worldwide to
control T vaporariorum on vegetable crops in greenhouses (van Lenteren & Woets
1988). Encarsia formosa tested under laboratory conditions with B. tabaci (strain B)
on poinsettia developed more slowly, exhibited higher mortality with reduced longev
ity, and was less fecund than parasitoids that developed on T vaporariorum (Boisclair
etal. 1990, Szabo etal. 1993).
Several studies on the use of E. formosa to control B. tabaci on poinsettia in green
houses suggest this parasitoid is effective, contrary to the laboratory findings [the
strain was not identified in these studies and others, but is assumed to be strain B
which became problematic on poinsettia in Europe after 1987 (see Boisclair et al.
1990, Szabo et al. 1993)]. Investigations by Benuzzi et al. (1990) in Italy, Albert &
Schneller (1989) and Albert & Sautter (1989) in Germany, and Stenseth (1993) in Nor
way concluded that E. formosa successfully suppresses B. tabaci (unidentified strain)
on poinsettia grown for the Christmas market when T vaporariorum is present. Work
by Parrella et al. (1991) in California, USA, reports that E. formosa is an ineffective
control agent for populations of B. argentifolii on poinsettia plants grown for cutting
production in the spring. Because poinsettia growing conditions in the northeastern
USA in the fall are more similar to those in Europe in the fall than to spring growing
conditions in California, there was a need to evaluate the ability of E. formosa to con
trol B. argentifolii populations on poinsettia in Massachusetts. To measure the
efficacy of E. formosa, we constructed lifetables for B. argentifolii in both the presence
and absence of the parasitoid.


Greenhouses and Cultivars

Two greenhouses at one commercial poinsettia producer in Massachusetts were
monitored to determine whether insecticides or E. formosa provided better control of
B. argentifolii. One greenhouse received E. formosa as a control measure and is des
ignated here as the biological control greenhouse. The second greenhouse was man
aged using synthetic pesticides and is designated as the insecticide greenhouse. Each
was a 170 m2 A' frame greenhouse (glass construction) with three benches in Cam
bridge, Massachusetts. The two side benches (1 m x 27 m) each held 156 pots (18 cm
diam) with 5 single-stem unpinched plants per pot. The middle bench (2 m x 26 m)
held 264 pots (19 cm diam) with 6 single-stem unpinched plants per pot for a total of

March, 1996

Hoddle & Van Driesche: Bemisia control with Encarsia formosa 3

576 pots and 3144 plants per greenhouse. The poinsettia cultivars were white and
marble Angelika'; red, pink and white 'Celebrate 2; and pink 'Gutbier V 14'. The
study started immediately after both greenhouses were filled with potted cuttings in
August, 1994. Some plants were removed during the test from both houses to satisfy
spacing requirements.

Population Density Estimation and Lifetable Construction

To estimate whitefly population densities, six leaves (2 from the bottom of the
plant, 2 middle, and 2 top) of 30 plants in each greenhouse were inspected weekly for
B. argentifolii. The number of eggs, first and second instars, third instars, fourth in
stars, red eyed pupae, and adults were recorded. T vaporariorum was not observed in
either greenhouse.
Three treatments were established in the biological control greenhouse: uncaged
plants (Treatment 1), cages without E. formosa (Treatment 2), and cages with E. for
mosa (Treatment 3). Treatment 2 acted as the control, and Treatment 3 was a check
for a cage effect on whitefly development in the presence of the parasitoid. In addition
to estimating whitefly densities on randomly selected leaves in Treatment 1, the fate
of marked cohorts of nymphs was determined. Cohorts were established by tagging
and numbering naturally-infested plants bearing first or second instar B. argentifolii.
Numbers were written on tagged leaves with an indelible marker beside young
nymphs. Numbered nymphs were examined weekly, and their developmental stage
recorded. Young nymphs (approximately 1-30 nymphs per leaf) found on 1 3 leaves of
each of 3-5 plants were recruited every week for the lifetable study. Observations were
continued until the nymphs had either died of unknown causes, disappeared, been
parasitized, or emerged as adult whiteflies. Parasitism was noted when the whitefly
pupa turned brown or a parasite had emerged. The recorded fates of all nymphs (204)
in Treatment 1 were used to create a partial lifetable for B. argentifolii.
For Treatments 2 and 3, nine pots (19 cm diam) with single-stem unpinched poin
settia plants (cultivars used; white Angelica,' red and pink'Celebrate 2') were selected
at random and enclosed by fine mesh bags (28 cm x 28 cm x 36 cm; mesh 6.2 x 6.2
threads per cm2). Each bag was supported by four 50 cm stakes that were driven into
the potting medium. A rubber band was used to seal the bottom of the bag against the
exterior of the pot. One male and one female adult B. argentifolii were released into
each bag. In Treatment 2, the resulting whitefly population was allowed to develop on
poinsettia in the absence of E. formosa. In Treatment 3, three E. formosa were intro
duced into each of the nine cages weekly.
For Treatments 2 and 3, whitefly population density estimates were made, and co
horts of nymphs established in the same manner as Treatment 1. Numbered nymphs
on tagged leaves were observed weekly for survivorship and parasitism. Mortality
data for numbered nymphs in Treatments 2 and 3 were used to create partial lifeta
bles for B. argentifolii populations on caged poinsettia plants in the presence and ab
sence of E. formosa.

Calculating Marginal Probabilities of Mortality and k-Values

Marginal attack rates were calculated to separate mortality from each observed
source (unknown death, disappearance, and parasitism). The marginal probability of
attack is the number of pests that would be attacked by an agent in the absence of all
other contemporaneous mortality agents. It is the net probability of dying (as opposed
to the crude probability of dying, which is the apparent mortality calculated from

Florida Entomologist 79(1)

numbers observed to die from a cause) (Royama 1981, Elkinton et al. 1992). The mar
ginal probability of attack was calculated for each factor (Table 1) as:

m, 1(1-d)d"d

where m,= marginal probability of attack from the ith cause, d,= death rate from the
ith cause and d= death rate from all causes combined (Elkinton et al. 1992).
Killing powers or k-values (the negative logarithm of the proportion surviving in
each stage) for each mortality factor were determined as:

k= -log,1(1 marginal probability of attack by the ith cause),

where k,= the k-value for the ith cause of mortality.

Wasp Releases, Percent Emergence and Emergence Pattern

Parasitoid releases in both the cages and open greenhouse began immediately af
ter the biological control greenhouse was filled with poinsettias. Bunting Biological
North America supplied release cards, each bearing 100 parasitized T vaporariorum
pupae. The number of cards put into the biological control greenhouse each week
ranged from 140 to 268. Cards were hung on strings stretched between the pots and
tied at the same height as the pot rims. In this position, wasps emerged below the fo
liage and were assumed to move upwards through the canopy searching for B. argen
tifolii nymphs.
Every week, all cards were removed from the biological control greenhouse before
new cards were put out. Ten cards were randomly selected from those recovered and
soaked in water and detergent for 30 min in the laboratory. Parasitized greenhouse
whitefly and exuviae were rubbed off the card with a size 2 insect pin, and the mean
number of parasitized greenhouse whitefly per card and the percent emergence of
wasps were determined for each weekly release. These values were used to calculate
the mean number of wasps released per plant per week. On two occasions, the emer
gence pattern of the parasitoid was determined by counting the number of wasps that
emerged from the cards each day in the laboratory.

Cost Analysis

The cost of biological control vs. the cost of insecticides was determined by analyze
ing insecticide application records for both the insecticide and biological control
greenhouses. The price of purchasing the required number of parasitoids was based
on an averaged estimate from suppliers of beneficial insects. Labor costs associated
with releasing parasitoids and applying insecticides were not included in the analy

Sales Inspection

At week 16 of the growing period, the finished plants were shipped to retailers. To
determine the final whitefly density, six leaves on 15 plants from both the biological
control and the insecticide greenhouses were inspected. The number of live nymphs
and pupae on each leaf was recorded.

March, 1996

Hoddle & Van Driesche: Bemisia control with Encarsia formosa 5

ato c'

o o C
CD -

o -

", CD


", o, c
H d



H ^


-c' m cn- ~
b~, c' c'

CC -C C ~ C

1 t -i t.~

Cm C Cm Cm

a)C a) C a) C a) cO

E^ LO cm


H a

O t-
CM 00





s ^i

0r S Cz]
r -0Z
I n^

.a &

s ^|
-5 ^

SX "2O
L s

-OD"^ :y=

*s "3 cq cqg

,, o i- ,, C 1

^" d <

Florida Entomologist 79(1)


Population Density Estimates and Lifetable Construction

Whitefly densities on caged poinsettia plants onto which one female whitefly had
been introduced at the start of the experiment (Treatment 2, Fig. 1A) increased
steadily, reaching 246 nymphs and pupae per plant by week 10. In contrast, whitefly
densities on caged plants inoculated with three adult E. formosa per week (Treatment
3) were substantially lower by week six (Fig. 1A) and averaged only 20 live nymphs
and pupae per plant in week ten, 8% of the recorded density on the control plants
(Treatment 2) (Fig. 1A).
Whitefly populations on uncaged plants in the biological control greenhouse re
mained below six nymphs and pupae per plant until week 7 of the experiment. The
number of immature whiteflies on these plants increased to approximately 39
nymphs and pupae per plant by week 10 (Fig. 1B). At that time parasitoid releases
were terminated, and two insecticide applications were made. In contrast, whitefly
densities in the insecticide greenhouse increased to 32 nymphs and pupae per plant
by week 4, but declined to low densities (< 5 nymphs and pupae per plant) by week six
(Fig. 1B). This low level of infestation was maintained in the insecticide greenhouse
through week 10 because of regular insecticide applications (see Fig. 1C).
The percentage of plants that were infested reached 100% in weeks 4 and 6 for the
insecticide greenhouse and the biological control greenhouse respectively. The per
centage of B. argentifolii-infested plants steadily declined in the insecticide green
house after week six. This trend was not observed in the biological control greenhouse
(Fig. 1C).
In the biological control greenhouse, 86% of the whitefly nymphs that were fol
lowed individually died prior to adult emergence (see footnote Table 1); however,
nymphal densities ultimately exceeded the grower's damage threshold for the crop. In
Treatment 2, in which a B. argentifolii population developed on caged poinsettia
plants in the absence of E. formosa, 33% of the nymphs died of natural causes (see
footnote Table 1). In Treatment 3, where three E. formosa per week were released into
identical cages, the level of nymphal mortality was 98% (see footnote Table 1).

Marginal Probabilities of Mortality and k-Values

Mortality from three factors parasitismm, unknown death, and disappearance) oc
curred contemporaneously. The marginal probability of attack and k-values for these
factors in Treatments 1 3 are presented in Table 1. Treatment 3 consistently exhibited
the highest levels of mortality for each of the immature lifestages (Table 1). The high
est observed k-values were those for pupae which exhibited high levels of parasitism
in Treatments 1 and 3 (Table 1).

Wasp Releases, Percent Emergence and Emergence Pattern

Number of parasitized pupae per card, percentage of wasps emerging, number of
release cards put into the greenhouse each week, number of wasps released per plant,
and number of wasps released per m2 are shown in Table 2. Two shipments of E. for
mosa exhibited different emergence patterns in the laboratory. Group one exhibited a
unimodal emergence pattern with wasp numbers peaking 5 days after receipt (mean
daily maximum temperature= 24.7 C + 0.7; mean daily minimum temperature= 23C
0.7) (Fig. 2A). Group two exhibited a bimodal emergence pattern, with wasp num

March, 1996

Hoddle & Van Driesche: Bemisia control with Encarsia formosa 7

Fig. 1A

--Treatment 3; cage with E.
50 formosa
)0 --Treatment 2; cage without
E. formosa


06 1& ,

1 2 3 4

Fig. 1B

--insecticide greenhouse

--Treatment 1; biological
control greenhouse

6 7 8 9 10


1 2 3 4 5 6 7 8 9 10

Fig. 1C

* *

100 -

40 / -0-Insecticide greenhouse
20 biological control
0 I

0 1 2 3 4 5 6

Weeks after potting

7 8 9 10

Figure 1. (A) The mean number of Bemisia argentifolii nymphs and pupae (
S.E.M.) on caged poinsettia plants in the absence (Treatment 2) and presence of E.
formosa (Treatment 3). (B) The mean number of Bemisia argentifolii nymphs and pu
pae ( S.E.M.) per plant in the insecticide house and the biological control house
(Treatment 1). (C) Percentage of plants infested with adult or immature stages of Be
misia argentifolii in the biological control and insecticide greenhouse; asterisk indi
cates dates of insecticide applications in the insecticide greenhouse.


Florida Entomologist 79(1)

March, 1996


< 7




CO (

^ K
g .

- Cj CM O' LO CD tN- X0

~' ~ o' C C C C Lf
~' ~ o' C C C C Lf
- - c cc c cc ~
mmcn c'jc'jc'jc'jc

C Cf Cf o CC C N
o o CJ C fl N -

i i, i C' Di iCJ N f

Ct 'J 'JLfl C~D L' CD Ll


biLf C Ct Cj C CJ D
Lf CflCt C Ct N
o "" o tC tCC





< S
0 3








Hoddle & Van Driesche: Bemisia control with Encarsia formosa 9

Fig. 2A

200 0 Group 1 E Group 2









1 2 3 4 5 6 7 8 9

Number of days after receipt of shipment

Figure 2. (A) Daily emergence of Encarsia formosa in the laboratory and cumula-
tive percent (B) emergence of Encarsia formosa in the laboratory.

bers peaking on days 2 and 5 after receipt (mean daily maximum temperature
24.4C + 0.7; mean daily minimum temperature= 23.3C + 0.3) (Fig. 2A). Over 97% of
E. formosa had emerged after 7 days (Fig. 2B). In the biological control greenhouse,

Florida Entomologist 79(1)

the mean daily maximum temperature was 22.8'C + 0.6; mean daily minimum tem
perature= 17C + 0.5.

Cost Analysis

The costs of controlling B. argentifolii with E. formosa or insecticides are pre
sented in Table 3. Weekly releases of 4-7 E. formosa per plant for nine weeks, followed
by two insecticide applications, were 9.5 times more expensive than using insecticides

Sales Inspection

At week 16 of the growing period, the numbers of live nymphs on plants from both
the biological control and insecticide greenhouse were low at the time of shipment.
The mean numbers of nymphs per leaf were 0.01 + 0.11 in the insecticide greenhouse
and 0.02 + 0.21 in the biological control greenhouse. There was no obvious difference
in foliage quality between the biological control and the insecticide greenhouses.


Release of high numbers of E. formosa (4-7 wasps per plant per week) did not suc
cessfully control a population of B. argentifolii on poinsettia, even though parasitoid
releases were initiated at the beginning of the growing period when the infestation of
B. argentifolii nymphs was very low (0.09 per leaf or 0.51 per plant).
The high mortality (98%) observed in Treatment 3 (caged plants with wasps) (Ta
ble 1) may have occurred because cages prevented wasps from abandoning plants,
thereby increasing residence and searching time. The differences between the ob
served k-values (Table 1) of Treatment 2 (caged plants with no wasps) and those of
Treatment 1 uncagedd plants) and Treatment 3 (caged plants with wasps), with re
spect to unknown death for all nymphal stages, may be due to aborted parasitism (in
older nymphs) or host feeding by E. formosa. In addition, some whitefly death ob
served in Treatment 3 may have resulted from superparasitism.
Another problem inherent with the use of cages to enclose single plants is the need
to introduce adult whiteflies. An introduction rate of just 1 female and 1 male per


Insecticide House Biological Control House

Total cost of sprays $268.47 $43.28
Total cost of E. formosa NA $2520.00
Total treatment cost $268.47 $2563.28
Treatment cost per plant $0.09 $1.02
Cost m2 $1.58 $15.08

'Insecticide costs are based on 1993 catalogue prices. The E. formosa price was based on a rate of $12.00 per
1000 wasps.
Insecticide costs per m2 in Massachusetts range from $0.32 -$2.26 in poinsettia crops [unpublished Univer
sity of Massachusetts Integrated Pest Management Program Annual Report (1994)].

March, 1996

Hoddle & Van Driesche: Bemisia control with Encarsia formosa 11

plant to establish an experimental population resulted in the caged plants having an
initial adult whitefly density nine times that of the biological control greenhouse; be
fore E. formosa was released, the mean number of adult whiteflies in the biological
control house was 0.22 + 0.45 adults per plant. Consequently, this may have exagger
ated the observed densities in the control cages (Treatment 2). However, this would
not affect comparisons between Treatments 2 and 3, as both were inoculated with
equal numbers of whiteflies.
The limited control provided by E. formosa was 9.5 times more expensive than in
secticides on a per m2 basis (Table 3). Albert & Sautter (1989) achieved cheaper con
trol of B. tabaci (the strain of whitefly was not identified) on poinsettia with E.
formosa than with chemicals, but T vaporariorum, a preferred host for E. formosa,
was present in the crop. This may have affected parasitoid population levels in the
At sale of the crop, whitefly densities in the chemical greenhouse and biological
control greenhouse were 0.01 + 0.11 and 0.02 0.21 nymphs per plant respectively.
Plants from both houses were marketed successfully. In the biological control green
house, insecticide use was reduced by 75% by release of wasps (from eight to two in
secticide applications), but pest control costs were increased from $0.09 to $1.02 per
Several agronomic practices associated with poinsettia production should favor bi
logical control of B. argentifolii in Massachusetts. First, poinsettias are grown from
June (if cuttings are being produced) until December and entire greenhouses are de
voted to poinsettia production. Monocultural production simplifies pest management
because B. argentifolii is the only arthropod causing foliar damage, and this nullifies
incompatible management programs for pest complexes (Heinz & Parrella 1994a,
Parrella et al. 1991).
Second, the majority of growers in Massachusetts purchase poinsettia cuttings in
July or August from suppliers who typically sell plants with very low densities of
adult and immature B. argentifolii. Therefore, initial B argentifolii densities are suf
ficiently low that a favorable ratio of parasitoids to whiteflies could be established.
Third, fungal diseases of poinsettia foliage can be controlled with fungicides that are
compatible with biological control agents (Parrella et al. 1991). Fourth, winters in the
northeastern USA prevent continual immigration of B. argentifolii into greenhouses
from outdoor host plants, and growers need only manage the whitefly population that
had established in the greenhouse before the onset of cold weather.
In view of these considerations, two major constraints to successful biological con
trol of B. argentifolii on commercially grown poinsettia in Massachusetts are: (1) the
commercial non-availability of an effective natural enemy for B. argentifolii, and (2)
lack of information as to which release strategies would maximize the impact of a
suitable control agent. A suitable release program should span the entire window of
pest susceptibility to ensure maximum mortality by host feeding and parasitism.
Variable release rates and timings may be necessary to achieve this objective as foli
age density, pest density, and levels of parasitism change over the season. Variable re
lease rates of a parasitoid may result in higher levels of parasitoid recycling through
reproduction. Natural reproduction in the greenhouse could augment weekly releases
and reduce the cost of parasitoid releases.
Complete reliance on biological control may not be feasible for this ornamental
crop, but incorporation of natural enemies in the context of an IPM program for poin
settia should be an attainable goal. Colonies of aphelinid parasitoids that attack B. ar
gentifolii exist at several research institutes in the United States. Further work in
greenhouses is needed to evaluate the efficacy of these parasitoids for B. argentifolii
control on poinsettia.

Florida Entomologist 79(1)


We thank J. Mason for her assistance in the field and E. Norberg for allowing this
trial to be run on his premises. The cooperation of Dr. R. Greatrex CIBA-Bunting Ltd,
France, and D. Cahn of Bunting Biological North America is gratefully acknowledged.
This research was supported by grants from the Massachusetts IPM Program, and
USDA/NRICGP grant #9402481.


ALBERT, R., AND H. SAUTTER. 1989. Parasitoids protect Christmas stars from white
flies. Deutscher Gartenbau 43: 1671-1673.
ALBERT, R., AND H. SCHNELLER. 1989. Successful biological control in ornamental
plants 1. Poinsettias (Euphorbia pulcherrima Willd. ex Klotzsch). Gesunde
Pflanzen 41: 389-395.
BELLOWS, T. S., T. M. PERRING, R. J. GILL, AND D. H. HEADRICK. 1994. Description of
a species of Bemisia (Homoptera: Aleyrodidae). Ann. Entomol. Soc. America.
87: 195-206.
BENUZZI, M., G. NICOLI, AND G. MANZAROLI. 1990. Biological control of Bemisia tabaci
(Genn.) and Trialeurodes vaporariorum (Westw) by Encarsia formosa Gahan
on poinsettia. SROP/WPRS Bull 13: 2731.
BOISCLAIR, J., G. J. BRUEREN, AND J.C. VAN LENTEREN. 1990. Can Bemisia tabaci be
controlled with Encarsia formosa? SROP/WPRS Bull. 13: 32-35.
Marginal attack rate, k-values and density dependence in the analysis of con
temporaneous mortality factors. Res. Pop. Ecol. 34: 29-44.
HEINZ, K. M., AND M. P. PARRELLA. 1994a. For silverleafwhiteflies on poinsettias ...
bios are in control. Greenhouse Grower 12: 32-39.
HEINZ, K. M., AND M. P. PARRELLA. 1994b. Poinsettia (Euphorbia pulcherrima Willd.
ex Koltz.) cultivar mediated differences in performance of five natural enemies
of Bemisia argentifolii Bellows and Perring, n. sp. (Homoptera: Aleyrodidae).
Biological Control 4: 304-318.
PARRELLA, M. P., T. D. PAINE, J. A. BETHKE, K. L. ROBB, AND J. HALL. 1991. Evalua
tion of Encarsia formosa for biological control of sweetpotato whitefly (Ho
moptera: Aleyrodidae) on poinsettia. Environ. Entomol. 20: 713-719.
Sweetpotato whitefly: prospects for biological control. California Agriculture
46: 25-26.
1993. Identification of a whitefly species by genomic and behavioral studies.
Science 259: 7477.
ROYAMA, T. 1981. Evaluation of mortality factors in insect lifetable analysis. Ecol.
Monogr. 5: 495-505.
SOUTHWOOD, T. R. E. 1978. Ecological methods with particular reference to the study
of insect populations. Chapman and Hall London. 524 pp.
STENSETH, C. 1993. Biological control of cotton whitefly Bemisia tabaci (Genn.) (Ho
moptera: Aleyrodidae) by Encarsia formosa (Hymenoptera: Aphelinidae) on
Euphorbia pulcherrima and Hypoestes phyllostachya. WPRS/SROP Bull. 16:
SZABO, P., J. C. VAN LENTEREN, AND P. W. T. HUISMAN. 1993. Development time, sur
vival and fecundity of Encarsia formosa on Bemisia tabaci and Trialeurodes va
porariorum. SROP/WPRS Bull. 16: 173-176.
VAN LENTEREN, J. C., AND J. WOETS. 1988. Biological and integrated control in green
houses. Ann. Rev. Entomol. 33: 239-269.

March, 1996

Castineiras et al.: Temperature response ofCeranisus menes 13


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


The development response to temperature of a Japanese uniparental strain and a
Thai biparental strain of Ceranisus menes (Walker) (Hymenoptera: Eulophidae) was
studied. The parasitoids were reared on first instar Thrips palmi larvae in incubators
at constant temperatures of 21, 23, 25, 27 and 29"C. Total developmental time de
creased with the increase of temperature from 35.1 to 21.9 days in females and from
33.4 to 18.8 days in males. Lowest mortality (12%) was recorded in both strains at
23'C and highest (95%) in the Japanese strain at 29C. Seventy-three percent of the
Thai parasitoids survived at 29C, but 39% mortality was observed at 21C. Percent
parasitism ranged from 23.8 to 28.9% at 25-29"C, but decreased to 11.5% at 210C. The
sex ratio (male:female) was not affected by temperature and averaged 1:1.9. A their
mal constant of 500 degree-days and a developmental threshold of 8C (from egg to
adult emergence) were obtained for both Japanese and Thai females. For the Thai
males, the thermal constant was 333.3 degree-days and the minimum threshold was

Key Words: Ceranisus menes, development, parasitism, mortality, Thrips palmi, bio
logical control.


Fue estudiada la respuesta a la temperature en una cepa uniparental japonesa y
en una cepa biparental tailandesa de Ceranisus menes (Walker) (Hymenoptera: Eulo
phidae), un parasitoide de Thrips palmi Karny (Thysanoptera: Thripidae). Los para
sitoides fueron criados en incubadoras a temperature constant de 21, 23, 25, 27 y
29"C. La duracidn del desarrollo disminuy6 con la temperature de 35.1 a 21.9 dias en
las hembras, y de 33.4 a 18.8 dias en los machos. La mas baja mortalidad (12%) fue
registrada en ambas cepas a 23C y la mas alta (95%) en la cepajaponesa a 29"C. El
73% de los parasitoides tailandeses sobrevivi6 a 29"C, pero a 21 C fue observado un
39% de mortalidad. El porcentaje de parasitismo estuvo en el rango de 23.8 a 28.9%,
pero baj6 hasta el 11.5% a 21C. La relacidn sexual (macho:hembra) no fue afectada
por la temperature y promedi6 1:1.9. Se obtuvo una constant t6rmica de 500 grades
dias y un umbral de desarrollo de 8C (desde el huevo hasta la emergencia del adulto)
en las hembras de ambas cepas. En los machos tailandeses la constant t6rmica fue
de 333.3 grados-dias y el umbral minimo de 13.7C.

Since its first detection in 1990, Thripspalmi Karny has become an important pest
of eggplant, potato, beans, peppers and cucurbits in South Florida (Seal et al. 1992).
The eulophid wasp Ceranisus menes (Walker) was selected as a possible biological
control agent for suppression of T palmi. In 1992-93, C. menes strains were intro
duced from Japan and Thailand into Dade County, Florida.

Florida Entomologist 79(1)

Ceranisus menes is an endoparasitoid of Thysanoptera with worldwide distribu
tion (Loomans 1995). Life cycle length of C. menes changes with temperature and host
(Loomans 1995). The duration of the egg-larval stage of C. menes is 14 days and adult
emergence occurs at 37 days at 20'C when reared on T palmi (Tagashira 1992). When
reared on Thrips tabaci Lind. at 21-22'C, the egg-larval stage takes 11 days and
adults emerge at 18 days (Sakimura 1937).
Under laboratory conditions, 80% parasitism by C. menes of Frankliniella intonsa
(Trybom) has been obtained (Murai 1990). A maximum of 75% field parasitism of T
palmiwas reported in Japan (Hirose et al. 1992).
Females are more abundant in nature, but sex ratio may change under laboratory
rearing conditions due to arrhenotokous parthenogenesis. Uniparental populations
have been reported from several countries (Loomans 1995).
The existence of biotypes of C. menes with differences in color patterns, tempera
ture response, host acceptance, developmental time, sex ratio and reproduction has
been documented (Loomans, 1995). Selection of biotypes adapted to particular condi
tions of host and climate can improve the efficacy of biological control programs. The
objective of this study was to evaluate the effect of temperature on developmental
time, percent parasitism and sex ratio of two biotypes of C. menes using T palmi as
the host. Data on temperature response will be used in laboratory rearing of the par
asitoid and in selection of strains for the biological control of T palmi in the field.


Two strains of C. menes were studied, a uniparental strain collected in Kyushu, Ja
pan in 1990 (Tagashira 1992) and a biparental strain collected in Chiang-Mai, Thai
land, by R. M. Baranowski and F D. Bennett (University of Florida) in January 1993.
A laboratory reared colony of the uniparental strain was brought into Florida by Y. Hi
rose (Kyushu University, Japan) in February 1992.
Groups of 20 first instar T palmi were taken from a laboratory colony and placed
on 25 mm eggplant leaf discs. The discs with the larvae were held with parafilm at the
top of 45 x 25 mm plastic tubes as described by Tagashira (1992). One parasitoid fe
male was placed in every tube and the open end of the tube was sealed with parafilm.
In the case of the Thai strain, as unmated females produce only males, pairs of C.
menes were kept for 6 h at 25'C in daylight in 70 x 10 mm vials to ensure mating be
fore females were placed in the plastic tubes. Tubes were maintained in incubators at
21, 23, 25, 27 and 29"C (+ 1C), and a photoperiod of 12:12 h (L:D). After twenty-four
hours, tubes were opened and female parasitoids were removed. Leaf discs with thrips
were transferred from the tubes into 500 mm petri dishes having a layer of plaster of
Paris in the bottom and covered with humid filter paper. Thrips larvae in the petri
dishes were kept in incubators at the conditions described above.
Last instar thrips were considered to be parasitized only when the parasitoid could
be clearly recognized through the body wall. Parasitized larvae were counted daily
and transferred to 25 x 8 mm moist filter paper strips. Filter paper strips with para
sitized larvae were placed in 70 x 10 mm vials. Vials were sealed with parafilm and
returned to the incubators. Percent mortality [(number of dead parasitoid larvae and
pupae per total parasitized larvae).100], percent parasitism [(number of parasitized
thrips larvae per 20).100] and sex ratio (male: female) were calculated.
Three replicates of 10 tubes containing a female wasp and 20 thrips larvae were
studied at each temperature. Developmental time data were square root transformed.
Percent mortality and parasitism data were arcsine transformed. Developmental
time data were analyzed using Statistical Analysis System general linear models
(SAS 1985) for a balanced ANOVA. Means were separated by Waller-Duncan k-ratio

March, 1996

Castineiras et al.: Temperature response ofCeranisus menes 15

t-tests. Developmental thresholds were estimated from linear regression equations of
the developmental rates against temperature, as described by Varley et al. (1973), us
ing Sigma Stat statistical software (Kuo et al. 1992).

Parasitoid developmental time decreased with increased temperature only be
tween 21 and 27'C (Tables 1 and 2). Thai males emerged one day before females at
25'C, but at lower or higher temperatures the difference in the dates of emergence in
creased (Table 2).
Japanese strain percent mortality increased at temperatures above or below 23C
(Table 3). In the Thai strain no significant difference was observed in mortality at 23
and 25'C, but above 25'C and below 23'C mortality increased. Only 4-6% of the Jap
anese parasitoids survived at 29C (Table 3). At 21'C and above 25'C pupae were fre
quently contaminated with fungi.
No statistical differences in percent parasitism were observed between 23 and
29"C in either strain, but parasitism decreased about 50% at 21'C compared with
temperatures equal to or above 23'C. (Table 3)
Sex ratios of the Thai strain reared at different temperatures were not signifi
cantly different (F= 0.30; df= 4,10; p= 0.87). The mean sex ratio (male:female) was
1:1.9 (+ 0.008).
Linear regression models show the effect of temperature on the rates of develop
ment of the parasitoids. Coefficients of determination equal to or above 84% were ob
trained for developmental rates against temperature (Table 4). The Japanese and Thai
strains showed different thermal constants in the time periods from egg to pupa, but
the number of degree-days needed for pupation and for total development were the
same for the females of the two strains. Thai males had lower thermal constants and
higher developmental thresholds than the Japanese and Thai females (Table 4).


The data in Tables 1 and 2 show that strains responded to temperature by increase
ing or decreasing their developmental time. Similar results on the duration of the egg
larval and pupal stages at comparable temperatures were reported for a Japanese
strain of C. menes reared on T tabaci (Sakimura 1937).


Developmental Time in Days' (Mean SD)
Temp ('C) Egg-pupa Pupa Total

21 12.51 + 0.33 a 22.56 + 0.07 a 35.07 + 0.39 a
23 10.71 + 0.37 b 18.86 + 0.30 b 29.57 + 0.09 b
25 9.80 + 0.26 c 16.94 + 0.15 c 26.68 + 0.11 c
27 8.18+ 0.16 d 14.15 + 0.06 d 22.45 + 0.08 d
29 8.59+ 0.39 d 14.43+ 0.31 d 22.72 + 0.33 d

'Means within a column followed by the same letter are not significantly different according to aWaller Dun
can k-ratio t-test on square root transformed data (p > 0.001, k-ratio = 100). Untransformed means are pre

Florida Entomologist 79(1)

-i Ll N- CT

March, 1996

Castineiras et al.: Temperature response ofCeranisus menes 17


% Mortality' (Mean SD) % Parasitism (Mean SD)

Temp ('C) Japanese Strain Thai Strain Japanese Strain Thai Strain

21 30.16 + 0.41 b 39.38 0.12 a 11.73 0.83 b 11.53 0.53 b
23 12.17 + 0.79 e 11.83 0.72 d 24.17+ 0.72 a 25.50+ 0.50 a
25 19.70 + 0.89 d 14.23+ 1.35 d 28.00+ 1.15 a 26.56+ 0.85 a
27 25.86 +0.96c 18.13 +0.36c 26.53 1.11 a 28.56 +0.89a
29 94.97 + 0.88 a 27.36+ 1.09 b 23.83+ 1.58 a 28.90+ 0.30 a

'Means within a column followed by the same letter are not significantly different according to aWaller Dun
can k-ratio t-test on arcsine transformed data (p > 0.001, k-ratio = 100). Untransformed means are presented.

Percent mortality was high (Table 3). According to Hirose et al. (1993), a high level
of mortality occurs between the larval and pupal stages in the laboratory due to han
Sakimura (1937) stated that females may kill some of the young host larvae by the
insertion of the ovipositor without laying eggs. That could explain the low percent par
asitism found in our experiments (Table 3).
Temperature did not affect the sex ratio in the biparental strain, but the mean sex
ratio of 1:1.9 was higher than other reports for C. menes: 1:1.5 (Sakimura 1937, Daniel
1986) and 1:1 (Murai 1990).
Minimum temperature thresholds of the Japanese parasitoids were between 5 and
8C (Table 4). Japanese parasitoids developed and pupated at 29C, but 95% of pupae
died (Table 3). Tagashira (1992) observed that wasps of the Japanese uniparental
strain reared on F intonsa did not emerge at 30'C. We suggest that the maximum
temperature threshold for the Japanese biotype of C. menes is near 30'C.
The Thai biotype is probably better adapted to warmer temperatures than the Jap
anese biotype. Minimum temperature thresholds of this strain ranged from 7.8 to
14.2'C (Table 4). The maximum temperature threshold for the Thai strain seems to be
over 29"C. At 29"C mortality was 27.4% and percent parasitism was 28.9%(Table 3).
Comparisons of the behavior of the Thai strain with former studies are not possible.
These parasitoids were found in Thailand in 1987 and only some field data are avail
able from Hirose et al. (1993).
The best temperature for laboratory rearing of C. menes is 25'C. At 25'C develop
ment from egg to adult was 26 days for Thai males and 27 days for females of both
strains, percent mortality fluctuated between 14-20% and percent parasitism aver
aged 27%. At 21 C parasitism averaged 12% and above 27'C mortality was generally
higher than 18%.
In South Florida, where the average temperature during the growing season (Sep
tember-April) is 21-28'C, temperature does not appear to be a limiting factor for field
establishment of C. menes. Considering their response to temperature, both strains
are good candidates for biological control of T palmi in the field.


We thank J. Pena and W Meyer, Tropical Research and Education Center, Univer
sity of Florida, for reviewing the manuscript, and J. B. Coulliette for technical assis

18 Florida Entomologist 79(1) March, 1996

M 2

g I o o~ cN co

oo QOr

2 o z oo Q oo


C o6

SLd C^ o 0 o
[^ LI ^ O o
E- ^ i LO i

0 <
H ^ r CD t-


H C c'J

4 0 c"
M? <



C tq

2 D

M ; C
~r 3 L

| 2.

Kn: mn OM oO
a? W 0 0 0 D Lr


S, Q

M -- -a

H Cflt-. ^
W m ooA o

Castineiras et al.: Temperature response ofCeranisus menes 19

tance. This research was supported by CSRS-03316 Grant Development of
Management Strategies for the Melon Thrips, Thrips palmi. This is Florida Agricul
tural Experiment Station Journal Series No R-04677.


DANIEL, M. A. 1986. Thrips-parasite interactions in some Panchaetothripine Thysan
optera (Insecta: Arthropoda). Proc. Indian Natl. Sci. Acad. 52: 437-444.
HIROSE, Y., M. TAKAGI, AND H. KAJITA. 1992. Discovery of an indigenous parasitoid of
Thrips palmi Karny (Thysanoptera: Thripidae) in Japan: Ceranisus menes
(Walker) (Hymenoptera: Eulophidae) on eggplant in home and truck gardens.
Appl. Entomol. Zool. 27: 465-467.
ICHPAN. 1993. Natural enemies of Thrips palmi and their effectiveness in the
native habitat, Thailand. Biol. Control 3: 15.
Kuo, J., E. Fox, AND S. MCDONALD. 1992. Sigma Stat. User's Manual. Jandel Scien
tific, San Rafael, CA.
LOOMANS, A. J. M., AND J. C. VAN LENTEREN. 1995. Hymenopterous parasitoids of
thrips. Wageningen Agricultural University Papers. 95: 89201.
MURAI, T. 1990. Rearing method and biology of thrips parasitoid, Ceranisus menes.
Bull. IOBC/WPRS. 13: 142-146.
SAKIMURA, K. 1937. On the bionomics of Thripoctenus brui Vuillet, a parasite of
Thrips tabaci Lind., in Japan. Kontyf 11: 370-390.
SAS INSTITUTE. 1985. SAS User's Guide: Statistics. 5th ed. SAS Institute, Cary, N.C.
TAGASHIRA, E. 1992. Host acceptance by Ceranisus menes (Walker) (Hymenoptera:
Eulophidae), a larval parasitoid of thrips, with reference to host suitability for
the parasitoid. M. S. Thesis, Fac. of Agric., Kyushu University, Fukuoka, 44 pp.
VARLEY, G. C., G. R. GRADWELL, AND M. P. HASSELL. 1974. Insect Population Ecology,
an Analytical Approach. Univ. of California Press. Berkeley and Los Angeles.

Tsai: Biology ofPeregrinus maidis


Fort Lauderdale Research and Education Center
University of Florida, IFAS
Fort Lauderdale, FL 33314


The development and oviposition of Peregrinus maidis (Ashmead) (Homoptera:
Delphacidae), a serious pest and the only known vector of maize stripe tenuivirus and
maize mosaic rhabdovirus in tropical and subtropical areas, was studied on the fol
lowing plants in the laboratory: corn (Zea mays L. var. Saccharata 'Guardian'), itch
grass (Rottboellia exaltata L.), rice (Oryza sativa L. var. Mars, Saturn, Nato, Bellevue,
Labelle, Labonnet, and Starbonnet), sorghum (Sorghum bicolor (L.) Moench var. AKS
614), goose grass (Eleusine indica (L.) Gaertn), oats (Avena sativa L.), rye (Secale ce
reale L.), gama grass (Tripsacum dactyloides L.), barnyard grass (Echinochloa crus
galli L.) and sugarcane (Saccharum officinarum L.). Peregrinus maidis nymphs did

Florida Entomologist 79(1)

not develop on rye, oats, rice and sugarcane, but the adults survived for various
lengths of time on these test plants. The average length of nymphal development on
corn, itch grass, sorghum, goose grass, barnyard grass and gama grass was 17.20,
17.87, 20.21, 24.97, 27.24 and 60.50 days, respectively Adult longevity (X SD) on
corn, gama grass, itch grass, sorghum, goose grass, and barnyard grass was 36.1 +
20.0, 42.7 16.6, 28.3 11.9, 7.6 6.4, 8.1 + 7.3 and 7.3 6.6 days, respectively Ovi
position rarely occurred on sorghum, goose grass and barnyard grass. The numbers of
eggs laid per day per female on corn, itch grass and gama grass was (X SD) 21.0
2.0, 6.4 6.6, 3.5 3.0 eggs, respectively; the numbers of eggs per female per life on
these respective plants was (X SD) 612 170.1, 146 156.7 and 48 + 45.6 eggs.

Key Words: Corn delphacid, corn, itchgrass, rice, sorghum, goosegrass, barnyard
grass, gama grass.


El desarrollo y la ovoposici6n de Peregrinus maidis (Ashmead) (Homoptera: Del
phacidae), una seria plaga y el fnico vector conocido del tenuivirus de la raya del maiz
y del rhabdovirus del mosaico del maiz en areas tropicales y subtropicales, fue estu
diado en el laboratorio en las siguientes plants: arroz (Oryza sativa L. var. Mars, Sa
turn, Nato, Bellevue, Labele, Labonnet y Starbonnet), avena (Avena sativa L.), cana
de azfcar (Saccharum officinarum L.), centeno (Secale cereale L.), maiz (Zea mays L.
var. Saccharata 'Guardian'), sorgo (Sorghum bicolor (L.) Moench var. AKS 614), Echi
nocloa crusgalli L., Eleusine indica (L.) Gaertn, Rottboellia exaltata L., y Tripsacum
dactyloides L. Las ninfas de P maidis no se desarrollaron en centeno, avena, arroz y
cana de azfcar pero los adults sobrevivieron diferente tiempo en estas plants. El de
sarrollo promedio de las ninfas en maiz, R. exaltata, sorgo, E. indica, E. crusgalliy T
dactyloides fue de 17.20, 17.87, 20.21, 24.97, 27.24 y 60.50 dias, respectivamente. La
longevidad de los adults (X SD) en maiz, T dactyloides, R. exaltata, sorgo, E. ini
dica y E. crusgalli fue de 36.1 20.0, 42.7 16.6, 28.3 11.9, 7.6 6.4, 8.1 7.3 y 7.3
+ 6.6 dias, respectivamente. La ovoposici6n raramente ocurri6 en sorgo, E. indica y E.
crusgalli. El numero de huevos puesto por dia por hembra en maiz, R. exaltata y T
dactyloides fu6 de (X SD) 21.0 2.0, 6.4 6.6, y 3.5 3.0, respectivamente._El nu-
mero de huevos por hembra puestos durante toda su vida en esas plants fue (X + SD)
612 + 170.1, 146 156.7 y 48 45.6.

The corn delphacid, Peregrinus maidis (Ashmead), is not only a major pest of corn
and sorghum (Namba & Higa 1971, Chelliah & Basheer 1965), it is also the only
known vector of two important maize viruses [maize mosaic rhabdovirus (MMV) and
maize tenuivirus (MStV)] (Nault & Knoke 1981, Tsai 1975), and it is of particular eco
nomic importance in the lowland humid tropics. It has even been suggested that its
introduction and the spread of two devastating viral diseases into Central America re
sulted in the collapse of the Mayan civilization (Brewbaker 1979; however, see Nault
Namba & Higa (1971) reported that P maidis was able to survive for various
lengths of time on napiergrass (Pennisetum purpureum Schumach), vaseygrass
(Paspalum urvillei Steud.), sugarcane (Saccharum officinarum L.), sorghum (Sor
ghum vulgare Pers.), sourgrass (Trichachne insularis Nees.), Californiagrass (Bra
chiaria mutica Stapf), Job's tears (Coix lacrymajobi L.), pangolagrass (Digitaria
decumbens Stent), and nutgrass (Cyperus rotundus L.). Chelliah & Basheer (1965)
stated that this insect utilized Pennisetum typhoides (Stapf. and Hubbard), Sorghum

March, 1996

Tsai: Biology ofPeregrinus maidis

halepenise (L.), Setaria italica (Beauv.), Echinochloa colona var. frumentacca (L.) and
Paspalum scrobiculatum (L.) as breeding or feeding hosts.
Earlier studies have shown that itch grass (Rottboellia exaltata L.) was a common
host for two corn viruses, MStV and MMV, and their vector, P maidis (Tsai 1975, Falk
& Tsai 1983, Nault & Knoke 1981). However, little is known of the relationship be
tween P maidis and its alternate hosts in south Florida. The role of alternate hosts
in the epidemiology of these diseases between growing seasons and the extension of
the host range of P maidis is of great significance in studying the ecology of this in
This paper reports the development and oviposition of P maidis on six host plants.


A stock colony of P maidis was maintained on corn (Zea mays L. var. Saccharata
'Guardian') in an insectary at 27+ 1C and a 12:12 (L:D) photoperiod for over 15 years.
P maidis eggs were excised from the midribs of corn plants and placed on cut leaf
pieces of corn and allowed to hatch. At least 30 newly hatched nymphs were singly
transferred to individual culture tubes (25 mm diam) containing fresh leaf pieces of
corn, itch grass (Rottboellia exaltata L.), sorghum (Sorghum bicolor (L.) Moench. var.
AKS 614), gama grass (Tripsacum dactyloides L.), sugarcane (Saccharum officinarum
L.), and seedlings of goose grass (Eleusine indica (L.) Gaertn.), barnyard grass (Echi
nochloa crusgalli L.), rye (Secale cereal L.), oats (Avena sativa L.) and rice (Oryza sa
tiva L. var. Mars, Saturn, Nato, Bellevue, Labelle, Labonnet and Starbonnet). The
opening of the culture tube was covered with a piece of stretched Parafilm to prevent
escape or desiccation. At the same time, a group of 70-80 newly hatched nymphs was
reared on the potted young plants of each test plant species or variety and used as re
placements for dead or missing insects. The ages of replacement insects were deter
mined by body sizes and morphological characteristics as described by Tsai & Wilson
(1986). All tests were conducted at 27 1C, 60% RH, and a photoperiod of 12:12 (L:D)
in three growth chambers (Percival Scientific Inc., Boone, IA) over a period of nine
months. Each insect was checked daily for molting, and plant tissue was replaced ev
ery two days or sooner. The dates of molting, duration of stadia, mortality, and adult
longevity were recorded. Differences in nymphal development and adult longevity on
different hosts were analyzed by Student-Newman-Keul (SNK) multiple range test
(Sokal & Rohlf, 1969).
Ten to 15 pairs of newly emerged adults were placed on single plants at the 4 to
5-leaf stage and covered by a plastic cage to test for oviposition. Pairs were removed
daily to new plants which were then dissected for egg counts.


Of the 17 species and varieties of plants tested, only corn, itch grass, gama grass,
goose grass and barnyard grass were able to support P maidis development. At least
60 first instar nymphs were singly tested on each of seven varieties of rice and one va
riety of sugarcane, oats and rye; the survivorships of test insects ranged from 2 to 5
days. A group of 32 adults was also tested individually on these plants. They only sur
vived (X + SD) 3.0 + 2.4, 4.0 3.2, 2.0 + 1.5, 5.0 3.2, 3.0 2.1, 3.0 1.8, 4.0 + 2.2,
4.0 3.5, 5.0 5.3, and 10.0 + 9.9 days on rice var. Mars, Saturn, Nato, Bellevue, La
belle, Labonnet, and Starbonnet, sugarcane, oats, and rye, respectively.
There were significant (P < 0.05) differences in total nymphal development times
among some hosts, with shortest development time on corn (17.20 1.50 days; X +

Florida Entomologist 79(1)

March, 1996

o C

- c'
+1 +1

. t- :

-g -

*^ c


01 00 CM
+ ++

+1 +1 +1
,- N- ^f

0 *^ -~

o C' N-

c'J Ncj

U C U O ~
cc cz CJ
cn ~c'j 'jcnc'

-c u
b, CMJ

-c -
U o 0
C'J Lf
c'J c' c' m

U U U 0
at N h'C
Lfl C' c' c' Lf

hcc h
C'J 0 C

Tsai: Biology ofPeregrinus maidis

SD) and longest on gama grass (60.50 10.30) (Table 1). In general, first instars had
higher survival rates, compared with other stages, on most plants tested (Table 2).
Cumulative percentage of survival from first instar to adult was higher on sorghum
(60.9%) than on other hosts (Fig. 1). The adults lived significantly longer on gama
grass (42.70 16.50 days), corn 36.10 20.00 days), and itch grass (28.30 + 11.90
days) than on sorghum (7.64 + 6.44 days), goose grass (8.00 + 7.32 days) and barnyard
grass (7.27 6.55 days; P < 0.05) (Fig. 2).
Adults rarely deposited eggs on sorghum, goose grass and barnyard grass. The
number of eggs laid per day per female on corn, itch grass, and gama grass was (X
SD) 21.0 2.0, 6.4 6.6, and 3.5 3.0 eggs, respectively; the number of total eggs per
female on these respective hosts was (X SD) 612 170.10, 146 156.75 and 48 +


This study showed that there was no significant difference in length of P maidis
nymphal development on sweet corn, itch grass and sorghum, but development times
were significantly shorter on these hosts than on goose grass, barnyard grass and
gama grass (Table 1). Although adult longevity and egg production on itch grass are
significantly less than on corn, the average total number of eggs produced by each fe
male on itch grass is still considerably higher when compared to other host plants.
Itch grass is considered to be one of the world's most serious weed pests and is cur
rently found in Florida, Louisiana, Texas, Arkansas, Alabama and Georgia, and as far
north as latitude 23 (Holm et al. 1977). Research has shown that itch grass is a good
host for MStV and MMV (Bradfute & Tsai 1983, Falk & Tsai 1983, Herold 1972, Gin
gery et al. 1981, Lastra 1977). It is the most important alternate host for P maidis
and MStV and MMV in south Florida between corn growing seasons (Tsai, unpub
Sorghum is grown throughout the eastern United States for grain and forage
(Bailey & Bailey 1976). Like itch grass, several sorghum species have been shown to
be hosts of MStV and MMV (Greber 1983, Herold, 1972, Gingery et al. 1981). Chelliah
& Basheer (1966) reported that sorghum was the preferred host for P maidis in lab
oratory tests. Peregrinus maidis has been reported as a serious pest on new sorghum
varieties in India (Agriwal et al. 1981). Sorghum species have thus been suggested as
the ancestral hosts for P maidis and MStV and MMV (Nault 1983). Although P mai



Host 1st 2nd 3rd 4th 5th

Z. mays 94.7 65.6 74.6 86.4 55.3
R. exaltata 70.5 85.0 73.5 80.0 58.3
S. bicolor 90.0 100.0 84.6 100.0 80.0
E. indica 82.9 82.8 65.8 71.1 60.4
E. crusgalli 86.4 44.4 69.0 83.3 88.0
T dactyloides 91.4 78.1 86.0 81.4 31.4

24 Florida Entomologist 79(1) March, 1996

dis feeding on sorghum was demonstrated in the present study, it may not be suitable
for breeding because few eggs were laid on this host.
Goose grass is found through the great plains and occasionally in California, Ore
gon, Utah and Arizona (Bailey & Bailey 1976). Gama grass and barnyard grass are
found throughout the eastern United States (Small, 1972). Gama grass, now estab
lished in Florida and Louisiana has the potential for spreading as far north as Min
nesota as a row crop competitor (Patterson & Quimby 1978; Patterson et al. 1979).
This study has demonstrated that P maidis could also feed and reproduce on gama
grass, but it is not as suitable a host as sweet corn and itch grass.
Both itch grass and gama grass flower and shed seeds year-round. Germination of
seed is staggered, resulting in continuous growth of new seedlings throughout the
year. The northward spread of these two weeds and the adaptation of MStV and MMV
to a perennial host or to other delphacid species with a broader host range and wider
distribution than P maidis might allow these viruses to become established in tem
operate regions of the United States.
Despite the short adult longevities on sorghum, goose grass and barnyard grass, P
maidis lived long enough to oviposit a few eggs on these plants. Tsai & Zitter (1982)
reported transovarial passage of MStV. This eliminates the need for an alternate host
for the virus, with newly hatched insects already carrying the disease agent.
The long distance spread of P maidis and two viruses by strong winds has been
proposed (Brewbaker 1979, Bradfute et al. 1981). It is possible that viruliferous P
maidis could be transported in a similar manner to the corn belt during the spring
and summer. Although the northward distribution of P maidis is probably prevented
by winter temperatures and the apparent absence of an overwintering stage (Tsai &
Wilson 1986), the adaptability of the insect and the presence of these alternate hosts
as possible overwintering sites may pose a serious threat to northern corn growers.



aI 60

" 0 -B- Z.mays
= 40
r -^ S.bicolor

20 E.indica
-- T.dactyloides

1 2 3 4 5

Nymphal Instar

Figure 1. Cumulative percentage survival of Peregrinus maidis nymphs reared on
six host plants.

Tsai: Biology ofPeregrinus maidis

Tripsacum dactyloides

Zea mays

Rottboellia exaltata

Sorghum bicolor

Eleusine indica

Echinochloa crusgalli -

0 10 20 30 40 50 60 70


Figure 2. Mean adult longevity (in days) of Peregrinus maidis on six host plants
(means SD).


Appreciation is extended to Ms. Lori A. Bohning for conducting various experi
ments and to K. H. Wang for statistical analysis. This research was supported in part
by grants from the Pioneer Hi-Bred International, Inc. and The American Seed Re
search Foundation, Washington, D.C. Florida Agricultural Experiment Stations Jour
nal Series # R-04761.


AGRIWAL, R. K., R. S. VERMA, AND G. S. BHARAJ. 1981. Screening of sorghum lines for
resistance against shoot bug, Peregrinus maidis (Ashmead) (Homoptera: Del
phacidae). JNKVV (Jawaharal Nehru Krishi Vishwa Vidyalaya) Res. J. 12: 116.
BAILEY, L. H., AND E. Z. BAILEY. 1976. Hortus Third. Macmillan Publishing Co., Inc.
1290 pp.
BRADFUTE, O. E., J. H. TSAI, AND D. T. GORDON. 1981. Corn stunt spiroplasma and vi
ruses associated with a maize epidemic in southern Florida. Plant Dis. 65: 837
BRADFUTE, O. E., AND J. H. TSAI. 1983. Identification of maize mosaic virus in Florida.
Plant Dis. 67: 1339-1342.
BREWBAKER, J. L. 1979. Diseases of maize in the wet lowland tropics and the collapse
of classic Maya civilization. Econ. Bot. 33: 101-118.
CHELLIAH, S., AND M. BASHEER. 1965. Biological studies of Peregrinus maidis (Ash
mead) (Araeopidae: Homoptera) on sorghum. Indian J. Entomol. 27: 466-471.

Florida Entomologist 79(1)

FALK, B. W., AND J. H. TSAI. 1983. Physiochemical characterization of maize mosaic
virus. Phytopathology 73: 1536-1539.
HEROLD, F. 1972. Maize mosaic virus. No. 94, in Descriptions of Plant Viruses. Com-
monw. Mycol. Inst., Assoc. Appl. Biol. Kew, Surrey, England. 4 pp.
HOLM, L. G., D. L. PLUCKNETT, J. V. PANCHO, AND J. P. HERBERGER. 1977. The world's
worst weeds, distribution and biology. The East-West Center. The University
Press of Hawaii. 609 pp.
GINGERY, R. E., L. R. NAULT, AND O. E. BRADFUTE. 1981. Maize stripe virus: Charac
teristics of a member of a new virus class. Virology 112: 99-108.
GREBER, R. S. 1983. Characteristics of viruses affecting maize in Australia, pp. 206
218 in D. T Gordon, J. K. Knoke, L. R. Nault, and R. M. Ritter, [eds.]. Proc. Int.
Maize Virus Dis. Colloq. and Workshop. Ohio Agric. Res. Dev. Cent., Wooster,
Ohio. 266 pp.
LASTRA, R. J. 1977. Maize mosaic and other maize virus and viruslike diseases in Ven
ezuela, pp. 30-39 in L. E. Williams, D. T Gordon, and L. R. Nault, [eds.], Proc.
Int. Maize Virus Dis. Colloq. and Workshop. Ohio Agric. Res. Dev. Cent.,
Wooster, Ohio, 145 pp.
NAMBA, R., AND S. Y. HIGA. 1971. Host plant studies of the corn planthopper, Peregri
nus maidis (Ashmead), in Hawaii. Proc. Hawaii Entomol. Soc. 21: 105-108.
NAULT, L. R., AND J. K. KNOKE. 1981. Maize vectors, pp. 77-84 in D. T Gordon, J. K.
Knoke, and G. E. Scott [eds.], Virus and viruslike diseases of maize in the
United States. Southern Coop. Series Bull. 247. p. 218.
NAULT, L. R. 1983. Origin of leafhopper vectors of maize pathogens in Mesoamerica,
pp. 75-82 in D. T Gordon, J. K. Knoke, L. R. Nault, and R. M. Ritter, [eds.], Proc.
Int. Maize Virus Dis. Colloq. and Workshop. Ohio, Agric. Res. Dev. Cent.,
Wooster, Ohio, 266 pp.
PATTERSON, D. T., AND P. C. QUIMBY, JR. 1978. Itchgrass: a potential noxious weed in
Mississippi. Mississippi Agric. For. Exp. Stn. Res. Rep. 3: 13.
PATTERSON, D. T., C. R. MEYER, E. P. FLINT, AND P. C. QUIMBY, JR. 1979. Tempera
ture responses and potential distribution of itch grass (Rottboellia exaltata) in
the United States. Mississippi Agric. For. Exp. Stn. Res. Rep. 27: 7782.
SMALL, J. K. 1972. Manual of the southeastern flora. Hafner Publishing Co. 1554 pp.
SOKAL, R. R., AND F. J. ROHLF. 1969. Biometry -the principles and practice of states
tics in biological research. W. H. Freeman and Co. 776 pp.
TSAI, J. H. 1975. Occurrence of a corn disease in Florida transmitted by Peregrinus
maidis. Plant Dis. Rep. 59: 830-833.
TSAI, J. H., AND S. W. WILSON. 1986. Biology of Peregrinus maidis with descriptions
of immature stages (Homoptera: Delphacidae). Ann. Entomol. Soc. America 79:
TSAI, J. H., AND T. A. ZITTER. 1982. Transmission characteristics of maize stripe virus
by the corn delphacid. J. Econ. Entomol. 75: 397-400.

March, 1996

Thorne et al.: Nasutitermes Nests and Nodules


'Department of Entomology
University of Maryland
College Park, MD 20742-5575

Smithsonian Institution
Museum Support Center
4210 Silver Hill Road
Suitland, MD 20746

3Department of Zoology
223 Bartram Hall
University of Florida
Gainesville, Florida 32611


Nest architecture of the arboreal Neotropical termites Nasutitermes acajutlae
(Holmgren) and N. nigriceps (Haldeman) is described, with special reference to carton
inclusions or nodules found within the normal gallery matrix of some nests. Nutrient
analyses of these nodules show that they have high cellulose and low cutin concentra
tions in comparison to normal nest carton. These data support the hypothesis that the
nodule inclusions serve as a form of facultative food storage in some nests of these ter
mite species. These cases appear to represent a rare situation in which food is not
stockpiled or cultured by termites, but rather some partially processed, masticated
food is incorporated into the nest matrix for future consumption.

Key Words: Termites, Nasutitermitinae, inclusions, food storage.


Se describe la arquitectura del nido de las termitas neotropicales Nasutitermes
.. ..:., .. 11l i.i i .... Ii y N nigriceps (Haldeman), con referencia especial a inclusiones
de carton o n6dulos encontrados dentro de la matriz de la galeria de algunos nidos. El
analisis de nutrients de los n6dulos muestra que estos tienen concentraciones altas
de celulosa y bajas de cutina, en comparaci6n con el carton normal de los nidos. Los
datos sostienen la hip6tesis de que las inclusiones de los n6dulos sirven como una
forma facultativa de almacenamiento de alimento en algunos nidos de termitas de
esas species. Estos casos parecen representar una rara situaci6n en la cual el ali
mento no es almacenado en pilas o cultivado por las termitas, sino masticado y par
cialmente procesado e incorporado a la matriz del nido para consume future.

The tropicopolitan termite genus Nasutitermes (Termitidae; Nasutitermitinae) is
the most speciose of all isopteran genera, containing approximately 75 described spe
cies from the Neotropics alone (Araujo 1977). Unlike most termites, many species of
Nasutitermes build arboreal carton nests composed of wood and salivary and fecal flu

Florida Entomologist 79(1)

ids (Light 1933), and occasionally other materials such as sand (Thorne & Haverty,
pers. obs.). Most other nest-building termites build mounds on the ground (e.g., Em
erson 1938), but nesting in trees has enabled species of Nasutitermes and several
other genera to colonize and exploit a new habitat.
Nasutitermes nigriceps (Haldeman) is a geographically widespread termite, rang
ing at least from Panama north throughout the lowland forests of Central America
into Mexico. It is also found on Jamaica and on Grand Cayman Island (Araujo 1977,
Thorne et al. 1994). N ..... .. (Holmgren), which is morphologically very similar to
N nigriceps, is found on Puerto Rico, the US and British Virgin Islands (BVI), Trin
idad, and Guyana (Emerson 1925,Araujo 1977, Thorne et al. 1994). There are isolated
reports of members of the N nigriceps "complex" from other parts of South America,
but a comprehensive taxonomic analysis of specimens is needed to verify species iden
tity of the South American fauna.
Despite the abundance of Nasutitermes arboreal nests, the chemical composition
of the carton material has not been examined in detail in any species (but see Becker
& Seifert 1962 for data on ash and lignin content). Knowledge of the composition of
the nest is fundamental in determining origin of nesting materials, cost of construct
tion, variation among colonies and species, and ability of the termites to allocate com
ponents of their diet for nest construction.
A distinctive feature of some N .- ..... and N nigriceps nests is the presence of
rounded carton inclusions or "nodules" within the typical gallery matrix (Hubbard
1877 pp. 268, 270, Andrews 1911 pp. 200-202, Emerson 1938 p. 264, Wolcott, cited in
Martorell 1945 p. 361). These nodules appear to be of a similar carton construction as
the rest of the nest, but they are a lighter brown color, are formed in dense concentric
sheaths (Fig. 1), and they may possibly serve as a form of food storage (Hubbard 1877;
Andrews 1911). Kemner (1929) interprets the presence of carton nodules in the Javan
termite Microcerotermes depokensis Kemner as food storage structures. Noirot (1959)
reported compact masses of wood fragments in the central cavity of nests of Globiter
mes sulphureus (Haviland). Some termite genera do store food as dried vegetative el
ements in specialized portions of their nests ("attics") [e.g., Hodotermitinae
(Hodotermes, Microhodotermes, Anacanthotermes); Rhinotermitidae (Psammoter
mes); Termitidae: Amitermitinae (certain Amitermes and Drepanotermes), Nasutiter
mitinae (certain species of Tumulitermes, Nasutitermes and 7 1..,. -. .. i. (Noirot
1970, Bouillon 1970). The "fungus growing" termites (certain Macrotermitinae) cul
ture fungus within the nest as a supplemental food source. Interestingly, some fungus
growing termites store vegetative materials within the nest before they are included
in the fungus garden (Pseudacanthotermes, Acanthotermes, some Macrotermes)
(Grasse & Noirot 1951). If the Nasutitermes nodules described in this paper are in
deed food reserves, they are not simply stored food but rather elements which have al
ready been masticated and partially processed by the termites, then positioned within
the nest matrix for future consumption.
In this paper we describe the architecture of N. ..:.. .. and N. nigriceps nests
from sites in Panama and the BVI. Observations of the nodule inclusions are reported.
Nutrient analyses of two nests without nodules and comparative chemical analyses of
nodule material versus the surrounding "normal" carton matrix of two nests with nod
ules are presented and reported.


Eight N nigriceps nests were collected within 5 km of the Panama Canal in 1980
and 1981; only one of these contained the distinctive nodules within the carton next
matrix. This arboreal nest was collected from the Gigante East Peninsula near Barro

March, 1996

Thorne et al.: Nasutitermes Nests and Nodules

Colorado Island on 7 April 1981. The entire nest was pried from the host tree, placed
within a plastic bag, and taken to the laboratory of the Smithsonian Tropical Re
search Institute on Barro Colorado Island. The nest was dissected by sequential shav
ing after being refrigerated for 24 hours to inactivate the termites (technique
described in Thorne 1984).
Nest carton from four colonies (one N. nigriceps nest collected near Barro Colorado
Island, Panama in 1981; three N. acajutlae nests collected in 1988 and 1989 on Guana
Island, BVI) was analyzed in 1989-1990. Two of the nests (the N. nigriceps nest from
Panama and a 1988 N. acajutlae nest from the BVI) contained nodules. Chemical com
position of both the nodules and a sample of the more typical dark carton material was
determined from those two nests, and samples of typical carton matrix from parts of
two additional Guana Island N. acajutlae nests (which did not contain nodules) were
also analyzed. Type of nest material examined is presented in Table 1.

Materials and Methods for Nutritional Analyses of Nest Samples

In the laboratory, samples were dried at 60 C to constant mass (approximately 24
h). Dried samples were ground to pass through a 1 mm screen in a Wiley mill. A por
tion of each sample was dried at 105 C to determine percent dry matter and then
placed in a muffle furnace for 3 h at 500 C to determine percent organic matter and
ash (an estimate of total mineral content). In vitro organic matter I'. I I ,I or per
cent fermentable substrate, was determined by the Tilley & Terry (1963) method as
modified by Moore & Mott (1974). This analysis consists of a 48 h incubation under
CO, at 39 C with an inoculant of steer rumen fluid followed by a 48 h acid-pepsin
treatment to remove undigested microbes. The percent of organic matter that disap
pears during the 96 h is the in vitro organic matter 11 -. 1 1.. 1. ,
Percentage of neutral detergent fiber (NDF: cellulose, hemicellulose, lignin and cu
tin) was measured by the Van Soest technique (Goering & Van Soest 1970) with deca
lin and sodium sulfite omitted (Golding et al. 1985). Analyses for percentages of acid
detergent fiber (ADF: cellulose, lignin and cutin), potassium permanganate lignin,
and cutin followed Goering & Van Soest (1970). Percent hemicellulose is estimated by
subtracting ADF from NDF Lipids were extracted with ethyl ether in a Goldfisch ap
paratus for 8 h. Percent concentrations of total (Kjeldahl) nitrogen and phosphorus
were measured with a block digester (Gallaher etal. 1975) and an automated Techni
con analyzer (Hambleton 1977). Energy content of food and feces was determined in
a bomb calorimeter following standard procedure (Parr Instrument Co. 1960).
One sample was analyzed from each source of nest material. In each analysis, two
subsamples were evaluated. Values for replicates of each sample were accepted within
1% relative error. Relative error is calculated as (a-b)/(a+b) where a and b are repli
cate values. Rarely, the values obtained for the duplicates were not within 1 relative
percentage, in which case a third subsample was analyzed. Table 1 reports the mean
of the analyzed values for each sample.


Nest Architecture

Nests built by N .../. i N. nigriceps can be among the largest of any arbo
real nesting Nasutitermes. Maximum dimensions of an ellipsoidal nest can approach
2 m in height and 1 m in girth (Thorne et al. 1994). The exterior of these nests is typ
ically medium to greyish brown in color and irregularly mottled, generally with

Florida Entomologist 79(1)

rather shallow bumps, unlike the dark nests with small bumps characteristic of the
exterior of N corniger (Motschulsky) and N. costalis (Holmgren) nests or the lighter
brown, smooth shell typical of N ephratae (Holmgren) (Thorne 1980; Haverty et al.
1990). Young nests may be difficult to identify to species, but differences in exterior
appearance make it possible to visually discriminate most mature nests of N acajut
lae and N nigriceps from those of N corniger, N costalis, or N ephratae. The outer car
ton shell of nests of all of these Nasutitermes species has small pinpoint holes, visible
if a piece is held up to a light. These holes presumably function in gas exchange.
The intercalated matrix of galleries within mature nests of N acajutlae and N ni
griceps tends to be larger (chamber diam up to 2.1 cm) and with thicker carton walls
(up to 0.6 cm near the exterior of a nest; exceeding 1.7 cm near the interior of the nest)
than in nests of arboreal Nasutitermes found sympatrically with one or both of these
species (N. columbicus, N. corniger, N. costalis, N ephratae). The royal "cell" within the
nest is often positioned near the central longitudinal axis of the nest, and frequently
located in or near a branch fork or knothole of the host tree. In small to medium sized
nests (< 60 cm diam) the royal cell is a distinctly thicker sphere or ellipsoid of layered
carton (generally up to 12-15 cm in diam) surrounding the royal chamber. In nests ex
ceeding 1 m on an axis the royal chamber can be embedded in the dense carton center
of the nest, with the royal cell becoming indistinct from the remainder of the central,
reinforced portion of the nest. We have not found distinguishing characters to differ
entiate nest architecture of N acajutlae from N nigriceps.
The carton-covered foraging trails built by large N acajutlae and N nigriceps col
onies are wider and less regular than in N. corniger N. costalis, or N. ephratae. Ter
mites occupying small to medium sized N acajutlae and N nigriceps nests frequently
build simple, linear trails 0.9-1.5 cm wide, thus they are indistinct from trails of N
corniger, N costalis, or N ephratae. However, trails leaving large nests of N acajutlae
and N nigriceps are often broad (up to 14 cm in width) and deep (up to 8 cm from the
tree to ceiling of the gallery). Trails from large nests are often highly irregular along
the edges. Occasionally a "floor" is built as well so that the trail becomes a tube that
can, for a limited distance, be separate from the tree or branch. As is typical for many
arboreal Nasutitermes, tunnels built on the exterior of tree branches are frequently on
the underside of the branch. We hypothesize that this minimizes disturbance by hard
rainfall or by creatures traveling along the tops of branches. Building galleries in the
"shade" of branches would also minimize desiccation from direct sunlight. A further
advantage would be that foraging tunnels built on the undersides of branches would
receive maximum moisture from rain running off the branch. This would be beneficial
for N acajutlae or N nigriceps since they often live in relatively dry habitats (Thorne
et al. 1994).

Description of Nodules and Nest Population

We describe nodule inclusions in three nests: one N nigriceps nest dissected in
April, 1981 in Panama, one N .. ...:.. .. I. I dissected in July, 1988 on Guana Island,
BVI, and a nest dissected on the island of Tortola, BVI in October, 1994 (nodules from
the latter nest were not analyzed for nutritional content).
Photographs of the interior of the N nigriceps nest collected in Panama are shown
in Fig. 1. The nest was generally spherical, about 46 cm in diam, which placed it in a
medium size category for conspecific nests in that area. Twenty nodule formations, 14
of which measured 3.0-4.8 cm in diam, but some as small as 1 cm diam, were removed
from the nest. All nodules were positioned within 4-10 cm of the nest exterior. The
nodules were of a uniform light brown color in contrast to the dark brown surrounding

March, 1996

Thorne et al.: Nasutitermes Nests and Nodules 31



c_ -_



32 Florida Entomologist 79(1) March, 1996

nest matrix. Nodule shape was generally spherical although some had distortions or
were irregular ellipses. The nest contained an active population of soldiers and work
ers, as well as a conspicuous brood of wingbud nymphs in the penultimate and ulti
mate instars. Many of the nymphs occupied the interiors of the nodule spheres. No
reproductive, eggs or immatures were found within the nest.
The N. acajutlae nest on Guana Island was irregularly ellipsoidal, measuring ap
proximately 1 m in height with a maximum diam of 75 cm. The nest contained a pri
mary queen, developing nymphs of a variety of instars, many eggs and white larvae,
and a large population of workers and soldiers. No primary king was retrieved but
that is typical during field dissections because mature kings are small enough to re
treat quickly and evade capture. The light-colored nodules were located in a zone sur
rounding the hard, inner core of the nest, all positioned at least 2 cm from the exterior
nest wall. Many of the nodules were scalloped, possibly indicating consumption by
termites. As with the N nigriceps nest, immature termites occupied the interior of
hollowed-out nodules.
An N. acajutlae nest collected on the island of Tortola, BVI was brought to us in
several pieces during a field trip in October 1994. This large nest, estimated to have
been just over a meter in height and about 80 cm in diam, contained eggs, white im
mature, soldiers, and workers. There were relatively few brachypterous nymphs, but
numerous mature alates were present. The primary queen and king were not recov
ered, but the presence of egg caches suggests that the reproductive were present in
the intact nest or in the portion of the host tree surrounded by the nest. This nest had
clusters of nodules positioned within the inner perimeter of the nest (Fig. 1 C, D)
Again, the outer 2 cm or more of nest material was dark, more typical carton matrix
with no nodules. Because the nest arrived in pieces, it was impossible to tell whether
nodules were built in the center core of the nest.
We did not do nutrient analyses of nodule material from the Tortola nest, but we
measured each of the 75 nodules that were retrieved. The distribution of nodule sizes
recovered from this nest is shown in Fig. 2. Some of the irregularly spherical nodules
in this nest were solid, dense material; most were hollowed to some extent as seen in
the Panama and Guana Island nests. Hollowed nodules contained large numbers of
immatures. Eight of the nodules from this nest were bilobed, as if two units had been
constructed or fused together.
The nodules from all three of these nests were generally similar in size, shape,
color, and position within the nest matrix. In each case immatures occupied excavated

Chemical Analyses of Nest and Nodule Material

The most striking aspect of the nutrient composition of nest materials (Table 1) is
the consistency among these nests. The only apparent differences are the higher cel
lulose and lower cutin concentrations in nodule samples than in carton samples and
the higher in vitro organic matter 1 -. 1. 1.11 values of the samples from Panama.
No statistical analyses were performed because only two samples were available in
each category. Hemicellulose was absent or present in only trace amounts in the sam
ples, so was not included in Table 1. Nutrient composition of the nest material appar
ently does not change with age. The composition of recently constructed normal
carton was very similar to that of old carton material from the North Bay and White
Bay Beach nests on Guana Island. The high organic matter content indicates that lit
tle, if any, soil or sand is incorporated into these samples of nest or nodule material.

Thorne et al.: Nasutitermes Nests and Nodules 33

S c 0 C CC 0 0 0 O O


C o LO NO N- 0z co 0 N c C

0c0O CC O! m

\3 3 M

0 C Ct -l 0 Ct c o\ LO Q -

< n -c5 c

S CD m tC CD t- o
O Q Q 6 ooo NSf

C C -Iz

z 1 -
2 U -5


0 C : C z
z C) z mm <


z oo
0 mn' CD^ Li ^cs 0 '
1- g'2 CT T T T1 C1 C1 C1 T 0 n
a? g MOcc

Florida Entomologist 79(1)

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Diameter (cm)

Figure 2. Size frequency distribution of nodules removed from the Nasutitermes
acajutlae nest collected on the island of Tortola, BVI in October 1994 (N = 75; x = 2.6
+ 1.2 cm).

Nasutitermes acajutlae and N. nigriceps are exceptional among termites in build
ing distinctive inclusions or nodules within the normal carton matrix of their nests.
The only other termite reported to build similar structures is the Javan termite Mi-
crocerotermes depokensis (Kemner 1929).
Two contrasts between the composition of nodules versus normal nest carton an
alyzed in this study may be biologically significant. First, the nodule samples have
lower cutin and higher cellulose percentages than do samples of the surrounding,
dark carton matrix. Cutin degradation is not possible for most organisms except some
specialized fungi; digestion of cutin by termites is unknown (Breznak, pers. comm.).
The differences in cutin and cellulose percentage may indicate that the termites are
constructing the nodules from materials with greater I,-. I. II1, The relatively high
cutin percentages in typical carton probably enhances water-proofing and construct
tion strength. It is unlikely that the difference in cutin abundance is due to transfer
of waxes from the insect exocuticle to the nest walls. The percentage of cutin in fresh,
newly constructed carton (having minimal opportunity for contact transfer of waxes
from passing insects) does not differ markedly from that of old, dense, interior carton

March, 1996

Thorne et al.: Nasutitermes Nests and Nodules

(see samples from North Bay and White Bay Beach nests, Guana Island, BVI in Table
A second distinction is that both the nodule sample and the normal gallery within
the nodule sample from Panama have higher in vitro 1 :' 1I1.1111 than do any of the
BVI samples. This may reflect species differences in carton processing, or a difference
in diet among the two populations (N acajutlae sampled from Guana Island were
feeding substantially on sea grape, Coccoloba uvifera, the diet of the N nigriceps from
Panama is unknown). Clearly, further sampling and geographic variation within each
species must be examined before differences of this type can be further evaluated.
Hubbard (1877) and Andrews (1911) hypothesized that these Nasutitermes nod
ules serve as food storage. Kemner (1929) came to a similar conclusion in the case of
Microcerotermes depokensis. The food storage hypothesis is supported by the higher
cellulose content of nodules in comparison to surrounding nest carton in both N. aca
jutlae and N. nigriceps. It is difficult to know the conditions under which the nodules
might be naturally consumed in a nest, but 0.3 g portions of both N. acajutlae and N.
nigriceps nodules offered to 100 workers of N acajutlae, N nigriceps, N costalis, and
Zootermopsis nevadensis (Hagen) were consumed in the laboratory within 24 h, and
consumed by 100 Reticulitermes flavipes (Kollar) workers within 48 hr (N=3 per spe
cies). These species did not consume the normal carton of either N ...:./. ... N. ni
griceps nests.
Termite nest material can be used as nutritional food reserves in some species.
Hegh (1922) commented that termites in mature colonies of Microcerotermes fuscotib
ialis '~ -. i 11) eat the internal walls of their nests during times of food stress. Noirot
(1970) reported that central walls of Cephalotermes rectangularis '-~ I 11) nests can
be used to culture the termites in the laboratory.
The function of nodules and circumstances under which they are constructed are
difficult to identify because they are found so rarely. In both Panama and the BVI, ex
amination of nests of approximately the same size, in the same local area, at the same
season never revealed another live colony with nodules. Because young were found
within the nodules of both N. acajutlae (white immatures instars 1 3) and N nigriceps
(developing alate nymphs) the nodule food reserves may be sequestered for juveniles.
Comparable nests with immatures, however, did not have nodules. Nodule construct
tion may be influenced by individual colony health, age, microhabitat, food resources,
caste proportions, or population size. Even among colonies producing nodules, they
may be ephemeral within a nest. Nodules may only be present seasonally, stockpiled
as food reserves and then consumed during times of high demand (as when alate
brood is present), when food is scarce, or when travel from the nest is physiologically
expensive (as in a drought). It is notable that the only two Nasutitermes species
known to construct these nest inclusions are the closely related species N acajutlae
and N nigriceps, both of which can occupy dry and thus potentially stressful environ
ments (Thorne et al. 1994). The facultative ability to store food in nodules, combined
with an exceptional desiccation tolerance of individuals, may contribute to the sur
vival of these two Nasutitermes species in arid or otherwise marginal habitats not col
onized by other members of the genus.


We extend sincere thanks to Dr. and Mrs. Henry Jarecki and the staff of The
Guana Island Club for their support and hospitality during the course of this project.
John A. Breznak, Michael I. Haverty, and James D. Lazell provided constructive input
and productive discussions. Field assistance was contributed by James Egelhoff,

Florida Entomologist 79(1)

Thomas Jarecki, and Ralph Rusher. Portions of this research were funded by The Con
servation Agency through a grant from the Falconwood Foundation, The Smithsonian
Tropical Research Institute, NSF grant DEB-8607407 to BLT, and by a cooperative
agreement between the University of Maryland and the Pacific Southwest Research
Station, U.S. Department of Agriculture.


ANDREWS, E. A. 1911. Observations on termites in Jamaica. J. Anim. Behav. 1: 193
ARAUJO, R. L. 1977. Catalogo dos Isoptera do Novo Mundo. Academia Brasileira de
Ciencias, Rio de Janeiro, 92 pp.
BECKER, V. G., AND K. SEIFERT. 1962. Ueber die chemische zusammensetzung des
nest -und galeriematerials von termiten. Ins. Soc. 9: 273-289.
BOUILLON, A. 1970. Termites of the Ethiopian Region, pp. 73-125 in K. Krishna and
F M. Weesner [eds.] Biology of Termites, Volume II. Academic Press, N.Y.
EMERSON, A. E. 1925. The termites of Kartabo, Bartica District, British Guiana. Zoo
logica 6: 291 459.
EMERSON, A. E. 1938. Termite nests. A study of the phylogeny of behavior. Ecol.
Monogr. 8: 247-284.
GALLAHER, R. N., C. O. WELDON, AND J. G. FUTRAL. 1975. An aluminum block digester
for plant and soil analysis. Soil Sci. Soc. American Proc. 39: 803-806.
GOERING, H. K., AND P. G. VAN SOEST. 1970. Forage fiber analyses (Apparatus, re
agents, procedures, and some applications). Agric. Handbook 379. U.S.D.A.,
Washington, D.C.
GOLDING, E. J., M. F. CARTER, AND J. E. MOORE. 1985. Modification of the neutral de
tergent fiber procedure for hays. J. Dairy Sci. 68: 2732-2736.
GRASSE, P. P., AND C. NOIROT. 1951. Nouvelles recherches sur la biologie divers Ter
mites champignonnistes (Macrotermitinae). Ann. Sci. Nat. Zool. Biol. Animale
[11] 13: 291-342.
HAMBLETON, L. G. 1977. Semiautomated method for simultaneous determination of
phosphorus, calcium and crude protein in animal feeds. J. Assoc. Off. Agric.
Chem. 60: 845-852.
HAVERTY, M. I., B. L. THORNE, AND M. PAGE. 1990. Surface hydrocarbon components
of two species of Nasutitermes from Trinidad. J. Chem. Ecol. 16: 2442-2450.
HEGH, E. 1922. Les Termites. Imprimerie Industrielle & Financiere, Bruxelles.
HUBBARD, H. G. 1877. Notes on the tree nests of termites in Jamaica. Proc. Boston
Soc. Nat. Hist. 19: 267-274.
KEMNER, N. A. 1929. Ans der Biologie der Termiten Java. 10th Congr. Int. Zool.,
Budapest, 2: 1097-1117.
LIGHT, S. F. 1933. Termites of Western Mexico. Univ. of California Publ. in Entomol.
MARTORELL, L. F. 1945. A survey of the forest insects of Puerto Rico. Part II. The Jour
nal of Agriculture of the University of Puerto Rico. The Agricultural Experi
ment Station, Rio Piedras, PR. 29: 355-608.
MOORE, J. E., AND G. O. MOTT. 1974. Recovery of residual organic matter from in vitro
digestion of forages. J. Dairy Sci. 57: 1258-1259.
NOIROT, C. 1959. Le nid de Globitermes sulphureus Haviland au Cambodge. Ins. Soc.
6: 259-269.
NOIROT, C. 1970. The nests of termites, pp. 73-125 in K. Krishna and F M. Weesner
[eds.] The Biology of Termites, Volume II. Academic Press, N.Y.
PARR INSTRUMENT CO. 1960. Oxygen bomb calorimetry and combustion methods.
Technical Manual Parr Instrument Company 130: 156.
THORNE, B. L. 1980. Differences in nest architecture between the Neotropical arbo
real termites Nasutitermes corniger and Nasutitermes ephratae (Isoptera: Ter
mitidae). Psyche 87: 235-243.

March, 1996

Thorne et al.: Nasutitermes Nests and Nodules 37

THORNE, B. L. 1984. Polygyny in the Neotropical termite Nasutitermes corniger life
history consequences of queen mutualism. Behav. Ecol. Sociobiol. 14: 117 136.
THORNE, B. L., M. I. HAVERTY, AND M. S. COLLINS. 1994. Taxonomy and biogeography
of Nasutitermes acajutlae and N nigriceps (Isoptera: Termitidae) in the Carib
bean and Central America. Ann. Entomol. Soc. America 87: 762-770.
TILLEY, J. M. A., AND R. A. TERRY. 1963. A two-stage technique for the in vitro diges
tion of forage crops. J. British Grassl. Soc. 18: 104-111.


Heath et al.: Papaya Fruit Fly Trapping System


'Insect Attractants, Behavior, and Basic Biology Research Laboratory,
Agricultural Research Service, United States Department of Agriculture,
Gainesville, FL 32604

2Centro de Desarrollo de Productos Bi6ticos, Km. 8.5 Carretera Yautepec Jojutla,
Yautepec, Morelos, Mexico

3University of Florida, IFAS, Tropical Research and Education, Homestead, Florida

4Instituto de Ecologia, A.C., Apartado Postal 63, 91000 Xalapa, Veracruz, Mexico

"APHIS-PPQ, United States Embassy Guatemala, APO Miami, Florida 34024


A membrane-based formulation method that provided a constant release rate of
synthetic pheromone for the papaya fruit fly, Toxotrypana curvicauda Gerstaecker,
was developed. Release rate measurements over 23 days indicated that lures loaded
with 5, 15, 25, and 50 pl of synthetic pheromone released an average of 120, 360, 580
and 1120 ng per hr and the half-life of the lures was estimated to be 67, 184, 300 and
48 days, respectively. Field tests conducted in Mexico compared efficacy of blank and
pheromone-baited sticky green spheres, cylindrical traps made from green opaque
plastic that either contained a toxicant or were coated with sticky material, and cy
lindrical traps prepared from green sticky paper. Green opaque traps containing a
toxicant and sticky paper traps captured approximately five times more papaya fruit
flies than either the sticky-coated green opaque traps or the sticky-coated green
spheres, and the presence of pheromone did not affect numbers of flies captured. Thus,
the combination of the green color and the cylindrical shape provided a visual cue suf
ficient for papaya fruit fly capture. The pheromone lure significantly increased trap
capture in similar tests conducted in Guatemala. Capture was highest in the sticky
paper traps and in sticky-coated spheres. Use of the membrane-based synthetic pher
omone in a cylindrical trap may provide an effective tool for monitoring papaya fruit

Key Words: Insecta, papaya fruit fly, pheromone formulation, trap.

Florida Entomologist 79(1)


Fue desarrollado un m6todo de formulaci6n basado en una membrana que provee
una velocidad constant de liberaci6n de la feromona sint6tica para la mosca frutera
de la papaya, Toxotrypana curvicauda Gerstaecker. Las medidas de la velocidad de li
beraci6n durante 23 dias indicaron que los cebos cargados con 5, 15, 25 y 50 Pl de la
feromona sint6tica liberaron un promedio de 120, 360, 580 y 1120 ng por hora y la vida
media de los cebos fue estimada en 67, 184, 300 y 48 dias, respectivamente. En expe
rimentos de campo llevados a cabo en Mexico se compare la eficacia de esferas verdes
adhesivas con y sin feromona, trampas cilindricas hechas con plastic verde opaco que
contenian un t6xico o estaban cubiertas con material adhesive, y trampas cilindricas
hechas de papel adhesive verde. Las trampas verdes opacas que contenian un t6xico
y las trampas de papel adhesive verde capturaron aproximadamente cinco veces mas
moscas de la papaya que las trampas verdes opacas cubiertas de goma o las esferas
verdes cubiertas de goma. La presencia de la feromona no afect6 el numero de moscas
capturado. La combinaci6n del color verde y la forma cilindrica proveyeron una pista
visual suficiente para la capture de las moscas fruteras. El cebo de feromona aument6
significavivamente la capture en pruebas similares conducidas en Guatemala. La
capture fue mas alta en las trampas de papel adhesive y en las esferas cubiertas de
goma. El uso de la feromona sint6tica con el m6todo de la membrana en la trampa ci
lindrica podria ser una herramienta efectiva para el monitoreo de las moscas fruteras
de la papaya.

Toxotrypana curvicauda Gerstaecker is the principal insect pest of commercial pa
paya (Carica papaya L.); it occurs throughout the tropical and subtropical areas of the
New World (Wolcott 1933). Studies of papaya fruit fly behavior in the field and in the
laboratory indicated the existence of a male-produced pheromone that mediates inter
actions with females (Landolt & Hendrichs 1983, Landolt et al. 1985). This male-pro
duced pheromone was identified by Chuman et al. (1987) as 2-methyl-6-vinylpyrazine
(2,6-MVP). Synthetic 2,6-MVP elicited the same behavioral responses from sexually
mature unmated female papaya fruit flies as did male-produced pheromone. Field tri
als demonstrated that the pheromone used with 12.7 cm diam sticky-coated green
spheres as visual fruit cues (Landolt et al. 1988) and a release rate of approximately
1 pg per hr (12 male equivalents) was optimal for papaya fruit fly attraction (Landolt
& Heath 1990). However, general field use of this trapping system has been limited
because of the difficulty in keeping the fragile, open glass capillaries, in which the
highly volatile pheromone is formulated, vertical in order to avoid spillage. Further
more, the short life span of the adhesive necessitates continuous trap maintenance.
Cylindrical traps, which were painted to provide a visual cue and were baited with a
food-based synthetic attractant, captured males and females of the Mediterranean
fruit fly, Ceratitis capitata (Wiedemann), and a number of Anastrepha spp. (Heath et
al. 1995). Cylindrical traps with appropriate visual cues may provide an alternative
to the spheres currently used for papaya fruit fly monitoring.
Herein, we report the application of a membrane-based formulation system that
provides a variety of release rates depending on the amount of chemical used. Field
tests were conducted to determine if the membrane-release formulation and green cy
lindrical traps could be used to overcome the shortcomings of the earlier trapping sys

March, 1996

Heath et al.: Papaya Fruit Fly Trapping System


Membrane-Based Formulation System

A 3 by 5 cm lure was prepared by folding a 6 by 5 cm piece of 6 mill impermeable
polyethylene (backing) in half. A 1.17 cm diam hole was cut in the center of the front
of the lure and a piece of 1 mill high density polyethylene (membrane) film (Consep
Inc., Bend, OR) was placed inside the lure. The bottom and sides were heat sealed to
form an envelope and to secure the membrane. The release area of the membrane was
reduced to a 5 mm diam circle by placing a piece of aluminum tape (United Tape Com-
pany, Cumming, GA) over the 1.17 cm hole in the front of the lure. A piece of rectan
gular filter paper and a plastic grid that were slightly less than 3 by 5 cm were placed
in the envelope to provide stability; synthetic pheromone was added and the top of the
lure was heat sealed. Lures were loaded with 5, 15, 25 and 50 Pl of >99% pure 2,6
MVP (Fuji Flavour Co. LTD., Tokyo, Japan). Lures were placed in a hood with a 0.25
cm per sec air flow and, beginning two days after filling, the release rates from at least
three lures of each load were measured every three to four days over a 23-day period.
Volatiles were collected and analyzed using collection systems and gas chromato
graphic conditions described previously (Heath & Manukian 1992, Heath et al. 1993).
Mean release rates for each pheromone load were used in linear regression analysis
to determine the change in release rate over time and the half life of each lure.
A field-test comparison of the membrane-based lures and the capillary lures used
by Landolt & Heath (1990) was conducted in Homestead, Florida, using sticky-coated
sphere traps. Both lures released approximately 1 gg 2,6-MVP per h. Solid styrofoam
spheres (12.7 cm diam) painted dark green were used (Great Lakes IPM, Vestaburg,
MI). Capillary lures were mounted in holes drilled in the spheres and lures were po
sitioned so that lure openings were 2-3 cm from the top of the sphere. A 4 by 7 cm piece
of 6 mill plastic (Faulkner Plastics, Gainesville, FL) was folded in half to form a tent.
The membrane-based lure was attached to the inside of the tent to protect the lures
from rain, and this assembly was placed on the top of the sphere (Fig. la). Paintable
sticky coating (Tangle Trap, The Tanglefoot Co., Grand Rapids, MI) was applied to the
outside of the sphere to retain responding flies. Ten pairs of sticky spheres baited with
either a capillary lure or a membrane-based lure were hung by wires from papaya leaf
petioles near fruit clusters located about 1.5-2 m above ground in trees along the out
side edge of a papaya grove. Traps were placed in the border rows because the papaya
fruit fly activity tends to be the highest and fruit damage the greatest along the bor
ders of a grove (Landolt & Heath 1988). The experiment was replicated over time;
traps were checked every 7-10 days per replicate. There were four consecutive repli
cates. Total numbers of males and females captured on the ten traps per lure type
were summed separately to give the number of each sex captured per replicate. Sep
arate t test comparisons were conducted for numbers of males and females captured
using Proc TTEST (SAS Institute 1985).

Cylindrical Traps

Cylindrical traps (Fig. lb,c) consisted of three major components; the main trap
body, which was a cylindrical container (9 cm diam by 15 cm long); two removable end
caps for quick access into the trap for bait replacement; and a wire hanger for holding
the trap together and supporting the complete assembly on a tree (Heath et al. 1995).
The trap bodies for the green opaque traps (Fig. Ib) were prepared from a rectangular

40 Florida Entomologist 79(1) March, 1996

piece (15.0 cm wide x 30.0 cm long) of green opaque plastic (0.025 cm thick, Faulkner
Plastics, Gainesville, FL), which was rolled to form a 9.0 cm diam cylinder. The green
opaque traps used one of two methods for catching insects that responded to the trap
and/or bait. One type used paintable sticky coating, as was used for the sphere traps,
applied to the outside of the trap. The second type used internally-placed toxicant
panels (Heath et al. 1995), which contained a feeding stimulant and a toxin, to kill in
sects after they entered the trap. The toxicant panels are coated with a solution (1.0:
0.5: 0.01) of paint [Hunter Green 100% acrylic latex paint (Glidden, Cleveland, OH)],
sugar [American Chemical Society grade sucrose (Mallinkrodt, Paris, KY)], and pes
ticide [technical grade methomyl (DuPont, Newark, DE); 98%(AI)]. Panels were air
dried for at least 48 h before use and were placed on the inside of both the top and bot
tom end caps using double-sided tape.
Sticky-paper traps were made from dark green fruit fly adhesive paper (FFAP)
supplied by the Atlantic Paste and Glue Co., Inc. (Brooklyn, NY). The efficacy of the
adhesive paper was determined in laboratory tests of the retention of alighting flies.
Flies were obtained as mature larvae from field-collected papaya fruit in Dade
County, Florida. Larvae exited the fruit and pupated in vermiculite (Landolt & Heath
1988). In tests conducted in a greenhouse, a piece of FFAP (12 by 12 cm) was placed
in a screen cage (30 by 30 by 30 cm) containing 30 sexually mature females. Males
were not tested because females are the primary target of the traps. An observer
counted the number of fly landings on the sticky paper over a one-h time period. Total


Figure 1. Illustration of three types of papaya fruit fly traps used in field studies
conducted in North and Central America. A) Solid styrofoam spheres were painted
with dark green paint and coated with paintable adhesive to retain flies. B) Opaque
cylindrical traps were constructed using green plastic. This trap either contained tox
icant panels or was coated with paintable adhesive to retain attracted flies. C) Sticky
paper cylindrical traps were similar to the opaque cylindrical traps, but the trap body
was made with dark green fruit fly adhesive paper. The adhesive paper was supplied
with protective paper (shown as white overlay) that was removed when traps were
placed in the field to expose the sticky trap surface.

Heath et al.: Papaya Fruit Fly Trapping System

numbers of flies remaining on the paper after the one-h time period were counted, and
ratio of landed to captured flies was determined. The test was replicated three times
over a period of three days.
The sticky-paper traps (Fig. Ic) were prepared similarly to the green opaque traps.
A rain guard was used with this trap to protect the paper trap body; it was made from
the top half of a 150 x 15 mm petri dish (P/N# 1058, Becton Dickinson, Lincoln Park,
NJ). A 5 cm length of polyvinyl chloride tubing (9.0 cm outside diam, Hughs Supply,
Gainesville, FL) was glued to the center underside of the petri dish, which provided a
holder for the trap body. The protective paper supplied with the adhesive paper was
removed when traps were placed in the field to expose the sticky trap surface.

Field Tests

A study was conducted at a papaya plantation on the grounds of the Centro de De
sarrollo de Productos Bi6ticos (CEPROBI) of the Instituto Polit6cnico Nacional (IPN),
Morelos, Mexico, from November 1994 through January 1995, the dry season during
the coolest time of the year. The native vegetation adjacent to the study site is classic
fled as "selva baja caducifolia" or lowland deciduous forest (Soria 1985). Gonolobus so
rodius (Asclepiadaceae) and Jacaratia mexicana (Caricaceae), which are native hosts
for the papaya fruit fly (Castrejon-Ayala & Camino-Lavin 1991), occur in the area.
The papaya plantation had two rows of papaya that serve as a trap crop and are sep
arated by 10 m from the main plot. The experimental plot was 177 by 63 m, with pa
paya trees planted every 3 m. Treatments consisted of four trap types: 1) sticky-coated
green spheres, 2) sticky-coated green opaque cylinders, 3) green opaque cylinders
with internally-placed toxicant panels, and 4) sticky-paper cylinders made with the
dark green FFAP material. These traps were either baited with membrane-based
pheromone lures, which released about 1 pg 2,6-MVP per h, or were unbaited. Lures
were taped to the inside of the opaque cylinders or attached to a plastic rain tent of
sticky-paper traps using double-sided sticky tape. Traps were placed near the imma
ture fruit within the papaya tree because papaya fruit fly females successfully ovi
posit in those fruit (Landolt 1985). The eight treatments (four trap types times two
bait treatments) were placed randomly within a block and treatments were moved se
quentially at time of sampling. There were six blocks placed in the trap crop, i.e.,
around the periphery of the papaya orchard. Traps were placed in every other tree
within a block, and there were 8-19 papaya trees without traps between the experi
mental blocks. The sticky-paper trap bodies were replaced weekly, the sticky-coated
traps were cleaned and recoated as needed. Pheromone lures were replaced after four
weeks of field use. The numbers of males and females captured per trap were recorded
weekly for six weeks (replicates) and then after two weeks for the final sample, for a
total of seven replicates.
Similar field tests were conducted in a papaya orchard at finca Cuilapa, Guate
mala. However, liquid protein-baited McPhail traps, bell-shaped glass traps with a
water reservoir (Newell 1936), were added because of local interest in using this trap
for papaya fruit flies. McPhail traps were baited with five torula yeast-borax pellets
(ERA Int., Freeport, NY) in 300 ml of water (Gilbert et al. 1984). Traps were placed in
papaya trees located around the periphery of the papaya orchard, with individual
traps placed near the smaller fruit within the tree, as above. All nine treatments (four
trap types, two bait treatments, plus McPhail traps) were placed randomly within a
block and treatments were moved sequentially at time of sampling. There were four
blocks placed in the periphery of the papaya orchard. There were at least ten papaya
trees without traps that separated the experimental blocks, and traps within a block

Florida Entomologist 79(1)

were placed in every third tree. Traps were sampled weekly Sticky-coated traps were
cleaned and recoated as needed, sticky-paper trap bodies were replaced weekly. Pher
omone lures were replaced after four weeks of field use. McPhail traps were cleaned
and protein solutions were replaced every two weeks. The numbers of males and fe
males captured per trap were recorded weekly for eight weeks, for a total of eight rep
Statistical Analysis. The total number of trapped flies per treatment was deter
mined from the sum of each sex collected per replicate. Thus, one replicate consisted
of the sum total captured in six traps (Mexico) or four traps (Guatemala) per trap
type. Sum total was converted to percentage trapped within replicate for statistical
analysis. Data were analyzed with two-way analysis of variance (ANOVA) with inter
action using Proc GLM (SAS Institute 1985) followed by LSD mean separation tests
(P = 0.05). Factors used in the ANOVA were bait (2 levels: pheromone baited or no
pheromone; data from McPhail traps were not included in this analysis) and trap type
(4 levels: sticky sphere, sticky green opaque, green opaque with toxicant or sticky-pa
per trap). One-way ANOVAs were conducted to test all nine trap type/bait combine
tions in the Guatemalan tests. Data were log-transformed (x + 1) to stabilize the
variance prior to analysis (Box et al. 1978). Sex of the trapped flies was considered to
be a response variable, not a treatment factor, therefore separate analyses were con
ducted for females, males and total (females plus males) papaya fruit flies captured in
each country.


Lures loaded with 5, 15, 25, and 50 ml released an average ( std) of 0.12 0.01,
0.36 + 0.02, 0.58 0.02 and 1.12 0.11 gg per h during the 23 days that release rates
were obtained (n= 16, 16, 16 and 15, respectively). The projected half lives of the lures
were 67, 184, 300 and 48 days, respectively. Lures prepared with 5, 15, or 25 Pl of ma
trial showed very little decrease during the time that release rates were measured
(Fig. 2). To cover the optimum range of pheromone release, we selected the 50 Pl
loaded pheromone lure, which released an average of about 1 gg per hr, for subse
quent field experiments. There were no significant differences in catch (average std)
on sticky spheres baited with capillaries or the membrane-based lure of either males
(2.0 3.4 versus 2.5 + 0.6) or females (7.8 + 4.5 versus 8.3 3.3).
Number of papaya fruit flies captured in Mexico was low, with an average of 15
flies captured among all six blocks (48 traps total) per replicate. Trap type signifi
cantly affected capture of females (F= 7.41; df 3, 48; P= 0.0004) and total flies (F=
3.79; df 3, 48; P= 0.0162), but not males (F= 1.39; df 3,48; P 0.26). A higher per
centage of females was captured on the green opaque traps with toxicant and on the
sticky-paper traps than on the traps that used paintable sticky coating (sphere-sticky,
opaque-sticky). Presence of the pheromone lure did not affect capture of the papaya
fruit flies, although pheromone-baited traps usually captured slightly more flies than
their unbaited counterparts (Fig. 3).
Capture of papaya fruit flies was also low in the tests conducted in Guatemala,
with an average of 17 flies captured among all four blocks (36 traps total) for the first
5 weeks of the study. Numbers trapped dropped to 0-3 for the last 3 weeks, therefore
these data were deleted and only the first five replicates were included in the analy
ses. Presence of pheromone lure and trap type significantly affected capture of fe
males, males and totals, and there was no interaction between these factors.
Percentage of papaya fruit flies captured in pheromone-baited traps was always
higher than capture in the same traps without pheromone (Fig. 4). Percentage of pa

March, 1996

Heath et al.: Papaya Fruit Fly Trapping System

paya fruit flies captured in the liquid protein baited McPhail traps was no higher than
that in the least effective traps tested. No females were captured in the sticky-coated
green opaque traps, and percentage captured in these traps was lower than in the
other three trap types. For both males and total flies, percentage captures in the
sticky-paper traps and the sticky-coated spheres were higher than in either of the
green opaque traps.
In laboratory trials, the FFAP material used for the sticky-paper traps was highly
effective in retaining flies that landed on the surface. In the three tests, nine, six and
four flies landed on the paper and all (19 out of 19) were captured on the sticky sur


These studies demonstrated that a membrane-based lure can be used to formulate
synthetic papaya fruit fly sex pheromone for field use, and that cylindrical traps may
be used to replace the sphere traps used previously. The cylindrical traps apparently
provided a sufficient visual cue because the unbaited cylindrical traps also captured
papaya fruit flies. Landolt & Heath (1990) found that although mated females re
sponded to the visual aspects of the spheres alone, virgin females represented most of
the females captured in response to pheromone. Number of males also increased di
rectly with pheromone dose, indicating that males responded to the pheromone pres
ence. It is not known why there was no effect of pheromone in the trials in Mexico as


- 1




-H- 50 pl
--*-- 25 pl
-0" 15 l1
-U- 5 pl


Figure 2. Average release rates (pg per h) of papaya fruit fly synthetic pheromone
which was formulated in membrane-based lures, over time. Regressions were deter
mined from lures (n=3) containing 5, 15, 25 and 50 il of 2 methyl 6 vinylpyrazine.

I L -

Florida Entomologist 79(1)




>^ 0

I- 30
C. 20

0 1o


0 30


b b

a al



Total (Females plus Males)

Sphere-sticky Opaque-sticky Opaque-toxicant FFAP-sticky

Figure 3. Results of trap capture of female (top), male (middle) and total (bottom)
papaya fruit flies captured in field trials conducted in Mexico (n 7). Traps tested (left
to right) were green sticky spheres, green opaque cylinders with sticky exterior, green
opaque cylinders with internally-placed toxicant, and green sticky-paper cylinders.
Traps were either baited with papaya fruit fly pheromone membrane-based lures
(shaded bars) or left unbaited (open bars). Pairs of bars within a graph headed by the
same letter are not significantly different (LSD mean separation test on log (x + 1)
transformed data, P= 0.05; non-transformed means presented).


March, 1996

Heath et al.: Papaya Fruit Fly Trapping System




ab b ab

ab ab
a a

Total (Females plus Males)


cc bc b

b ab
a ab


a 40

- 30
2 20
Q 10
~ 0


< 30




McPhail FFAP-sticky

Figure 4. Results of trap capture of female (top), male (middle) and total (bottom)
papaya fruit flies captured in field trials conducted in Guatemala (n= 5). Traps tested
(left to right) were green sticky spheres, green opaque cylinders with sticky exterior,
McPhail traps, green opaque cylinders with internally-placed toxicant, and green
sticky-paper cylinders. Except for the McPhail traps, which were baited with aqueous
torula yeast plus borax solution, the traps were either baited with papaya fruit fly
pheromone membrane-based lures (shaded bars) or left unbaited (open bars). Bars
within a graph headed by the same letter are not significantly different [LSD mean
separation test on log (x+ 1) transformed data, P= 0.05; non-transformed means pre


Florida Entomologist 79(1)

was observed in the trials in Guatemala. A number of factors, including the demo
graphics of the fly population (e.g. presence or absence of virgin females in the popu
nation) and environmental parameters at each site, may have affected the results. The
mated status of the females captured was not determined in our studies.
The membrane-based lure can be formulated to provide a multitude of release
rates, and release rates were directly related to amount of 2,6-MVP added to the lure.
The lure used in trials reported herein was based on optimal release rate for sticky
coated spheres (Landolt & Heath 1990). It is not known if this release rate is optimal
for traps other than sticky spheres. A lower release rate may improve catch in the
opaque cylindrical traps containing the toxicant panel because this trap requires that
the flies enter the trap to be captured. Additional studies are needed to examine a
range of pheromone doses for use in cylindrical traps that use toxicant or sticky ma
The use of the FFAP material provided a facile method to prepare cylindrical
traps. This material was easily fabricated and the sticky material did not adhere to
personnel who contacted the adhesive. Exposure over time indicated that it was im
pervious to rain when the material was used as described. The adhesive material did
not drip or run as was observed when paintable sticky coating was used. It should be
noted, however, that in experiments conducted with sticky-paper traps for other te
phritid pests in southern Florida, we observed that on occasion small birds and liz
ards were either blown into the trap or came in contact with the FFAP material. While
the sticky-paper trap caught non-target insects in studies in both Mexico and Guate
mala, we did not observe the capture of small animals such as was observed in Flor
ida. The captured insects and animals can be remove from the FFAP material with
the use of mineral oil. This process had no observed detrimental effect on the animals
if the animals were removed shortly after capture. Another problem with this, as well
as other sticky-coated traps, is that wind-borne dust or debris may coat the trap and
limit the longevity of the adhesive in the field. The green opaque trap used with the
toxicant provided an alternative to a sticky trap, although it may not be as sensitive
under all environmental conditions.
Papaya fruit flies were captured in protein-baited McPhail traps in this study in
Guatemala, but in low numbers. This suggests that volatiles from protein baits may
be attractive to papaya fruit flies. If so, addition of protein bait volatiles may provide
synergists that could further improve trap capture of the pheromone-baited traps. It
is envisioned that the information presented here may provide a facile method to
monitor papaya fruit fly infestations. This would result in decreased pesticide appli
cation and, potentially, offer a method of control through trapping alone.


We thank Charles Powell (USDA/ARS, Gainesville, FL), Sanjay Kulkarni, John
Howell, Yasmin Cardoza (Univ. of Florida), Vicot Castrej6n and Jaime Pinero (CEP
ROBI, Mexico) for their assistance; D. Martinez (Laboratorio de Botanica, Univer
sidad Aut6noma del Estado de Morelos, Cuernavaca, Mexico) for plant identification;
and Peter E. A. Teal (USDA/ARS, Gainesville, FL), Jennifer L. Sharp (USDA/ARS, Mi
ami, FL), Timothy C. Holler (USDA/APHIS, Gainesville, FL) and two anonymous re-
viewers for critical review and helpful comments on the manuscript. The authors
thank Michael Bentivegna, Jr. and the Atlantic Paste and Glue Co., Inc. (Brooklyn,
NY), for providing the fruit fly adhesive paper; Janice Gillespie, Edie Christensen and
Consep Membranes, Inc. (Bend, OR) for providing materials for the lures. We thank

March, 1996

Heath et al.: Papaya Fruit Fly Trapping System

Gordon Tween and the staff at the USDA, APHIS-International Services, U.S. Em
bassy, Guatemala and Pedro Rendon and the staff at the USDA, APHIS-PPQ, Guate
mala for their support of this research. Partial financial support was provided by the
International Foundation for Science (Grant C/1741-1), CEPROBI and Instituto de
Ecologia, A. C. (Proyecto Ecologia y Comportamiento Animal). This article reports the
results of research only. Mention of a proprietary product does not constitute an en
dorsement or recommendation for its use by USDA.


Box, G. E. P., W. G. HUNTER, AND J. S. HUNTER 1978. Statistics for experimenters. An
introduction to design, data analysis, and model building. J. Wiley & Sons, New
York, New York.
CASTREJON-AYALA, F., AND M. CAMINO-LAVIN. 1991. New host plant record for Toxot
rypana curvicauda (Diptera: Tephritidae). Florida Entomol. 74: 466.
CHUMAN, T., P. J. LANDOLT, R. R. HEATH, AND J. H. TUMLINSON. 1987. Isolation, iden
tification, and synthesis of male-produced sex pheromone of papaya fruit fly,
Toxotrypana curvicauda Gerstaecker (Diptera: Tephritidae). J. Chem. Ecol. 13:
GILBERT, A. J., R. R. BINGHAM, M. A. NICOLAS, AND R. A. CLARK. 1984. Insect trapping
guide. Pest Ii. i ,. I -- 11. i .. Projects, State of California Department of
Food and Agriculture, Sacramento, California.
HARRIS, E. J., S. NAKAGAWA, AND T. URAGO. 1971. Sticky traps for detection and sur-
vey of three tephritids. J. Econ. Entomol. 64: 6265.
HEATH, R. R., AND A. MANUKIAN. 1992. Development and evaluation of systems to col
lect volatile semiochemicals from insects and plants using a charcoal-infused
medium for air purification. J. Chem. Ecol. 18: 1209-1226.
HEATH, R. R., N. D. EPSKY, P. J. LANDOLT, AND J. SIVINSKI. 1993. Development of at
tractants for monitoring Caribbean fruit flies (Diptera: Tephritidae). Florida
Entomol. 76: 233-244.
MEYER. 1995. Development of a dry plastic insect trap with food-based syn
thetic attractant for the Mediterranean and the Mexican fruit fly (Diptera: Te
phritidae). J. Econ. Entomol. 88: 1307-1315.
LANDOLT, P. J. 1985. Papaya fruit fly eggs and larvae (Diptera: Tephritidae) in field
collected papaya fruit. Florida Entomol. 68: 354-356.
LANDOLT, P. J., AND R. R. HEATH. 1988. Effects of age, mating, and time of day on be
havioral responses of female papaya fruit fly, Toxotrypana curvicauda Gers
taecker (Diptera: Tephritidae) on synthetic sex pheromone. Environ. Entomol.
17: 4751.
LANDOLT, P. J., AND R. R. HEATH. 1990. Effects of pheromone release rate and time of
day on catches of male and female papaya fruit flies (Diptera: Tephritidae) on
fruit model traps baited with pheromone. J. Econ. Entomol. 85: 2040-2043.
LANDOLT, P. J., AND J. HENDRICHS. 1983. Reproductive behavior of the papaya fruit
fly, Toxotrypana curvicauda Gerstaecker (Diptera: Tephritidae). Ann. Entomol.
Soc. America 76: 413-417.
LANDOLT, P. J., R. R. HEATH, AND J. R. KING. 1985. Behavioral responses of female pa
paya fruit flies, Toxotrypana curvicauda Gerstaecker (Diptera: Tephritidae) to
male-produced sex pheromone. Ann. Entomol. Soc. America 78: 751-755.
A sex pheromone-based trapping system for the papaya fruit fly, Toxotrypana
curvicauda Gerstaecker (Diptera: Tephritidae). J. Econ. Entomol. 81: 1163
NEWELL, W. 1936. Progress report on the Key West (Florida) fruit fly eradication
project. J. Econ. Entomol. 29: 116-120.

48 Florida Entomologist 79(1) March, 1996

SAS INSTITUTE. 1985. SAS/STAT guide for personal computers, version 6 edition. SAS
Institute, Cary, North Carolina.
SORIA, R. G. 1985. Flora de Morelos. Descripcidn de species vegetables de la selva baja
caducifolia del Canon de Lobos, Mpio. de Yautepec; Serie Ciencias Naturales y
de la Salud. Program Floristico-Faunistico. Universidad Autdnoma del Estado
de Morelos. Cuernavaca, Morelos, Mexico. 165 p.
WOLCOTT, G. N. 1933. An economic entomology of the West Indies. Clay and Sons,
Bungay, Suffolk.


48 Florida Entomologist 79(1) March, 1996


Department of Entomology, University of Georgia
Athens, GA 30602-2603


The fall armyworm, Spodoptera frugiperda (J. E. Smith), includes morphologically
indistinguishable corn and rice strains. The two strains were surveyed for diagnostic
restriction patterns in mitochondrial DNA (mtDNA) using 25 restriction endonu
cleases. Polymorphic mtDNA restriction patterns were identified for BstNI, HinfI and
Mspl. The Mspl pattern was the most distinctive since the molecular size of each DNA
fragment differed between the two strains. Analyses of laboratory and field-collected
insects showed the Mspl mtDNA pattern to be a diagnostic marker for corn and rice
strain insects. Strain identification by the Mspl mtDNA profile correlated exactly
with nuclear DNA markers. Since no HaeIII sites are present in fall armyworm
mtDNA, a double-digest of total fall armyworm DNA using HaeIII and Mspl allowed
the direct detection of mtDNA restriction fragments from total DNA on a stained aga
rose gel. In contrast to conventional techniques utilizing mtDNA markers, this rapid
and simple procedure does not require the isolation of mtDNA, or avoids the use of
DNA blots and labeled mtDNA.

Key Words: Spodoptera, fall armyworm, mitochondrial DNA, strain identification.


El gusano trozador, Spodoptera frugiperda (J. E. Smith), posee cepas de maiz y
arroz morfoldgicamente indistiguibles. Esas dos cepas fueron muestreadas para el
diagndstico mediante patrons de restriccidn en DNA mitocondrial (mtDNA), utili
zando 25 endonucleasas de restriccidn. Los patrons polimdrficos del mtDNA de res
triccidn fueron identificados para BstNI, Hinfl y Mspl. El patron Mspl fue el mas
distintivo debido a que el tamano molecular de cada fragmento de DNA difiri6 entire
las dos cepas. Los analisis de insects de laboratorio y colectados en el campo mostra
ron que el patron del Mspl mtDNA es un marcador diagndstico para los insects de las
cepas de maiz y arroz. La identificacidn del perfil del Mspl mtDNA se correlacion6
exactamente con los marcadores nucleares de DNA. Debido a la ausencia de sitios
HaeIII en el mtDNA del gusano trozador, una double digestion del total del DNA
usando HaeIII y Mspl permiti6 la deteccidn de los fragments de restriccidn del

Lu & Adang: Fall Armyworm Strain Identification

mtDNA a partir del DNA total en un gel de agarosa tenido. En contrast con las tc-
nicas convencionales que utilizan marcadores de mtDNA, este process rapido y simple
no require del aislamiento del mtDNA, o evita el uso de blots de DNA y de mtDNA

The fall armyworm, Spodoptera frugiperda (J. E. Smith), is a major pest on corn
(Zea mays L.), sorghum (Sorghum vulgare pers.) and bermudagrass (Cynodon dacty
ion pers.) in the southeastern United States. The insect is an occasional pest on many
other crops, including cotton (Gossypium hirsutum L.), peanut (Arachis hypogea L.),
millet (Pennisetum glaucum Pers.), alfalfa (Medicago sativa L.), rye (Secale cereale
L.), rice (Oryza sativa L.) and soybean (Glycine maxMerr.) (Sparks 1979). Difficulties
in the control of fall armyworms have been attributed to its wide range of host plants,
vast geographical distribution, and rapid and long distance movement (Knipling
Pashley (1986) discovered by allozyme analyses that fall armyworm populations
consist of two host strains. The corn-strain prefers corn, sorghum and cotton, while
the rice-strain prefers rice and bermudagrass. Subsequent investigations of the nu
clear and mitochondrial genomes using restriction fragment length polymorphism
(RFLP) techniques revealed significant differences between the two strains (Pashley
1989, Lu et al. 1992). Further evidence for the genetic separation of the strains was
the discovery of a unique repeated DNA sequence in the genome of rice strain insects
(Lu et al. 1994). Reproductive isolation mechanisms exist between the two fall army
worm strains (Pashley et al. 1987; Pashley et al. 1992). Genetic separation and barri
ers to interbreeding support the species status for the fall armyworm strains (Pashley
et al. 1992).
The practical impact of the sympatric strains is not clear, but several studies sug
gest the strain diversity complicates pest management. Quisenberry & Whitford
(1988) demonstrated that bermudagrass bred for resistance to the corn strain insects
was still susceptible to rice strain. As efforts are directed towards developing fall ar
myworm-resistant corn (Wiseman & Davis 1979, Williams et al. 1989, Wiseman &
Isenhour 1988), the use of characterized fall armyworm strains may be crucial. Other
biological differences between the strains, including dispersal pattern (Pashley et al.
1992) and response to pesticides (Pashley et al. 1988), may influence population mon
itoring studies (Barfield et al. 1980, Pair et al. 1986) and pest control strategies.
The major objective of this study was to identify restriction enzyme patterns for
mtDNA that distinguish the corn and rice strain insects. Pashley (1989) reported dif
ferences in the number of restriction fragments for several restriction enzymes. We
tested additional enzymes and report that Mspl digested mtDNA reveals a diagnostic
pattern for the strains. This Mspl pattern is directly detected on agarose gels when a
second endonuclease HaeIII is included to digest nuclear DNA. Due to its simplicity
this Mspl/HaeIII co-digestion method will be useful in the identification of fall army
worm strains.


Insect Sources

Both laboratory-reared and field-collected fall armyworms were used. Sources de
scribed in Lu et al. (1992) include the corn strain (designated as C) and the rice strain

Florida Entomologist 79(1)

(R) colonies from Louisiana State University (a gift from S. Quisenberry); a colony (M)
from the USDA-ARS laboratory at Mississippi State University (a gift from F Davis);
and colony (I) formerly maintained by D. Isenhour at the University of Georgia,
Coastal Plains Experimental Station. Population P was collected from corn plants in
Tifton, Georgia, in 1990. Other insect sources included a laboratory colony estab
lished with bermudagrass-collected fall armyworms (provided by R. Mcpherson,
USDA Tifton, Georgia) and fall armyworms collected from 2 sorghum and 5 corn fields
near Athens, Georgia in 1992.

Total DNA Isolation

Total DNA was isolated from individual fifth instar larvae as described in Lu et al.
(1992). Isolated DNA samples in TE buffer (10 mMTris, 1 mMEDTA, pH 8.0) were
stored at 20'C until needed.

Mitochondrial DNA Isolation and 32Plabeling

Fifteen larvae (4.2 g) were homogenized with about 20 strokes in a Dounce tissue
homogenizer with 20 ml of homogenization buffer [200 mMmannitol, 70 mMsucrose,
500 mMTris-HC1 pH 7.5, 100 mMEDTA, and 0.2 mg/ml proteinase K (Sigma, St.
Louis, MO) (Zehnder et al. 1992)]. The homogenate was centrifuged at 1,600 x g for 10
min to pellet the cellular debris. The supernatant was centrifuged again at 17,000 x
g for 30 min to pellet mitochondria. The mitochondrial pellet was washed once by re
suspending the pellet with 20 ml of homogenization buffer, centrifuged at 1,600 x g for
10 min, and the supernatant decanted and centrifuged at 40,000 x g for 15 min. The
final mitochondrial pellet was resuspended in 3 ml of buffer containing 100 mMNaC1,
500 mMTris-HC1 pH 8.0, 10 mMEDTA pH 8.0, and then mixed with 0.375 ml of 10%
sodium dodecyl sulfate (SDS). To the above mixture, 6.8 g CsCl and 0.25 ml of ethid
ium bromide solution (10 mg/ml) were added, and the total volume was adjusted to
7.15 ml with TE buffer. The mixture was centrifuged at 161,000 x g for 20 h. The DNA
bands were visualized under UV light and the lower mtDNA band was removed. Iso
lated mtDNA was extracted with 1 butanol to remove ethidium bromide according to
procedures described (Sambrook et al. 1989). Purified mtDNA was labeled with 32P
dCTP using the random primer procedure (Feinberg & Vogelstein 1984), and labeled
mtDNA was used as a probe to hybridize with restriction digested total DNA.

Restriction Digestion and Electrophoresis

Complete digestion of total DNA was carried out by using 2 units of enzyme (Boe
hringer-Mannheim Laboratories, Indianapolis, Ind.) per gg of DNA in the supplied
buffer at 37'C for 16 h. In the case of double digestion, the same enzyme/DNA ratio
was used for each enzyme. Digested DNA was electrophoretically fractionated in a 1%
agarose gel (Bio-Rad, Richmond, Calif) in TAE buffer (40 mM Tris-acetate, 1 mM
EDTA, pH 7.5). After electrophoresis, the agarose gel was stained with ethidium bro
mide and DNA visualized under UV.

Southern Blot Procedures

Following electrophoresis, DNA was transferred to nylon membranes (GeneScreen
Plus, DuPont) by the method of Southern (1975). DNA was immobilized on mem
branes by baking at 80C for 2 h.

March, 1996

Lu & Adang: Fall Armyworm Strain Identification

Hybridizations were carried out as follows: membranes were prehybridized with
sheared salmon sperm DNA (100 g/ml) in hybridization buffer consisting of 3X SSC
(1X SSC 0.15 MNaCI, 0.015 MNa citrate) containing 0.1% SDS at 65'C for 6 h. La
beled mtDNA was denatured by boiling for 20 min and then added to the prehybrid
ized filter in hybridization buffer. Hybridization was for at least 6 h at 65'C. Filters
were washed once in 2X SSC, 0.1% SDS for 20 min and twice in 1X SSC, 0.1% SDS for
20 min at 65'C. Filters were air-dried and exposed to X-ray film (Kodak) at 80C with
intensifying screens.


We screened for polymorphic mtDNA restriction patterns between the corn and
rice strains of fall armyworm by digesting total DNA from both strains with 25 re
striction enzymes followed by probing with 32P-labeled mtDNA. The preliminary
screening data is not shown. Among the restriction enzymes tested, a single cleavage
site was observed for EcoRV, HincII, Pvull, Sall and Scal. The size of fall armyworm
mtDNA was estimated as approximately 14.8 kb. This value is slightly smaller than
the 16.3 kb size estimated by Ke & Pashley (1992). BamHI, HaeIII, Hpal and SmaI
did not digest fall armyworm mtDNA. Restriction enzymes having 2 or more cleavage
sites included Alul, BstNI, Dral, EcoRI, FokI, HhaII, Hinfl, HindIII, HpaII, MspI,
Ndel, Nrul, PstI, Rsal, Taql and Xbal. Of the 16 restriction enzymes that generated
more than two mtDNA fragments, only three enzymes (BstNI, HinIl and Mspl) pro
duced restriction profiles that differed between the corn and rice strains (Fig. 1). The
Mspl and Hpall patterns are the same because the two enzymes share the same rec
ognition site (designated the Mspl pattern). The BstNI and HinIl mtDNA patterns
were previously described by Pashley (1989), while the Mspl pattern is first described
in this study. The Mspl pattern is very distinctive between the two strains, with the
corn strain pattern consisting of 4 restriction fragments of 5.4, 4.3, 3.8 and 1.3 kb, and
the rice strain pattern consisting of 2 fragments of 10.4 and 4.4 kb (Fig. 1).
Further, we explored the Mspl mtDNA restriction pattern as a diagnostic marker
for corn and rice strain insects. We determined the Mspl mtDNA patterns for fall ar
myworms from five populations identified previously as rice (R) or corn (C, I, M, and
P) strain populations (Lu et al. 1992). Fifteen to 20 individual larvae were selected
from each population and total DNA extracted. Fig. 2 shows a Southern blot of repre
tentative samples of Mspl digested DNA probed with 32P-labeled mtDNA. Each insect
belonged to one of two mtDNA haplotypes; one represented by the corn strain and the
other by the rice strain (Fig. 2). We also examined the mtDNA from insects of un
known strain status. Larvae were collected from corn and sorghum fields, and from a
laboratory colony initiated from bermudagrass-collected insects. The analyses showed
that 32 of 36 fall armyworms (over 85%) collected from the corn and sorghum fields
had the corn strain Mspl restriction pattern (data not shown). All of the 38 insects
from the bermudagrass-originated colony had the rice strain pattern (data not
shown). These results are consistent with the host preference of strains as reported by
Pashley et al. (1988). Corn strain fall armyworms prefer corn and sorghum, while the
rice strain insects prefer rice and bermudagrass. However, the above result also
showed that a small fraction of rice strain insects can be found on corn and sorghum,
indicating some overlap in host usage occurs (Pashley 1989).
Total DNA preparations from the field-collected and bermudagrass colony insects
were analyzed using a repeated DNA marker found only in rice strain insects (Lu et
al. 1994). All the total DNA samples that reacted with the rice strain-specific marker
showed the rice strain mtDNA pattern, while the remainder displayed the corn strain

Florida Entomologist 79(1)











Figure 1. Polymorphic mtDNA restriction patterns between the fall armyworm
corn (C) and rice (R) strain generated by BstNI, Hinfl and Mspl, and detecting by
probing total DNA blot with 32P-labeled mtDNA. Molecular sizes (bp) are indicated at
the left.

mtDNA pattern (data not shown). These results correlate the Mspl mtDNA pattern
with known nuclear DNA markers and establish the Mspl mtDNA pattern as a diag
nostic marker for fall armyworm host strains.
We felt that the Mspl strain marker would be more useful if the time consuming
and expensive Southern blotting steps could be eliminated. This was accomplished us
ing the following rationale and approach. We observed that fall armyworm mtDNA is
uncut by HaeIII, while genomic DNA is digested by HaeIII to fragments smaller than
4 kb in molecular size (data not shown). Also, after Mspl digestion of total fall army
worm DNA, several mtDNA restriction fragments were partially discernible in an
ethidium bromide stained agarose gel (data not shown). These observations led us to
postulate that the diagnostic Mspl mtDNA pattern might be visible on a stained gel
if the molecular size of the background genomic DNA fragments was reduced. Figure
3 depicts a stained agarose gel of fall armyworm DNA samples after treatment with
Mspl and HaeIII. As expected the genomic DNA fragments migrated further down the
agarose gel and uncovered the mtDNA fragments. The two mtDNA fragments (10.4
and 4.4 kb) of the rice strain pattern, and 3 of the 4 fragments (5.4, 4.3 and 3.8 kb) of
the corn strain pattern were easily detected. The 1.4 kb fragment in the corn strain
pattern was hidden by the bulk genomic DNA fragments in the lower part of the gel.
We have termed this the "3-band pattern" for the corn strain and the "2-band pattern"
for the rice strain.

March, 1996

Lu & Adang: Fall Armyworm Strain Identification 53


*I" V $-ma

""* .

5.1 -
3.1 -
2.1 -

1.1 -


Figure 2. A Southern blot showing the Mspl mtDNA pattern of fall armyworms
from different populations as indicated. Population R is the rice strain. Populations C,
M, I and P were previously identified as corn strain populations (Lu et al. 1992).


The results of a previous study (Pashley 1989) and this study show that the corn
and rice strains of fall armyworm have distinct mtDNA RFLPs for BstNI, Hinfl and
Mspl. Since the diagnostic Mspl mtDNA pattern can be detected by a simple double






Figure 3. A stained agarose gel of Mspl and HaeIII digested total fall armyworm
DNA, showing the diagnostic mtDNA restriction pattern on the upper part of the gel.
Fall armyworm populations and DNA samples are the same as in Fig. 2.

Florida Entomologist 79(1)

digest of total DNA using Mspl and HaeIII, the method does not require the isolation
of mtDNA, or the use of Southern blot analysis. Thus, this very straightforward Mspl/
HaeIII digestion technique facilitates the rapid detection of fall armyworm strains.
Since mtDNA is maternally inherited, a concern is that mtDNA markers may not
correctly identify intraspecific pest populations, particularly when these populations
are sympatrically distributed and interbred. In these situations, insect populations
with certain mtDNA genotypes may simply represent maternal lineages, but not nu
clear genotypes. To address this concern, we analyzed fall armyworms with both
mtDNA and nuclear DNA markers (this study, Lu et al. 1992, Lu et al. 1994). Our re
sults demonstrated that the two types of markers agreed completely, indicating that
fall armyworm mtDNA genotypes correspond to their nuclear genotypes. These re
sults support Pashley's conclusion that gene flow between the sympatric fall army
worm strains is very limited (Pashley 1989).
Fall armyworm strains were shown to be selective in host usage (Pashley 1986),
however, it should not be assumed that all the insects collected from a given host will
belong to the host strain. Pashley (1989) reported finding a small number of corn
strain insects on bermudagrass and rice strain insects on corn and sorghum plants.
By using the mtDNA and nuclear DNA markers, we also identified a small number of
fall armyworms taken from corn and sorghum plants to be rice strain insects. Further
studies are needed to gain insight into the mechanism of strain differentiation and its
impact on pest management.


The authors are grateful to Drs. D. Isenhour, G. Kochert, and M. Bass for their sup
port. This research was funded by a HATCH project through the University of Georgia
College of Agriculture and Environmental Sciences.


BARFIELD, C. S., J. L. STIMAC, AND M. A. KELLER 1980. State-of-art for predicting
damaging infestations of fall armyworm. Florida Entomol. 63: 364-373.
FEINBERG, A. P., AND B. VOGELSTEIN. 1984. A technique for labeling DNA restriction
fragments to a high specific activity. Anal. Biochem. 132: 613.
KE, L. D., AND D. P. PASHLEY. 1992. Characterization of fall armyworm mitochondrial
DNA (Lepidoptera: Noctuidae). Arch. Insect. Bio. Phisil. 21: 263-269.
KNIPLING, E. F. 1980. Regional management of the fall armyworm-A realistic ap
proach? Florida Entomol. 63: 468-479.
Lu, Y. J., M. J. ADANG, D. J. ISENHOUR, AND G. D. KOCHERT. 1992. RFLP analysis of
genetic variation in North American populations of the fall armyworm moth
Spodoptera frugiperda (Lepidoptera: Noctuidae). Molecular Ecology 1: 199
Lu, Y. J., G. D. KOCHERT, D. J. ISENHOUR, AND M. J. ADANG. 1994. Molecular charac
terization of a strain specific repeated DNA sequence in fall armyworm
Spodoptera frugiperda (Lepidoptera: Noctuidae). Insect Molecular Biology. 3.
1986. Fall armyworm distribution and population dynamics in the southeast
ern states. Florida Entomol. 69: 468-487.
PASHLEY, D. P. 1986. Host-associated genetic differentiation in fall armyworm (Lepi
doptera: Noctuidae): A sibling species complex? Ann. Entomol. Soc. America 79:
PASHLEY, D. P. 1989. Host-associated differentiation in armyworms (Lepidoptera:
Noctuidae): An allozymic and mitochondrial DNA perspective, pp. 103-114 in

March, 1996

Lu & Adang: Fall Armyworm Strain Identification

H. D.Loxdale, and J. den Hollander, [ed.], Electrophoretic Studies on Agricul
tural Pests. Clarendon Press, Oxford.
PASHLEY, D. P., A. M. HAMMOND, AND T. N. HARDY. 1992. Reproductive isolating
mechanisms in fall armyworm host strains (Lepidoptera: Noctuidae). Ann. En
tomol. Soc. America. 85: 400-405.
PASHLEY, D. P., AND J. A. MARTIN. 1987. Reproductive incompatibility between host
strains of the fall armyworm (Lepidoptera: Noctuidae). Ann. Entomol. Soc.
America 80: 731-733.
1988. Two fall armyworm strains feed on corn, rice and bermudagrass. Louisi
ana Agriculture 30: 89.
QUISENBERRY S. S., AND F. WHITFORD. 1988. Evaluation of bermudagrass resistance
to fall armyworm (Lepidoptera: Noctuidae): influence of host strain and dietary
conditioning. J. Econ. Entomol. 81: 1463-1468.
SAMBROOK, J., E. F. FRITSCH, AND T. O. MANIATIS. 1989. Molecular cloning: A Labo
ratory Manual. 2nd ed. Cold Spring Harbor Press, New York.
SOUTHERN, E. M. 1975. Detection of specific sequences among DNA fragments sepa
rated by gel electrophoresis. J. Mol. Biol. 98:503-517.
SPARKS, A. N. 1979. A review of the biology of the fall armyworm. Florida Entomol. 62:
WILLIAMS, W. P., P. M. BUCKLEY, AND F. M. DAVIS. 1989. Combining ability for resis
tance in corn to fall Armyworm and southwestern corn borer. Crop Sci. 29: 913
WISEMAN, B. R., AND F. M. DAVIS. 1979. Plant resistance to the fall armyworm. Flor
ida Entomol. 62: 123-130.
WISEMAN, B. R., AND D. J. ISENHOUR 1988. Feeding responses of fall armyworm lar
vae on excised green and yellow whorl tissue of resistant and susceptible corn.
Florida Entomol. 71: 243-249.
ZEHNDER, G. W., L. SANDALL, A. M. TISLER, AND T. O. POWERS. 1992. Mitochondrial
DNA diversity among 17 geographic populations of Leptinotarsa decemlineata
(Coleoptera: Chrysomelidae). Ann. Entomol. Soc. America. 85: 234-240.

Florida Entomologist 79(1)


'Departamento de Zoologia. Facultad de Biologia. Universidad de Salamanca.
37071-Salamanca (Spain)

2Museo Nacional de Ciencias Naturales. C/J. Gutierrez Abascal,
2. 28006-Madrid (Spain)


The mature larvae of Chrysis gracillima Forster, 1853 and Omalus biaccinctus
(Buysson, 1893) are described and compared with others known in the tribes Chrysi
dini and Elampini, respectively. Additionally, new data are reported on the biology of
Trichrysis cyanea (L., 1758).

Key Words: Cleptoparasites, biology, preimaginal states.


Se described las larvas maduras de Chrysis gracillima Forster, 1853 y Omalus
biaccinctus (Buysson, 1893) y se comparan con otras conocidas de las tribus Chrysi
dini y Elampini, respectivamente. Adicionalmente, son reportados nuevos datos sobre
la biologia de Trichrysis cyanea (L., 1758).

The family Chrysididae includes some 3,000 species (Kimsey & Bohart 1990), most
of which are ectoparasitoids or cleptoparasites of Hymenoptera Aculeata. Their biol
ogy is not very well known, most contributions referring to isolated data on their pos
sible hosts. However, particularly outstanding are the workers in the case of the
Chrysididae that attack stem-nesting Sphecidae and Eumenidae: Trautmann (1927),
Enslin (1929), Clausen (1940), Van Lith (1953, 1955, 1956, 1958), Grandi (1959,
1961), Krombein (1967), Danks (1970), Spradbery (1973), Iwata (1976), and
Medvedev (1978).
The mature larvae of 15 Chrysidinae have been described. These species belong to
eight genera, five of which are Chrysidini and three are Elampini. Studies on the ma
ture larvae of Palearctic Chrysis ignita (L., 1758); Chrysura dichroa (Dahlbom, 1854);
Hedychrum rutilans Dahlbom, 1854; Praestochrysis shanghaiensis (Smith, 1874);
Trichrysis cyanea (L., 1758); or Holarctic Omalus aeneus (F., 1758) and Pseudomalus
auratus (L., 1758)), have been carried out by: Enslin (1929); Mar6chal (1923); Soika
(1934); Maneval (1936); Parker (1936); Grandi (1959, 1961); Yamane (1976); Evans
(1987) and Asis et al. (1994).
Janvier (1933) described the mature larva of a Neotropical species (Chrysis gran
dis (Brull6, 1846)), and Evans (1987) dealt with the major morphological features of
all the above mentioned species (with the exception of Chrysis ignita, Hedychrum ru
tilans and T cyanea), the Nearctic species Chrysis cembricola Krombein, 1958; Chry

March, 1996

Tormos et al.: Larvae of Chrysididae

sis cessata Buysson, 1891; Chrysis nitidula F., 1775; Chrysis inflata Aaron, 1885;
Chrysis inaequidens Dahlbom, 1854, and the Nearctic-Neotropical species Caenochry
sis doriae (Gribodo, 1874) and Chrysis smaragdula F, 1775.
In the course of a study on the fauna of stem-nesting Hymenoptera on the north
ern subplateau of the Iberian Peninsula, data were obtained on the biology of T cya
nea. Mature larvae of Chrysis gracillima F6rster, 1853 and Omalus biaccinctus
(Buysson, 1893) were also collected.
For Omalus Panzer, Pemphredoninae (Sphecidae) are accepted as hosts (Kimsey &
Bohart 1990). For hosts of T cyanea, these authors cite Larrinae and Pemphredoninae
(Sphecidae). The mature larvae of C. gracillima and 0. biaccinctus have been previ
ously undescribed.


A total of 2,043 trap nests was placed in different sites of the provinces of Burgos,
Cuenca, Soria, Teruel and Valencia (Spain). The trap nests, comprising stems of Ail
anthus altissima (Miller) Swingle and of Phragmites australis (Cav) Trin ex Steudel
(length=20-30 cm, diam=2-5 mm), were placed in the field in mid spring of 1992 and
1993 and were collected at the end of autumn of the same year. After opening them in
the laboratory, the contents of each cell were transferred to glass vials that were kept
at 6-8 C over the winter. During the following spring the vials were transferred to a
culture chamber at 28 C to elicit emergence of the images; it was thus possible to
know the identity of the occupants of the nests and their parasitoids.
Some of the mature larvae were fixed and preserved in 70% alcohol for later study
and description. The methodology used in their preparation was similar to that em
played by Asis et al. (1994). The descriptions employ the terminology and organization
used by Evans (1987).
The cells were numbered from the exterior to the interior (cell 1 outermost) even
though chronologically the innermost cell was the one first completed.


In all of the nests examined by us that contained two or more parasitized cells, the
parasites were always of the same species.

Chrysis gracillima Forster

Three larvae were collected from a nest with three cells, all of them parasitized. Of
these, two were reared to obtain images while the other one was placed in 70% alcohol
for the description that follows. The absolute measurements refer to the specimen
from Vilviestre (Soria) (reference: 93010101).
Mature larva. General aspect (Fig. 1). Body robust (length = 4.3 Im, width = 1.9
tm), with the abdominal segments divided into two annulets by a transverse crease.
Anus small, terminal, as a transverse slit. Pleural lobes developed. Integument with
eighteen minute setae (1 12 pm) on each segment, distributed in a line between spi
racles (Fig. 2).
Spiracles (Fig. 3). With a well-developed peritreme; atrium simple, naked. First
pair (mean diam = 42 pm, n=10) slightly larger than the rest (mean diam = 35.7 pm,
n 18).
Head (Fig. 4). Width = 838 mm, height = 600 pIm, with sparse and short setae (1
19 pm) very numerous near insertion of mandibles. Coronal suture and parietal

Florida Entomologist 79(1)

March, 1996

0.5 mm

2 mm


0.1 mm


I.2 mm
i \

0.2 mm

Figures 1-6. Mature larva of C. gracillima F6rster: 1, general aspect; 2, segment
dorsumm) with the distribution of setae; 3, spiracle (atrium and subatrium); 4, head in
frontal view; 5, labrum; 6, labium and maxilla (oral face). Mature larva of Omalus bi
accinctus (Buysson).

bands absent. Antennal orbits elliptical (57 pm x 47 tm); antennal papilla short (1=9.5
pm), with three sensillae at center.
Mouthparts. Labrum (Fig. 5) (w = 209 mm) emarginate, with fourteen marginal se
tae (1 = 19 pm). Epipharynx naked. Mandibles (1 = 266 pm, w = 142 pm) tridentate.
Maxillae (1 260 pm, w = 104 pm), with three setae on external part (1 28.5 pm), me
sal margin papillose. Maxillary palpi much wider than long (1 9.5 pm, w = 28.5 pm);

2 mm

2 '''''

0.2 mm

Tormos et al.: Larvae of Chrysididae

galeae present (1= 9.5 m, w = 9.5 pm). Labium (Fig. 6) (1= 95.2 prm, w=228 pm) with
short palpi (1=19 prm, w=38 pm); spinneret a transverse slit (1= 57.1 pm). Maxillae
and labium with pigmented bands.

Omalus biaccinctus (Buysson)

Imagos were obtained from three nests: one of Passaloecus gracilis (Curtis, 1834)
which had six cells (cell 1 being parasitized, from which a male emerged). Another
nest of Passaloecus sp. contained three cells, all of them parasitized; the mature larva
of cell 2 was transferred to 70% alcohol for later study; a male emerged from cell 1 and
in cell 3 the development of the chrysidid was not completed although its larva did
construct a cocoon. The third nest contained nine parasitized cells; the mature larvae
of cells 1, 2 and 3 were transferred to 70% alcohol for later study; two females emerged
from cells 4 and 9, and in cells 5, 6, 7, and 8 the development of the chrysidid was not
completed, although the larvae did construct a cocoon. The absolute measurements of
the description of the larva refer to the specimen from Cofrentes (Valencia) (reference:
Mature larva. General aspect (Fig. 7). Body robust (1 4.3 Im, w = 1.8 tm); abdom
inal segments not divided into annulets. Fourth abdominal segment humped dorso
laterally Anus terminal, as a transverse slit. Pleural lobes scarcely developed.
Integument microspinulose (1 4 pm), with sparse minute setae (1 14 pm) (Fig. 8a).
Spiracles (Fig. 8b) with a well-developed peritreme; atrium simple, lined with
weak ridges. First pair (mean diam = 75.3 prm, n = 10) slightly larger than the rest
(mean diam = 66.6 mm, n = 18).
Head (Fig. 9).W = 914 m, h = 885 pm with sparse, short setae (1 19 pm). Coronal
suture and parietal bands present. Antennal orbits circular (d = 47.5 pm) located be
low middle of head; antennal papilla (1 =19 pim, w = 18 [tm) rather long, with three
small sensillae at center (Fig. 10). Clypeolabral suture slightly emarginate.
Mouthparts. Labrum (Fig. 11) W = 324 prm, emarginate, with ten short setae (1 =
10 pm) and six marginal sensillae (w = 9 pm). Epipharynx naked. Mandibles (1 257
ptm, w = 125 pm) tridentate. Maxillae (1 105 prm, w = 63 pm) with a few setae on ex
ternal part (1 9 pm); margin mesally papillose. Maxillary palpi slightly wider than
long (1= 29 prm, w = 30.4 pm); galeae present (1= 19 prm, w = 9 pm). Labium (Fig. 12)
(1= 142 pIm, w = 171 tm) with four setae at lower face (1 10 pm); palpi short (1 19
tm, w = 29 pm); spinneret a transverse slit (1 = 42 pm). Hypopharynx with four sen
sory pores (d = 7 pm). Maxillae and labium with strong, pigmented bands.

Trichrysis cyanea (Linnaeus)

Data on the biology of this species (hosts, mechanisms of parasitism, sex-ratio)
have been provided by Enslin (1929) and Danks (1970). Various hosts have been cited
by Trautmann (1927), Micheli (1929), Hamm & Richards (1930), Grandi (1931), and
Medvedev (1978).
Individuals were obtained from nests of Trypoxylon attenuatum Smith, 1851; T
beaumonti Antropov, 1991; figulus (L., 1758); Trypoxylon sp.; Pemphredon lethifera
(Schuckard, 1837) and Psenulus pallipes (Panzer, 1978). Table 1 shows the incidence
of the Chrysididae in each of the hosts.
This was the most abundant cleptoparasite, occurring in 42 nests with 177 cells, of
which 62 were parasitized (35%). The number of cells affected per nest and the para
sitism index as a function of the position of cell in the nest are shown in Tables 2 and
3, respectively. With respect to the mortality observed in the 62 parasitized cells, in 7

Florida Entomologist 79(1)

2.5 mm

March, 1996

A S'

0.1 mm



0.5 mm

0.2 mm

Figures. 7-12. Mature larvae of 0. biaccinctus: 7, general aspect; 8a, setae of integ
ument (detail); 8b, spiracle (atrium and subatrium); 9, head in frontal view; 10, an
tennal papilla; 11, labrum; 12, labium and maxilla (oral face).

Tormos et al.: Larvae of Chrysididae



Number of Total Number of
Nests Number of Parasitized
Host Affected Cells Cells

Trypoxylon attenuatum 12 60 16(27%)
Trypoxylon beaumonti 14 67 28(42%)
Trypoxylon figulus 5 23 5(22%)
Trypoxylon sp. 9 17 11(65%)
Pemphredon lethifera 1 2 1(50%)
Psenulus pallipes 1 8 1(13%)

(11%) the development of the Chrysididae was not completed, although in all cases
the larva constructed a cocoon. Of the 51 images obtained (four larvae were preserved
in 70% alcohol), 21 were male and 30 female. Although the sex-ratio obtained is 0.7 m:
1.0 f, this is not significantly different from a 1:1 ratio as observed in many haplo-dip
loid species (X2_ 1.607; d.f.= 1; 0.3 > p > 0.2). Furthermore, no significant differences
were seen in the sex-ratio as a function of the position of the parasitized cell (X2 9.82;
d.f. 6; 0.2 >p> 0.1).
Comparing these data with those obtained by Krombein (1967) for other species of
Chrysididae, it may be concluded that T cyanea has a low degree of host specificity
and that it is quite effective as a parasitoid. The proportion of parasitized nests with
respect to the total number of nests of each species obtained was as follows: T attend
uatum (16%), T beaumonti (10%), T figulus (38%), Trypoxylon sp. (38%), P lethifera
(2%), and P pallipes (17%); only 8 of the 38 affected nests with two or more cells had
50% or more of the cells parasitized and the mean number of cells affected per para
sitized nest was 1.48 (35%). The sex-ratio was similar to that reported by Danks


Like the rest of the described species of Elampini (Hedychrum rutilans, 0. aeneus
and Pseudomalus auratus), 0. biaccinctus displays antennal orbits located near or be
low middle of head; well-developed antennal papillae; and entire (not divided into two
rings) body segments that are slightly humped in the dorsolateral direction. In this
species, the back of the abdominal segments is very humped.
The scanty and protuberant marginal sensilla of the labrum and the presence of
galeae are morphological characters shared with 0. aeneus and Pseudomalus auratus.


Nests with 1 affected cell 28 (67%)
Nests with 2 affected cells 10 (24%)
Nests with 3 affected cells 3 (7%)
Nests with 4 affected cells 0
Nests with 5 affected cells 1 (2%)

Florida Entomologist 79(1)


Number of Sex
Total Number Parasitized
Cell Position of Cells Cells male female

Cl 42 23 (55%) 9 12
C2 38 14 (37%) 2 9
C3 32 13 (41%) 3 7
C4 20 5 (25%) 5
C5 16 3 (19%) 1
C6 13 3 (23%) 1 1
C7 5 1 (20%) 1
Total 166 62 (36%) 21 30

Although their presence is quite probable in Hedychrum rutilans, these characters
are not noted in the work of Maneval (1936).
The number of mandibular teeth is variable: 4-5 in Pseudomalus auratus, 3 in
Hedychrum rutilans and 0. biaccinctus, and 2 in 0. aeneus.
Chrysis gracillima has the segments of the body divided into two annulets, anten
nal orbits situated in normal position and marginal sensillae of the labrum numerous
and minute. These characteristics are shared by the described species of Chrysidini:
Caenochrysis doriae, Chrysis cembricola, C. cessata, C. grandis, C. ignita, C.
inaequidens, C. inflata, C. nitidula, C. smaragdula, Chrysura dichroa, Praestochrysis
shangaiensis, and T cyanea. All species have tridentate mandibles with the exception
of T cyanea, in which the mandibles are quadridentate.
Other generalized characters are the presence of galeae and short antennal papil
lae, although both structures are absent in Chrysura dichroa (Grandi, 1961). Praesto
chrysis shangaiensis does not have antennal papillae, and Parker (1936) does not
represent the galeae.
Our scanty knowledge of the mature larvae of the Chrysidoidea does not permit
the establishment of primitive vs. specialized states of the various characters. In this
respect, it is highly significant that in his cladistic analysis of the Chrysidoidea, Car
penter (1986) did not use any preimaginal morphological character. However, from
the characters studied it may be deduced that in some Chrysidinae (sensu Kimsey &
Bohart 1990), the galeae are apparently lacking, thus a specialized feature. In the
Elampini the low position of the antennal orbits and long antennal papillae may also
represent states of specialized character.


We are much indebted to H. E. Evans (Colorado State University, U.S.A.), L. S.
Kimsey (University of California, Davis, U.S.A.) and Karl V. Krombein (National Mu
seum of Natural History, Washington, U.S.A.), for their comments on the manuscript.
Grants from the DGICYT (PB91 0351 C02) supported the study

March, 1996

Tormos et al.: Larvae of Chrysididae


AsiS, J. D., J. TORMOS, AND S. F. GAYUBO. 1994. Biological observations on Trypoxylon
attenuatum and description of its mature larva and its natural enemy Trichry
sis cyanea (Hymenoptera: Sphecidae: Chrysididae). J. Kansas Entomol. Soc. 67:
CARPENTER, J. M. 1986. Cladistics of the Chrysidoidea (Hymenoptera). J. New York
Entomol. Soc. 94: 303-330.
CLAUSEN, C. P. 1940. Entomophagous Insects. Hafner Pub. Co. New York.
DANKS, H. V. 1970. Biology of some stem-nesting aculeate Hymenoptera. Trans. R.
Entomol. Soc. London 122: 323-399.
ENSLIN, E. 1929. Beitrage zur Metamorphose der Goldwespen. Zeitschr. wissen
schaftl. Insektenbiologie XXIV: 116-130.
EVANS, H. E. 1987. Order Hymenoptera, pp. 597-710 in F. W. Stehr [ed.] Immature In
sects. Kendall/Hunt Publishing Company. Dubuque. Iowa.
GRANDI, G. 1931. Contributi alla conoscenza biological e morfologica degli imenotteri
melliferi e predatori XII. Bol. Lab. Entomol. R. Ist. Sup. Agr. Bologna 4: 1872.
GRANDI, G. 1959. Contributi alla conoscenza degli Imenotteri Aculeati. Bol. Inst.
Entomol. Univ. Bologna 28: 239-292.
GRANDI, G. 1961. Studi di un entomologo sugli imenotteri superior. Boll. Inst. Ento
mol. Univ. Bologna 25.
HAMM, A. H., AND O. W. RICHARDS. 1930. The biology of the British fossorial wasps of
the wasps of the families Mellinidae, Gorytidae, Philanthidae, Oxybelidae and
Trypoxylonidae. Trans. Entomol. Soc. London 78: 95-131.
IWATA, K. 1976. Evolution of Instinct. Comparative Ethology of Hymenoptera. Amer
ind Publishing Co. Pt. Ltd. New York.
JANVIER, H. 1933. Etude biologique de quelques Hym6nopteres du Chili. Ann. Sci.
Nat. Zool. 16: 210-356.
KARLIN, S., AND S. LESSARD. 1986. Theoretical studies on Sex Ratio Evolution. Princ
eton University Press. New Jersey.
KIMSEY, L. S., AND R. M. BOHART. 1990. The Chrysidid Wasps of the World. Oxford
University Press. Oxford.
KROMBEIN, K. V. 1967. Trap-nesting Wasps and Bees: Life Histories, Nests and Asso
ciates. Smithsonian Press. Washington.
MARECHAL, P. 1923. Note sur 1'6tat larvaire et 1'6tat nymphal de Chrysis ignita L.
Bull. Soc. Entomol. Belg. 5: 103-107.
MANEVAL, H. 1936. Nouvelles notes sur divers Hym6nopt6res et leurs larves. Rev.
Fran. Entomol. 3: 1832.
MEDVEDEV, G. S. 1978. Keys for the identification of the insects of the European part
of the USSR. Acad. Sciencies.USSR. Zoological Institute. (in Russian).
MICHELI, L. 1929. Note biologiche e morfologiche sugli imenotteri (I). Bol. Soc. Ento
mol. Ital. 7: 34-43.
PARKER, D. E. 1936. Chrysis shangaiensis Smith, a parasite of the oriental moth. J.
Agric. Res. 52: 449-458.
SOIKA, A. G. 1934. Etudes sur les larves des hym6nopteres. Ann. Soc. Entomol.
France. 103: 337-344.
SPRADBERY, J. P. 1973. Wasps: An account of the biology and natural history of social
and solitary wasps. Sidwick & Jackson Biology Series. London.
TRAUTMANN, W 1927. Die Goldwespen Europas. G. Uschmann Weimar.
VAN LITH, J. P. 1953. De Nederlandse metselwespen. Lev Nat. 12: 231-233.
VAN LITH, J. P. 1955. De Nederlandse Spilomena-soorten (Hym.: Sphecidae). Ento
mol. Ber. 15: 525-527.
VAN LITH, J. P. 1956. Hoplomerus (Hoplomerus) spinipes (L.) en H. (Spinicoxa) reni
formis (Gmel.). Entomol. Ber. 16: 259-263.
VAN LITH, J. P. 1958. Opmerkingen over Chrysididae (3). Entomol. Ber. 58: 231-232.
YAMANE, S. 1976. Mature larva of Chrysis ignita (Hymenoptera: Chrysididae). New
Entomol. 25: 2124.

Florida Entomologist 79(1)


Delaware Agricultural Experiment Station
Department of Entomology and Applied Ecology
College of Agricultural Sciences, University of Delaware
Newark, DE 19717-1303


Adult Podisus maculiventris (Say) consumed no adult ladybird beetles (Harmonia
axyridis Pallas) and 30-35% of larval ladybird beetles in no-choice laboratory trials.
Nymphal P. maculiventris appeared to be more agile than adults in petri dishes and
attacked 65% of both larval and adult H. axyrides; however, they only killed and con
sumed 50% of larvae and 20% of adult ladybird beetles. Ladybird beetle larvae were
aggressive and often escaped, whereas adult beetles were usually rejected by the
nymphal predators, suggesting unpalatability Overall, P. maculiventris took more
than four times longer to capture ladybeetle larvae than to capture fall armyworms
(Spodoptera frugiperda [J. E. Smith]). In nature, while some ladybird beetles are un
doubtedly consumed by P. maculiventris, most probably escape predation, either by
evasion or through lack of palatability

Key Words: Harmonia axyrides, prey selection, defense, palatability, refugia, no
choice tests.


Los adults de Podisus maculiventris (Say) no consumieron adults del coccinelido
Harmonia axyridis Pallas y solo depredaron el 30-35% de sus larvas en ensayos de ali
mentaci6n obligada. Las ninfas de P. maculiventris parecieron ser mas agiles que los
adults en las places de petri y atacaron el 65% de las larvas y adults de H. axyridis;
sin embargo, estas solo mataron y consumieron el 50% de las larvas y el 20% de los
adults del coccinelido. Las larvas del coccinelido fueron agresivas y a menudo esca
paron, mientras que los adults usualmente eran rechazados por las ninfas predator
ras, lo cual sugiere su impalatabilidad. En general, P. maculiventris necesit6 4 veces
mas tiempo para capturar un coccinelido que para capturar un gusano trozador (Spo
doptera frugiperda (J. E. Smith)). En la naturaleza, mientras algunos coccinelidos son
sin duda consumidos por P. maculiventris, la mayoria probablemente escape a la de
predaci6n, ya sea por evasion o por falta de palatabilidad.

The predaceous hemipteran, Podisus maculiventris (Say), has a broad host range,
consisting primarily of soft-bodied, slow-moving larvae of Coleoptera and Lepidoptera
(Esselbaugh 1948, Mukerji & LeRoux 1965). Among those listed as prey species, how
ever, are at least four species of predaceous Coccinellidae (McPherson 1980). These
would seem to be unlikely prey for P. maculiventris, because of their typically agile,
rapid movements and the variety of chemical defenses generally present in coccinel
lids (Dettner 1987).
Podisus maculiventris is a candidate for augmentative release (e.g., Hough Gold
stein 1996), but before large numbers are released, its potential impact on beneficial

March, 1996

Hough-Goldstein et al.: Podisus Predation on Coccinellids 65

arthropods should be evaluated. In this study, we examined the potential for P. mac
uliventris adults and nymphs to feed on larval and adult ladybird beetles.


Podisus maculiventris were provided initially by the Maryland Department of Ag
riculture (Plant Protection Section), Annapolis, MD; these were reared in our labor
tory for one or two generations on fall armyworm, Spodoptera frugiperda (J. E.
Smith), before use in trials. Fall armyworms were obtained from the DuPont Stine
Haskell Laboratory, Newark, DE, where they were reared on artificial diet. Colorado
potato beetles, Leptinotarsa decemlineata (Say), were field-collected as adults, and
larvae were reared on greenhouse-grown potatoes ('Superior'). Ladybird beetle (Har
monia axyridis Pallas) adults and larvae were collected from mixed, weedy vegetation
on the University of Delaware Experiment Station Farm immediately before use.
Four no-choice trials were conducted between 11 and 25 July, 1995. No-choice ex
periments were used exclusively, because these are more likely to result in consump
tion and thus are a more stringent test of prey suitability than choice tests. Predators
were fed in the colony 24 h prior to the start of each experiment. For each trial, P. ma
culiventris were placed individually in plastic petri dishes approximately 2 h before
the trial began. Between 8 and 20 (usually 10) dishes were set up for each prey type,
and one potential prey was placed in each dish. The dishes were observed continue
ously for 1 h after prey were introduced and then checked for prey consumption every
half hour for the next 6 h. Dishes were left at room temperature (approximately 24 C)
overnight and checked again for consumption 20 h after set-up.
Two trials were run using P. maculiventris adults and two using fourth instars. For
each P. maculiventris life stage, half the trials used arenas consisting of 9 by 1.3 cm
plastic petri dishes, and half used 11 by 2.5 cm plastic dishes containing a piece of 9
cm filter paper, folded in half with the edges folded up as a refugium.
The first trial compared fall armyworm larvae, Colorado potato beetle larvae, la
dybird beetle larvae, and ladybird beetle adults as potential prey. Subsequent trials
eliminated Colorado potato beetle, since both it and fall armyworm were palatable.
Results of encounters (consumption or no consumption) were evaluated using
Fisher's Exact Test (SAS Institute 1990). Times to capture (in min) of different types
of prey for those P. maculiventris that consumed prey during each 20-h trial were com-
pared using Analysis of Variance (ANOVA; PROC GLM, SAS Institute 1990). Time of
capture was assigned as soon as the predator inserted its proboscis firmly and se
curely into the prey without either the prey escaping or the predator rejecting the
prey Attacks that occurred between 1 h and 6 h were assigned times as soon as they
were observed (e.g., a capture noted at 1.5 h was assigned 90 min, even though it may
have occurred somewhat sooner). Attacks that occurred overnight were conservea
tively) assigned a time of 7 h (420 min), even though they may actually have occurred
somewhat earlier or considerably later. Time of capture over all trials for all predators
that consumed fall armyworms or ladybird beetle larvae was also analyzed by preda
tor stage (adult or fourth instar), type of arena (9-cm or 11 cm with refugium) and
prey type (fall armyworm or ladybird beetle larva), using ANOVA (SAS Institute


In small dishes without refugia, P. maculiventris adults ate 100% of the fall army
worms and Colorado potato beetle larvae, 35% of the ladybird beetle larvae, and none
of the ladybird beetle adults (Table 1). Consumption was similar in the larger dishes

Florida Entomologist 79(1)

with refugia where adult P maculiventris ate 100% of the fall armyworms, 30% of the
ladybird beetle larvae, and none of the ladybird beetle adults. In both trials, consump
tion rate varied significantly by prey (P<0.0001, Fisher's Exact Test). The time to cap
ture was significantly shorter for fall armyworms than for ladybird beetle larvae in
the 11-cm arenas (F 17.27, df= 1,P= 0.0010; ANOVA, Table 1). A similar trend was
observed in the 9-cm dishes (F = 3.27, df 2, P= 0.0581, Table 1).
In 9-cm and 11 cm dishes, respectively, P maculiventris fourth instars killed and
consumed 60 and 80% of the fall armyworms, 30 and 70% of the ladybird beetle lar
vae, and 10 and 30% of the ladybird beetle adults during the 20-h observation period
(Table 1). Differences in nymphal consumption by prey type were not significant for
the 9-cm dish test (P 0.080) or the 11 cm dish test (P 0.111); however, the differ
ence was significant for the two nymph tests considered together (P= 0.007, Fisher's
Exact Test). The time to capture (for those that consumed prey) did not differ signifi
cantly by prey type for nymphs in 9-cm dishes (ANOVA, F = 0.32, P= 0.5883, df = 1),
but approached significance in 11 cm dishes (F = 3.36, P= 0.0624, df = 2) with fall ar
myworms on average captured more quickly than ladybird beetle larvae or adults (Ta
ble 1).
In addition to ladybird beetles that were killed and consumed by P maculiventris
nymphs, several other ladybird beetle larvae (2 in the 9-cm arena and 1 in the 11 cm
arena) and nearly half of the ladybird beetle adults (4 in the 9-cm arena and 5 in the
11 cm arena) were attacked and apparently sucked by the P maculiventris nymphs
for one or more brief periods, and then abandoned. Thus, the nymphs actually caught
and "sampled" 65% of the ladybird beetle larvae and 65% of the ladybird beetle adults
(over both dish sizes), but only consumed and killed 50% of the larvae and 20% of the
adult ladybird beetles. Ladybird beetle larvae often attempted to bite the predator's


Time (min.) to
No. Consumed Capture (mean
Predator Stage Arena Prey' (N) SEM)

Adult 9-cm FAW 8 (8) 48 + 23
CPB 9 (9) 144+ 54
LBBL 7 (20) 253 + 79
LBBA 0 (10)
Adult 11-cm, refugia FAW 10 (10) 84 + 42
LBBL 6 (20) 367 + 53
LBBA 0 (10)
Fourth instar 9-cm FAW 6 (10) 23+ 19
LBBL 3 (10) 7 4
LBBA 1 (10)
Fourth instar 11 cm, refugia FAW 8 (10) 77 + 33
LBBL 7 (10) 289 + 70
LBBA 3 (10) 146+ 137

FAW, fall armyworm; CPB, Colorado potato beetle; LBBL, ladybird beetle larva; LBBA, ladybird beetle adult.

March, 1996

Hough-Goldstein et al.: Podisus Predation on Coccinellids 67

proboscis and frequently escaped, whereas adult ladybird beetles were usually re
jected by the predator.
Analysis of variance over all experiments on time of capture by predators that con
sumed fall armyworms or ladybird beetle larvae indicated a trend toward more rapid
capture by P maculiventris nymphs than by adults (Table 2). Prey were captured
more rapidly in 9-cm dishes without refugia than in 11 cm dishes with refugia (Table
2); fall armyworms were captured much more quickly than ladybird beetle larvae (Ta
ble 2).


Adult P maculiventris appeared to be unable to capture ladybird beetle larvae or
adults successfully because of their lack of agility (in small petri dishes) compared
with the quick-moving prey. Those adult predators that did capture ladybird beetle
larvae, however, did not subsequently reject them.
Podisus maculiventris nymphs appeared to be considerably more agile (in petri
dishes) than P maculiventris adults or, perhaps, they were simply more highly moti
vated to feed than the adult predators. Both ladybird beetle larvae and adults were
caught by the predator nymphs. Once caught, however, nearly a quarter of the lady
bird beetle larvae and 30% of the ladybird beetle adults either escaped or were re
jected by the nymphs after one or more feeding bouts. Ladybird beetle larvae
responded aggressively to predation attempts in a manner similar to that described
by Marston et al. (1978) for corn earworm, Heliothis [Helicoverpa] zea (Boddie). Adult
ladybird beetles did not respond aggressively to attack, and thus were apparently un
palatable to the predator. Some of the "sampled" ladybird beetle larvae or adults may
have died eventually due to injection of salivary digestive enzymes (Cohen 1990,
1995), although we did not observe any obvious ill effects during several days in the


Time (min) to
Variable N (mean SE) F' P

Predator Stage
Adult 31 168 +33 2.65 0.1099
Fourth instar 24 116+ 33
9-cm 24 97+ 32 6.15 0.0165
11 cm, refugia 31 183+ 33
FAW 32 62 17 28.63 0.0001
LBBL 23 262 40

'Overall F value = 12.25 (df3, 51), P= 0.0001 (PROC GLM, SAS Institute 1990).
FAW, fall armyworm; LBBL, ladybird beetle larva.

Florida Entomologist 79(1)

Over all trials, P maculiventris took more than four times longer to capture lady
bird beetle larvae than to capture fall armyworms, probably due to the agility and
speed of the ladybird beetle larvae in evading the predators. Fall armyworms, in con
trast, were slow-moving, with little defense evident against predation. Because the
predators were reared on fall armyworms prior to the trials, however, it is also possi
ble that some adaptation to this prey species may have occurred.
Arena size and presence of refugia primarily affected time to capture rather than
total prey captured, with prey overall captured nearly twice as quickly in the smaller
dishes that lacked refugia than in the larger dishes with refugia. The larger dishes
with refugia undoubtedly reflect natural conditions somewhat better than the small
empty dishes. However, even the larger dishes did not allow flight, an obvious means
of escape for adult ladybird beetles in the field.
In summary, this species of ladybird beetle (Harmonia axyrides), which was by far
the most abundant coccinellid species in our study area in July of 1995, was clearly
not a preferred prey for P maculiventris. Undoubtedly some ladybird beetles are
eaten by the predator in nature; indeed, during specimen collection for these experi
ments, one P maculiventris nymph was observed feeding on an H. axyrides larva. Un
der some circumstances, local losses of coccinellids due to P maculinventris predation
could be high. In general, however, most larvae and especially adults of this coccinel
lid species are likely to escape predation by most P maculiventris. We predict that
other predaceous coccinellids will show similar behavioral and chemical predation
avoidance characteristics.

We thank C. Holko and P. Tipping, Maryland Department of Agriculture for pro
viding Podisus maculiventris to initiate our colony, and D. Tallamy for valuable com-
ments on a draft of the manuscript. Published as Paper No. 1573 in the Journal Series
of the Delaware Agricultural Experiment Station, Contribution No. 681 of the Depart
ment of Entomology and Applied Ecology, University of Delaware, Newark.

COHEN, A. C. 1990. Feeding adaptations of some predaceous Hemiptera. Ann. Ento
mol. Soc. America 83: 1215-1223.
COHEN, A. C. 1995. Extra-oral digestion in predaceous terrestrial Arthropoda. Annu.
Rev. Entomol. 40: 85-103.
DETTNER, K. 1987. Chemosystematics and evolution of beetle chemical defenses.
Annu. Rev. Entomol. 32: 1748.
ESSELBAUGH, C. O. 1948. Notes on the bionomics of some midwestern Pentatomidae.
Entomol. America 28: 173.
HOUGH GOLDSTEIN, J. A. 1996. Use of predaceous pentatomids in integrated manage
ment of the Colorado potato beetle (Coleoptera: Chrysomelidae), in M. Coll and
J. Ruberson [eds.], Predatory Heteroptera in agroecosystems: their ecology and
use in biological control. Thomas Say Publ., Entomol. Soc. America, Lanham,
of five species of soybean caterpillars to attack by the predator, Podisus macu
liventris. Environ. Entomol. 7: 5356.
MCPHERSON, J. E. 1980. A list of the prey species of Podisus maculiventris (Hemi
ptera: Pentatomidae). Great Lakes Entomol. 13: 1724.
MUKERJI, M. K., AND E. J. LEROUX. 1965. Laboratory rearing of a Quebec strain of the
pentatomid predator, Podisus maculiventris (Say) (Hemiptera: Pentatomidae).
Phytoprotection 46: 40-60.
SAS INSTITUTE. 1990. SAS user's guide, version 6, 4th ed. SAS Institute, Cary, NC.

March, 1996

Scientific Notes


Barry University, School of Natural and Health Sciences,
11300 N.E. Second Avenue, Miami Shores, FL 33161, USA

While working on the coastal dunes at Holly Beach in Cameron Parish, Louisiana
during the summer of 1995, I had the opportunity to observe predation by the drag
onfly Erythemis simplicicollis (Say) on the cicada Diceroprocta delicate (Osborn). An
individual D. delicate that had just flown from its perch was captured by an E sim-
plicicollis and was being consumed in the surrounding tall vegetation. The dragonfly
appeared to have been drawn to the movement of the cicada as it flew from its perch.
Under similar conditions I witnessed another cicada being attacked by two dragon
flies: both dragonflies rose from their perches and collided with the cicada as it ap
preached the edge of a dune. However, this attack was unsuccessful and the cicada
escaped. The specific identity of these dragonflies was not determined. A colleague of
mine counted at least 17 species of Odonata at Holly Beach that day, and we could not
make a positive identification of the individuals that attacked the cicada.
I have been unable to locate other references to dragonflies using a non-periodical
cicada species as prey in North America. Fitch (1855), Riley (1885), Marlatt (1907),
Felt (1912), and McAtee (1921) have reported dragonflies feeding on periodical cicadas
(Magicicada spp.). However, most carnivorous animal species (see list in Marlatt
1907) use the superabundant food source that periodical cicadas represent during an
emergence where local population densities are often greater than three million cica
das per acre (Dybas & Davis 1962). Diceroprocta delicate is apparently the first non
periodical North American cicada species reported to be prey for dragonflies. Cicadas
have also been reported as prey of dragonflies in New Zealand (Myers 1929), Afghan
istan (Hay 1840), and South Africa (Distant 1897). The relatively low population num
bers of non-periodical cicadas combined with their large body size (D. delicate body
length is about 20.5 mm and length to wingtip is 29.8 mm; the captured E simplici
collins was 50.8 mm long) may make cicadas difficult targets for capture and perhaps,
therefore, influence the paucity of reports of dragonflies using cicadas as prey
Sidney W. Dunkle identified the E simplicicollis specimen, and Sr. John Karen
Frei of Barry University provided financial support.


This paper reports the dragonfly Erythemis simplicicollis capturing the cicada
Diceroprocta delicate as prey. It is unusual in that it represents a non-periodical ci
cada being used as prey by a dragonfly


DYBAS, H. S., AND D. D. DAVIS. 1962. A population census of seventeen-year periodical
cicadas (Homoptera: Cicadidae: Magicicada). Ecology 43: 432-443.
DISTANT, W. L. 1897. Zoological rambles in and around the Transvaal. Zoologist (4)1:
FELT, E. P. 1912. Twenty-seventh report of the state Entomologist 1911. New York St.
Mus. Bull. 155: 1198.

70 Florida Entomologist 79(1) March, 1996

FITCH, A. 1855. The seventeen-year locust, Cicada septendecim, Linnaeus. Trans.
New York St. Agric. Soc. 14: 742-753.
HAY, R. G. 1840. Notes on the wild sheep of the Hindoo Koosh, and a species of cicada.
Jour. Asiatic Soc. Bengal 53: 210-233.
MARLATT, C. L. 1907. The periodical cicada. USDA Bur. Entomol. Bull. 71: 1181.
McATEE, W. L. 1921. The periodical cicada, 1919; brief notes for the District of Colum
bia region. Proc. Entomol. Soc. Washington 23: 211-213.
MYERS, J. G. 1929. Insect singers: A natural history of the cicadas. George Routledge
& Sons, London.
RILEY, C. V. 1885. The periodical cicada. USDA Div. Entomol. Bull. 8: 146.

Florida Entomologist 79(1)


'Entomology and Nematology Department, University of Florida
Gainesville, FL 32611

2Plant Pathology Department, University of Florida, Gainesville, FL 32611

3Wildlife and Range Science Department,
South West Florida Research and Education Center, Immokalee, FL 33934

4Universidade Estadual Paulista do Jaboticabal, Sao Paulo, Brazil

Tropical soda apple, Solanum viarum Dunal (Solanaceae), a herbaceous annual
weed (Aranha et al. 1982) native to South America (Nee 1991), is considered a serious
weed threat and has been included in the list of noxious weeds by the Florida Depart
ment of Agriculture and Consumer Services (Florida Dept. Agric. and Consumer Ser
vices 1994). This weed has spread into other geographical regions including Central
America, the Caribbean, India, China, Africa (Coile 1993, Chandra & Srivastava
1978), and south Florida (Mullahey et al. 1993). It was recently found (July 1994) in
southern Mississippi by J. Byrd (Extension weed specialist, Mississippi State Univ,
pers. comm.). Solanum viarum is spreading rapidly, invading pasture lands in south
Florida. In 1992, infestations of pastures by S. viarum were estimated to be near
150,000 acres (Mullahey et al. 1993). A census of Florida ranchers in 1993 (J. J. M. un
published) found that nearly 400,000 acres of pastures and natural systems were in
fested by S. viarum causing an annual production loss to Florida ranchers of $11
million. The rapid spread in south Florida can be partially attributed to the plant's
ability to grow in sandy loam soils (Chandra & Srivastava 1978), great reproductive
potential (Mullahey et al. 1993), and effective seed dispersal by cattle and wildlife
(Mullahey & Colving 1993).
How this South American plant entered Florida is unknown. Earliest records in
dicate that it was initially observed in Hendry county (southwest Florida) in the
1980s (Mullahey et al. 1993). No insects have been reported feeding on this plant in
An exploratory survey was conducted by the authors during 8-14 June, 1994 to
record insects feeding on S. viarum plants in Sao Paulo and Parana states, Brazil and

March, 1996

Scientific Notes

southeast Paraguay Collections were made on natural stands of S.viarum on road
sides and grasslands. Insects found on the plant were trapped by hand. Plant parts
(fruits, leaves, stems) were examined for insects. Fruits and stems were dissected to
collect insects feeding on these internal parts. Identification to species level was de
termined mainly with those insects that were actually feeding on S. viarum. The in
sects collected on S. viarum at the different sites are listed in Table 1. The major
insects collected included Neoleucinodes elegantalis Guenee (Pyralidae), a fruit borer
that feeds inside S. viarum fruits destroying a great proportion of the seeds. The host
feeding range of this insect is not known (J. Vasconcelos, Universidade Estadual
Paulista do Campinas, pers. comm.). Gallo et al. (1988) listed N elegantalis as a to
mato pest in Brazil. Based on host data provided by Buzzi (1994), S. viarum repre
sents a new host record for the cassidines Metriona elatior Klug and Gratiana
boliviana Spaeth (Chrysomelidae), the larvae and adults of which feed on S. viarum
leaves. Sweetpotato, Ipomoea batatas L. (Convolvulaceae), is also listed as host of M.
elatior (Buzzi 1994). The genus Metriona includes six species (M. argentina Spaeth,
M. tenella Klug, M. bifossulata Boheman, M. erratic Boheman, M. vilis Boheman, M.
bilimeki Spaeth) that occur in Central America and/or South America (Buzzi 1988,
Maes & Staines 1991) and for most of them the host ranges are unknown or income
pletely known. Likewise, the host-feeding range of G. boliviana is unknown. A com
plete description of this cassidine is provided by Buzzi (1995). Few attempts have
been made to use cassidines for biological control of Solanum weeds. Gratiana lute
scens Boheman and G. pallidula Boheman were studied in South Africa (Siebert 1975)
as potential biological control agents of Solanum elaeagnifolium Cav. but research ef
forts were not continued because the insects' host range included eggplant, S. melon
gena L.
In June 1994, the lace bugs (Hemiptera: Tingidae), Corythaica sp. near cyanthicol
lis (Costa) formed small aggregations on the underside of S. viarum leaves. This insect
was most commonly found on S. viarum leaves. The genus Corythaica includes their
teen species that occur in the neotropics (Drave & Ruhoff 1965) and for most of them
the plant host ranges are unknown or incompletely known. Amblyophallus macuala
tus Funkhonser (Homoptera: Membracidae) was the second most abundant species
found on S. viarum stems and leaves. The butterfly Mechanitis lysimnia Fabricius
(Nymphalidae: Ithominae), probably polyphagous, was collected by Pitelli, a junior
author (R. A. P.), around Jaboticabal, Sao Paulo. The larvae seem to be associated with
extensive S. viarum leaf chewing damage. The known geographical distribution of M
lysimnia extends from Mexico to northern Argentina (D'Abrera 1984, DeVries 1987),
including Brazil. The other minor insects collected on S. viarum were not causing sig
nificant damage to the plants and most are probably polyphagous (Coccidae, Core
idae, Fulgoridae, Pentatomidae, Rhopalidae). Diabrotica speciosa Germar is a key
pest in Brazil. Larvae and adults feed on roots and foliage of common beans, Phaseo
lus vulgaris L., field corn, Zea mays L., and potatoes, Solanum tuberosum L.
This survey in Brazil and Paraguay indicated that a diverse group of insects (phy
tophagous and others) is associated with S. viarum and at least 5 species, M. elatior,
N elegantalis, and M. lysimnia, the tingid Corythaica sp. and the membracid A. mac
ulatus, probably cause significant damage to S. viarum plants in South America. Par
aguayan technicians (Centro Tecnologico Agropecuario) indicated that S. viarum is
not considered an important weed in the region. In Brazil, S. viarum is an occasional
invader of field crops and its geographical distribution includes Goias, Minas Gerais,
and the southern states (Costa et al. 1985). Further surveys in the weed's native re
gion, basic biological studies, and preliminary host range tests need to be imple
mented to start a biological control project for S. viarum in Florida.

Florida Entomologist 79(1)


Part Insect
Family Species Collected Stage' Collection Site2

Mechanitis lysimnia

Metriona elatior

Chrysomelidae Gratiana boliviana
Chrysomelidae Diabrotica speciosa



Corythaica sp.

Amblyophallus maculatus


fruit L Taiuva, SP
Divinolandia, SP
Ciudad del Este
leaf L,P Jaboticabal, SP
leaf L Divinolandia, SP
leaf P,A Divinolandia, SP
Ciudad del Este
leaf A Divinolandia, SP
leaf A Taiuva, SP
Divinolandia, SP
Ciudad del Este
leaf A Ciudad del Este
leaf A Taiuva, SP
Divinolandia, SP
Ciudad del Este
leaf A Divinolandia, SP
leaf A Taiuva, SP
leaf A Ciudad del Este
stem N,A Taiuva, SP
leaf A Taiuva, SP
leaf stem N,A Taiuva, SP
Divinolandia, SP
Ciudad del Este
leaf A All sites
leaf stem A All sites

'L=larva, N=nymph, P=pupa, A adult.
2SP=Sao Paulo state, Brazil; Ciudad del Este in south-east Paraguay

We thank G. R. Buckingham (USDA-ARS), J. H. Frank, and D. H. Habeck (Depart
ment of Entomology and Nematology, University of Florida) for reviewing the manu
script. We also thank J. Vasconcelos (Universidade Estadual Paulista do Campinas,
Sao Paulo) for identifying M. elation N elegantalis, and M. lysimnia; Z. J. Buzzi and
A. Sakakibara (Universidade Federal do Parana in Curitiba, PR) for identifying G. bo
liviana and A. maculatus, respectively; S. E. Halbert (State of Florida Department of
Agriculture & Consumer Services-Division of Plant Industry, Gainesville) for identi
fying the Corythaica sp. Specimens of several of the species identified were retained
by the taxonomists responsible for identification. Others were deposited in the insect
museum of the Universidade Federal do Parana in Curitiba, Brazil, and the Univer
sity of Florida Entomology & Nematology Department insect collection in Gainesville,
Florida. This manuscript is published as Florida Agricultural Experiment Station
Journal No. R-04336.

Pyralidae Neoleucinodes elegantalis


March, 1996

Scientific Notes


An exploratory survey was conducted in Brazil and Paraguay to record insects
feeding on Solanum viarum Dunal (Solanaceae). A list of insects collected is included.
The survey indicated that a diverse group of phytophagous insects is associated with
S. viarum, and some of them may have potential as biocontrol agents of S. viarum in


ARANHA, C., O. BACCHI, AND H. F. LEITAO. 1982. Plantas invasoras de cultures. Vol.
II. Institute Campineiro de Ensino Agricola. Campinas, Brasil. 597 p.
BUzzI, Z. J. 1988. Biology of neotropical Cassidinae, p. 559-580, in P. Jolivet, E. Petit
pierre, T H. Hsiao [eds.]. Biology of Chrysomelidae. Kluwer Academic Publish
ers, Netherlands.
BUzzI, Z. J. 1994. Host plants of neotropical Cassidinae, p. 205-212 in P. H. Jolivet, N.
L. Cox, and E. Petitpierre [eds.]. Novel aspects of the biology of Chrysomelidae.
Kluwer Academic Publishers, Netherlands.
BUzzI, Z. J. 1995. Redescricao de Gratiana boliviana Spaeth, 1926. (Coleoptera: Chry
somelidae: Cassidinae). Revista Brasileira de Zoologia (In press).
CHANDRA, V., AND S. N. SRIVASTAVA. 1978. Solanum viarum Dunal syn. Solanum
khasianum Clarke, a crop for production of solasadine. Indian Drugs 16: 53-60.
COILE, N. C. 1993. Tropical soda apple, Solanum viarum Dunal: the plant from hell.
Botany Circular No. 27. Florida Dept. Agric. & Consumer Services, Division of
Plant Industry
COSTA, J., E. SANTOS, E. FROM, N. L. COSTA, AND M. C. CUNHA. 1985. Ervas danin
has do Brasil. Solanaceae I. EMBRAPA. Brasilia, Brasil. 58 p.
D'ABRERA, B. 1984. Butterflies of South America. Hill House. Victoria, Australia. 256
DEVRIES, P. J. 1987. The butterflies of Costa Rica and their natural history. Princeton
University Press, New Jersey. 328 p.
DRAVE, C. J., AND F. A. RUHOFF. 1965. Lacebugs of the world: A catalog (Hemiptera:
Tingidae). Smithsonian Institution. Bull. 243. Washington D.C. 634 p.
control agents. Chapter 5B-57. Introduction or release of plant pests, noxious
weeds, arthropods, and biological control agents. Vol. 2, p. 698-26.
R. A. ZUCCHI, S. BATISTA, AND J. DJAIR 1988. Manual de Entomologia Agricola.
Editora Agronomica Ceres Ltda. Sao Paulo. 649 p.
MAES, J. M., AND C. L. STAINES. 1991. Catalogo de los Chrysomelidae (Coleoptera) de
Nicaragua. Revista Nicaraguense de Entomologia 18: 153.
MULLAHEY, J. J., AND D. L. COLVING. 1993. Tropical soda apple: A new noxious weed
in Florida. University of Florida, Florida Cooperative Extension Service, Fact
Sheet WRS-7.
MULLAHEY, J. J., M. NEE, R. P. WUNDERLIN, AND K. R. DELANEY. 1993. Tropical soda
apple (Solanum viarum): A new weed threat in subtropical regions. Weed Tech
nology 7: 783-786.
NEE, M. 1991. Synopsis of Solanum section Acanthophora: A group of interest for
glyco-alkaloides, pp. 258-266 in J. G. Hawkes, R. N. Lester, M. Nee, N. Estrada
[eds.] Solanaceae III: Taxonomy, chemistry, evolution. Royal Botanic Gardens
Kew, Richmond, Surrey, UK.
SIEBERT, M. W. 1975. Candidates for the biological control of Solanum elaeagnifolium
Cav. (Solanaceae) in South Africa. 1. Laboratory studies on the biology of Gra
tiana lutescens (Boh.) and Gratiana pallidula (Boh.) (Coleoptera, Cassididae).
J. Entomol. Soc. South Africa. 38: 297-304.

Florida Entomologist 79(1)


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

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

Keifer (1969) described an eriophyid mite Tegolophus perseaflorae sent to him by
C. W. Fletchman collected from Persea gratissima from Recife, Pernambuco, Brazil.
Dr. Fletchman reported this mite caused flower damage and decreased fruit produce
tion. In 1977, Dr. R. Baranowski (UF/TREC, Homestead, FL) collected this mite in the
bud of avocado, Persea americana Miller, at Irupana, Bolivia.
In May 1991, excessive flower drop and fruit deformation was observed in avocado
trees in the vicinity of Homestead, Dade Co., Florida. In a preliminary survey, an av
ocado orchard was sampled in May of 1991. Sampling consisted of collections of ten
floral clusters, fruitlets and buds from each of 10 trees. Tegolophus perseaflorae were
collected from buds and fruits. The mites were observed feeding on buds, causing ne
crotic spots on apical leaves, and subcircular, irregular openings on mature leaves.
Mites were also found in petioles, the underside of leaves and fruitlets (Fig. 1) The
mite is also reported to feed on the peduncle, calyx and stylar area (Medina et al.
1978, Jeppson et al. 1975). Feeding by this mite on fruitlets may cause fruit deforma
tion and discoloration.
A preliminary survey was initiated in June 1991 through May 1992 to determine
the relative frequency of T perseaflorae on fruits, leaves and flowers. Initially, ten
fruits, buds, and flower clusters were collected twice a month from each orchard. They
were placed in an ice chest (about 10 C) and transported to the laboratory where the
mites were counted. Voucher specimens were identified by the junior author. A total
of 508 T perseaflorae were collected from leaf buds and fruitlets. A significant differ
ence in frequency of mites was observed between buds and fruits [Chi-square 0.05 (1)
3.84]. More mites were observed on the buds (x + SE = 5.85 1.06) than in fruits
(2.29 0.72). Population peaks (18 -35 mites per bud) were observed from March to
May 1992. These warm dry months correspond with blooming and fruit formation on
avocado in the area. No mites were observed on flowers, fruits or leaves from June
through February. The above accounts would imply that warm weather is most favor
able for this eriophyid and that the presence of developing avocado plant organs
might influence its development.
Description. (See Figs. 2 -7.) Adult females are 155-170 pt long, about 37 pt thick,
abdominal thanosome with about 55-60 rings; rostrum 20 pt long, curved down, shield
design not clear; forelegs 24 t long; tibia 5 t long, tarsus 4.5 p, and claw 5.5 t long;
featherclaw 5-rayed; lateral sets 18 t long on about 5-7 behind the shield; first ventral
seta 38 t long on ring 20, second ventral seta 31 t long on ring 36. Telesome with 5-6
rings, the granules fine. Accessory seta 3 pt long. Female genitalia 19 t wide, 12 t long,
covering flap with about 18 close-set longitudinal ribs; genital seta 13 t long.
The presence of this mite in Florida represents another piece in the "puzzle" in the
continuous appearance of neotropical pests in this state. Since, avocado grafting ma
trial is often transported by commerce, it may be that T.perseafloraeis an immigrant
species that was introduced in Florida, unintentionally, by human transport.
Florida Experimental Station Journal Series No. R-04406.

March, 1996

Scientific Notes 75





Florida Entomologist 79(1)

March, 1996



Ji' 6

Figures 2 -7. Adult female Tegolophus perseaflorae (2), side of anterior section(3),
left foreleg (4), featherclaw (5), lateral rings and microtubercles on thanosome (6), fe
male genital structures and coxae (7) (after Keiffer).

The status, damage and description of Tegolophus perseaflorae Keifer, a newly in
produced mite into southern Florida, are discussed.

JEPPSON, L. R., H. H. KEIFER, AND E. E. BAKER. 1975. Mites injurious to economic
plants. Berkeley, California, University of California Press. 614 pp.
KEIFER, H. H. 1969. Eriophyid studies, California Dept. of Agric. C-1:1 20.
MEDINA, C., E. BLEINROTH, J. TANGO, AND W. Do CANTO. 1978. Abacate. Inst. Tecnol.
Alimentos, Secr. Agric., Sao Paulo 212 p.

Scientific Notes


Department of Entomology, University of Massachusetts, Amherst, MA, 010003

The oothecal parasitoid Comperia merceti (Compere) has been used as a biological
control agent against the brownbanded cockroach, Supella longipalpa (F) (Ortho
ptera: Blattellidae), in environments where pesticide use is undesirable (Slater et al.
1980). In an experimental study of brownbanded cockroach populations, Coler et al.
(1984) released C. merceti in 1978 in two insect rearing rooms at the University of
Massachusetts at Amherst. In one room, cockroach populations were augmented dur
ing the study by provision of food. Parasitism at that site increased as the brown
banded cockroach population density increased, followed by the collapse of the
cockroach population (Coler et al. 1984). After the end of the Coler et al. study in 1983,
no further parasitoid releases were made in the building in which these rearing rooms
were located. This note reports the status of the parasitoid populations in 1993 at the
two sites used by Coler et al. (1984), fifteen years after the initial release and ten years
after the last parasitoid release. Insect rearing has been conducted continuously in
these same rearing rooms with little change since the initial parasitoid releases. In
this note we demonstrate the ability of this parasitoid to persist under such conditions
without augmentation for long periods and to cause high levels of mortality to host
On 8 and 22 November, 1993, S. longipalpa oothecae were collected from each of
the two insect rearing rooms used in the Coler et al. study. These rooms were of mod
erate size (39.8 m3 and 35.7 m3) with loose construction providing cockroach harbor
age. Rooms were searched thoroughly for oothecae; search areas included walls,
shelving, molding, along electrical conduits, inside light timing boxes, and the under
side of tables. Oothecae inside electrical conduits, in deep wall cracks, or around in
sect rearing cages could not be retrieved. All oothecae (both currently live oothecae
and older, emerged or dead ones) encountered in accessible areas were collected and
examined for parasitism. Most oothecae contained neither live cockroaches nor para
sitoid stages, but rather were oothecae which had either died, or from which cock
roaches or parasitoids had previously emerged. These oothecae were probably no
more than two to three years old, as the rearing rooms are periodically repainted. In
the laboratory, all oothecae were examined under a stereoscope for either wasp emer
gence holes or the oviposition stalks present on the surface of parasitized oothecae.
Oothecae from which cockroach nymphs had emerged were easily identified by open
ing the oothecae and observing the remnants of hatched eggs. Oothecae from which
neither cockroaches nor wasps emerged were held for 60-70 days and those producing
either cockroach nymphs or parasitoids recorded. Oothecae from which no emergence
occurred were dissected. Of the 556 oothecae collect, only three were classified as hav
ing died, with nothing emerged. The remaining oothecae all resulted in emergence of
hosts or parasitoids, either in the field prior to collection, or during rearing.
As a separate measure of parasitism, oothecae from a laboratory colony were ex
posed at one study site as "trap hosts" for a 14 day period (22 November-6 December,
1993). Trap host oothecae were evenly distributed on walls and shelving. These ooth
ecae were then recovered and reared to determine the rate of parasitoid attack during

Florida Entomologist 79(1)


% Parasitism

Type of Host Room No. 1 Room No. 2

Old oothecae 35.5 (6.4)' (214)2 92.6 (2.8) (312)
Live oothecae 46.2 (27.1) (13) 88.2 (15.4) (17)
Trap host oothecae 45.8 (19.9) (24)3

'95% confidence interval for parasitism.
Number of oothecae examined.
'Number of oothecae deployed as trap hosts, 22 Nov 6 Dec., 1993.

the exposure interval, which was approximately half of the total period for oothecal
When comparing parasitism of the total live oothecae versus total oothecae show
ing previous emergence, no significant difference was observed (Chi2 0.004, d.f.1, p-
value 0.95) (Table 1). Percent parasitism of oothecae deployed as trap hosts was
lower than that of live oothecae collected in the same room. However, trap host ooth
ecae were exposed for 14 days, approximately half of the normal exposure period of

Figure 1. Comperia merceti preparing to oviposit in ootheca of Supella longipapla
(Photograph courtesy of R. Coler).

March, 1996

Scientific Notes

oothecae to parasitoid attack (LeBeck 1985). In addition, trap host oothecae were de
played at a time immediately following the removal of many wasps from the study site
through collection of existing live oothecae.
Data on cockroach densities were not collected, and it is not clear to what degree
the existing parasitoid populations contribute to lowering cockroach densities. These
observations show that C. merceti, once established at a site with a permanent popu
lation of brownbanded cockroaches, is able to persist without need for periodic re
lease. These observations also indicate that levels of mortality from parasitism under
such conditions can be large, suggesting that even non-augmented populations of this
wasp may be of value in suppressing brownbanded cockroaches at some sites, such as
animal rearing or breeding facilities, and zoos, where pesticide use is not desired.


A non-augmented population of the brownbanded cockroach (Supella longipalpa)
oothecal parasitoid Comperia mercetiwas observed to have persisted at two indoor ur
ban sites in Massachusetts for 10 years. Levels of parasitism were high (36-93%), sug
gesting that this parasitoid has the potential to contribute to pest suppression at
some types of urban sites, even when repeated releases of the parasitoid are not made.


COLER, R. R., R. G. VAN DRIESCHE, AND J. S. ELKINTON. 1984. Effect of an oothecal
parasitoid, Comperia merceti (Compere) (Hymenoptera: Encyrtidae), on a pop
ulation of the brownbanded cockroach (Orthoptera: Blattellidae). Environ. En
tomol. 13: 603-606.
LEBECK, L. 1985. Host-parasite relationships between Comperia merceti (Compere)
(Hymenoptera: Encyrtidae) and Supella longipalpa (F) (Orthoptera: Blattel
idae). Ph.D. Dissertation, 104 pp.
SLATER, A. J., M. J. HURLBERT, AND V. R. LEWIS. 1980. Biological control of brown
banded cockroaches. California Agric. 34: 1618.

Florida Entomologist 79(1)


(eds). 1995. Biological Control in the Western United States: Accomplishments and
Benefits of Regional Research Project W 84, 1964-1989. University of California, Di
vision of Agriculture and Natural Resources; Oakland. xii + 356 p. ISBN 1 879906-21
X. Paperback. (From ANR Publications, 6701 San Pablo Ave, Oakland, CA 94608
1239, $25.00 plus shipping. A hardback edition is available at $45.00).

This book is principally a series of progress reports, representing the research ef
forts of scientists from 13 states and 2 territories affiliated with the regional research
project designated "W 84." Most, but not all, contributors represent Land Grant Uni
versities in the western U.S.; scientists from the U.S. Department of Agriculture, state
Departments of Agriculture, and foreign countries are also included in the research
effort. The pests are mostly exotic species that became established within the last cen
tury, and the emphasis of the biological control research, not surprisingly, is classical
biological control. Also included are recommendations for further research on each
The first section of the book is devoted to history of the regional project, relation
ship of biological control to IPM, relationship of biological control to population and
evolutionary ecology, and the impacts of biological control. This introductory material
provides the theoretical and practical framework for the research reports, and is
richly illustrated with examples from the geographic region.
Most of the book consists of case histories of biological control efforts directed to
ward arthropod pests. The arthropod research reports total 57, and include various
mite, thrips, bug, psyllid, whitefly, aphid, scale, mealybug, beetle, caterpillar, and fly
projects. Nearly all involve crop pests, with only a few ornamental pests targeted.
In addition to arthropod pests, there are many research reports addressing weed
biological control. Most of the weed studies, which total 22, involve rangeland weeds,
but a few crop-infesting weeds are included. One-half of the weeds considered are in
the family Asteraceae, but the remaining species represent a very diverse group of
There is no attempt to include biological control of plant pathogens in this compi
lation; presumably these research efforts function under the auspices of another re
search project. Microbial organisms are not completely excluded, however, as plant
pathogens were introduced for biological control of some weed species, such as rush
Most of the research reports are very concise but informative. A particular
strength is the extensive bibliography accompanying most reports. Overall, the re
ports make an excellent introduction to ongoing research in biological control in the
western United States. Anyone contemplating research on one of these pests would
benefit not only from reviewing the research reports, but considering the authors' rec
commendations for additional research.
A very nice component of the book is summary tables (appendixes 1 & 2) that sum
marize the classical biological control attempts by biological control agent, target
pest, location, and level of success. The book is complete with a comprehensive subject
index. The only detracting feature is the inconsistent designation of higher taxa of
beneficial arthropods. Some authors note species names only, whereas others include
family, and sometimes order, designations. Although it takes additional space to pro
vide this additional information, it makes for a more useful publication.
Biological Control in the Western United States makes a nice companion volume
for the report of regional project S-192, Classical Biological Control in the Southern

March, 1996

Book Reviews 81

United States (D, H. HabeckF.F. D. Bennett. and J. H. Frank, eds.. 1990). Both volumes
provide updates on progress in biological control, but they take different approaches
The former is organized by taxon, with each report being fairly complete for each pest
considered. The latter is more integrative, emphasizing the entire commodity system,
and the treatment of each pest is therefore more brief The former also is a more hand-
some and durable volume. The title is clearly labeled on the binding, which makes re-
trieval from a busy bookshelf or cluttered desk more feasible. Unfortunately, with
increased quality comes a substantial cost, as the latter sells for a mere $6. Classical
Biological Control in the Southern United States was reviewed earlier in this journal
(74: 167-8; 1991) by D. Rosen.
This book will serve as a handy reference for nearly anyone interested in biological
control of arthropods and weeds.
John L. Capinera
Department of Entomology & Nematology
University of Florida
Gainesville, FL 32611-0620

Book Reviews

SMETANA, A. 1995. Rove beetles of the subtribe Philonthina of America north of Mex-
ico (Coleoptera: Staphylinidae): Classification, phylogeny and taxonomic revision.
Memoirs on Entomology. International 3: i-x, 1-946. Available from Associated Pub-
lishers, PO. Box 140103, Gainesville. FL 32614-0103 at $125.

Almost everyone who has worked on insect communities of natural or agricultural
lands in America north of Mexico has encountered members of the genus Philonthus
or its allies (subtribe Philonthina). All too often identification to species level was un-
available, in part because the existing keys were mainly from the 19th century and
were grossly inadequate, and in part because for over three decades neither the USDA
nor the Smithsonian Institution has employed a systematist to work on Staphylin-
idae. I know this because, although I am not employed as a systematist (I merely pro-
fess a side-interest in systematics of Staphylinidae as a means of identifying the
species with which I have worked as an ecologist), I have been asked to identify in my
spare time very many specimens of this group. Ales Smetana, in this masterly book,
the culmination of many years of research, has come to the aid of the entomological
community and made identification of adults of Philonthus and allies possible. Now I
can say to inquirers: "buy a copy of the book and make your own identifications."
Alet Smetana is an immigrant to Canada employed as a systematist by Canada
Agriculture. Having written taxonomic works in Czech, German and French, he now
writes remarkably well in English. Over 40 years ago, employed as a forestry ento-
mologist in what was then Czechoslovakia, he began to publish systematic works on
the subfamily Staphylininae of Staphylinidae. Since arrival in Canada, his major
works were funded by his employers, and published in Memoirs of the Entomological
Society of Canada It is a sad reflection on the reduced status of insect systematics in
Canada that Canada Agriculture found itself unable to support the cost of this publi-
cation, his most important work. Therefore, this masterpiece was published in Flor-
After the introductory taxonomic history, begin 13 pages on natural history of
Philonthina. The importance of Philonthina suggests there should be many more
pages on natural history. but I suspect that lack of a huge number of citable studies
reflects one thing: taxonomy is the underpinning for all other entomological endeav-
ors, and that if entomologists (basic and applied) who have encountered Philonthina

Florida Entomologist 79(1)

in the field in America north of Mexico had been able to identify specimens, there
would have been many more publications.
The systematic section [descriptions, distribution (with maps) and keys to identi
fiction] occupies over 700 pages, and then there are almost 150 pages of high-quality
illustrations, mainly of diagnostic characters but including some habitus illustra
tions. Remaining parts are a systematic index and a cladistic analysis.
Smetana states that his cladistic analysis is hindered by lack of understanding of
groups belonging to and allied to Philonthus worldwide. His next endeavor will be a
worldwide analysis. This new analysis may cause changes in generic assignment of
the species of America north of Mexico. Beware, reader, that the generic assignment
of species names in this book is not set in stone! Yes, it would have been "nice" if the
worldwide cladistic analysis had been done before this book appeared in print so that
the generic assignments would have been less subject to future change. But I for one
am extremely pleased to see the appearance of this book now, and I will gladly live
with the need for future changes.
Doubtless some people will be deterred by the price of the book. But it has 946
pages, so consider the average price of a technical book of half that number of pages
produced by a "big name" publisher (I won't name them -you know them), and you
will realize that this book is a bargain. It is simply that Philonthus and allies are a
large subtribe within a very large family. Knowledge of systematics of very many of
the other subtribes of the family is extremely poor in America north of Mexico, and
much worse in all other regions except Europe. If only Smetana were to live 946 years,
he might present us with many other much-needed volumes!
I have extremely few criticisms and will air only one. Smetana states that many
species of Philonthina were introduced to America north of Mexico. In fact, very few
species were introduced, and these recently, by USDA entomologists, as natural ene
mies of pest flies whose larvae inhabit cattle-dung. Most of the species that Smetana
regards as "introduced" arrived by unknown means (not by purposeful introduction)
and should be classed as immigrants [see Frank & McCoy Florida Entomol. 73: 1-9
(1990), 76: 153 (1993), 78: 21 35 (1995)]. But, then, most humans in America north of
Mexico also are immigrants, or at least of immigrant stock.
J. H. Frank
Entomology & Nematology Dept.
University of Florida
Gainesville, FL 32611 0630

March, 1996

Edited Minutes-78th Annual Meeting


The fourth and final 1994-95 Executive Committee meeting was held on August 6,
1995, at the Cariari Hotel in San Jose, Costa Rica. President Ellen Thoms called the
meeting to order at 4:30 PM on August 6, 1995.
The 1994-95 Annual Business Meeting of the Society was called to order by Pres
ident Thoms at 5:00 p.m., Tuesday, August 8, 1995. A total of 41 Society members were
present. Minutes of the 1994 Annual Business Meeting at the Indian River Plantation
Resort in Stuart, Florida, were accepted as published in Florida Entomologist 78(1):
207-217. Final reports from the various standing and ad hoc committees of the Society
are presented herein. President Thoms passed the gavel to the new president, Russ
Mizell. The meeting was recessed at 5:50 p.m. and reconvened at 7:00 p.m. Proposed
changes to the FES Bylaws were reviewed by David Hall, Chairman of the Bylaws
Committee. Joe Eger moved that the Bylaws changes be accepted, seconded by Ellen
Thoms; the changes were unanimously accepted by voice vote. No further business
was discussed. The meeting adjourned at 7:15 p.m.

January 1, 1994 to December 31, 1994*

Operating Income
Membership Dues 20,082.50
Subscriptions 6,570.00
Contributions 95.00
Entomology Directories 50.94
Promotional Materials (T Shirts, Posters) 475.34
Miscellaneous 123.32
Less: Refunds (25.50)
Total Operating Income 27,372.60

Other Income
Interest Income 2 477.95
Total Other Income 2 477.95

TOTAL INCOME 28,850.55

Contract Labor 12,452.70
Annual Meeting 4,602.31
T Shirts 3,300.06
Grants/Scholarships 2,500.00
Journal Printing 1,332.11
Editing 1,167.37
Depreciation Expense 717.06
Contributions 450.00
Student Activities 300.00
Postage 248.47
Miscellaneous 200.00

Florida Entomologist 79(1)

Bank Charges 196.75
Office Expenses 144.41
Newsletter 113.86
Dues & Subscriptions 90.00

NET INCOME 2.035.45

Cash-In-Bank 63,439.21
Petty Cash 100.00
TOTAL CASH 63,539.21

Total Fixed Assets (Office Equipment) 1 079.59

TOTAL ASSETS 64.614.80

Full 438
Student 74
Sustaining 48
Honorary 9
Emeritus 11

*As of January 1, 1994, the Society switched to a fiscal year which runs from January
1 to December 31.



The Program for the Seventy-Eighth Annual Meeting of the Society consisted of 5
symposia, 2 mini-symposia, 120 submitted papers, 6 student contest papers. The key
note address was presented by Rodrigo Gamez and the President's Address was pre
sented by Ellen Thoms.


The Honors and Awards Committee recommended and the Florida Entomological
Society recognized the following individuals at the 78th Annual Meeting for their con
tribution to entomology and the Society.

Entomologist of the Year: Jorge E. Pena

Dr. Pena is internationally recognized for his research in the biology, behavior,
ecology, economic thresholds, monitoring methodology and management of a variety
of arthropod pests. He has published more than 50 refereed papers, many on newly

March, 1996

Edited Minutes-78th Annual Meeting

introduced pests in Florida. Some of the diverse pests he has studied are the banana
moth, the rotten sugarcane weevil, the banana weevil, broad mites, the cycad weevil
and the Anona wasp. Dr. Pena has assisted other scientists internationally and has co
ordinated searches for parasitoids. He has served the Florida Entomological Society
in many capacities including most recently as President.

Annual Achievement Award for Extension and Industry: Philip Stansly

Dr. Stansly was recognized for his superior service to Florida producers in the area
of pest management. He has consistently provided accurate and timely information to
citrus and vegetable producers throughout south Florida. Most recently, Dr. Stansly
has provided information to vegetable producers on the management of the silverleaf
whitefly and to citrus producers on the use of biological control for management of the
citrus leaf miner.

Teacher of the Year, Higher Education Category: James E. Lloyd

Dr. Lloyd was selected because he is regarded as a model innovative educator in
how science is most effectively taught. His dedication to teaching is exemplified by the
considerable time he devotes to specialized field trips and on one-on-one assistance for
individualized activities with insects. Dr. Lloyd has developed a course in firefly biol
ogy for the University of Florida's honors program. This course has rapidly become
very popular for the exceptional students in the program.

Teacher of the Year, K 12 Category: Zane Greathouse

Mr. Greathouse was selected as one of 22 middle school teachers statewide to par
ticipate in a National Science Foundation-sponsored Insect Field Biology Institute. At
the Institute he was selected as one of the most effective teachers of the group. At
Wiles Elementary School (Gainesville) where he is a teacher, he has taught hundreds
of students about insects in an extensive outdoor laboratory. He has conducted work
shops to train other teachers in the effective use of insects in the classroom.

Recognition of the Outgoing President for Outstanding Dedicated Service:
Ellen Thorns

President Thoms was recognized for outstanding, dedicated service as President of
the Florida Entomological Society for the 1994-1995 year, culminating in the 78th An
nual Meeting.


We have further consulted with Robert Respess, Tax Accountant, who has pre
pared IRS tax return forms for the first time in the history of the society. The forms
had to be signed by an elected officer of the society, and this was done by Everett
Mitchell, Vice President. The forms were forwarded to Ann Knapp for the addition of
a check for $342.00, i.e., the amount of our taxes. Because we were late, there may be
an additional $20.00 charge.

Florida Entomologist 79(1)

During the course of our road towards being tax exempt, Mr. Respess brought it to
our attention that a tax return need not be filed if the society's annual income was less
than $25,000. Because the society's income had always been less than $25,000, we
slowed our course slightly in our pursuit of tax exempt status. However, for the first
time, in 1994, our income exceeded $25,000! This resulted in the two other society
firsts mentioned above, i.e. filing a tax return, and paying tax to the federal govern
Mr. Respess finally submitted a bill for his services to FES. For 2 years of filling out
and filing forms and consulting, he charged FES only $300. This is quite a bargain.
As a note, Mr. Respess stated that even in tax exempt status, FES will have to con
tinue to file what is termed a notification tax return to IRS. This is just for IRS's in
formation. So, our records must still be kept and the returns filed even though no tax
is paid.
We are continuing towards becoming tax exempt. Because FES has a tax exempt
status in the state of Florida (only for items purchased for resale), the federal status
should not be too involved. We will continue to pursue this, but would like another
round of support from the Executive Committee. The tax committee has been active
for at least two years, which seems like a long time. However, I think we have made
quite a bit of progress towards our goal. With the Executive Committee's consent we
will continue our work.


In all, 6 FES student members received travel grants of $240 for the annual meet
ing in Costa Rica: Kevina Vulinec, Julieta Brambila, Dini Miller, Marco Toapanta, Avi
Eitam, and Guangye Hu.
Three out of seven student applicants were awarded FES $500 scholarships: Dini
Miller, Eliane Quintela, and Robin Goodson.
Three students received mini-grants: Dini Miller, Oscar Liburd, and Denise


Resolution No. 1: WHEREAS the 78th Annual Meeting of the Florida Entomo-
logical Society at Cariari Hotel and Country Club, San Jose, Costa Rica, has enjoyed
outstanding facilities and hospitality which immensely contributed to the success of
the meeting, AND WHEREAS Ms. C. Vasquez, Local Sales Executive, generously gave
her time and effort to welcome the Society to Cariari Hotel and Country Club at the
opening of our 78th Annual Meeting, AND WHEREAS J. Arias generously gave his
time and effort to plan the field trips with INBio, CATIE, EARTH, and OS, AND
WHEREAS INBio, CATIE, EARTH, and OTS faculty and/or personnel generously
gave their time and effort to plan and guide the field trip and share their organize
tions with the Society, THEREFORE, BE IT RESOLVED that the Secretary of the So
city be instructed to forward a copy of the resolution to C. Vasquez and the president
of the Society forward letters of appreciation to C. Vasquez, Cariari Hotel and Country
Club; J. Arias; Dr. C Schnell, OTS; Dr. R. Gamez, INBio; Dr. O Ramirez, CATIE; and
Prof. E. Alvarado, EARTH.
Resolution No. 2: WHEREAS E. Mitchell and the Local Arrangements commit
tee have provided excellent organization and facilities for the 78th Annual Meeting of
the Society, AND WHEREAS R. Mizell, III, and the Program Committee have pre

March, 1996

Edited Minutes-78th Annual Meeting

pared a well-balanced, high quality program for the Society's meeting, AND
WHEREAS R. Gamez, Director of INBio, presented an excellent keynote address,
AND WHEREAS the speakers who presented papers, both invited and submitted,
shared their outstanding work and ideas with our Society, AND WHEREAS excellent
and timely symposia and workshops were organized by W. Klassen, W. Whitcomb, J.
Knapp, J. H. Tsai, and R. K. Jansson, AND WHEREAS the Committee on Student Ac
tivities encouraged excellent student participation in, and contributions to, our An
nual Meeting, THEREFORE BE IT RESOLVED that the Society expresses its deepest
appreciation to these individuals.
Resolution No. 3: WHEREAS President E. M. Thoms and other members of the
Executive Committee have provided our Society with dedicated leadership and in
valuable service, AND WHEREAS A. C. Knapp has once again done an outstanding
job as Business Manager, AND WHEREAS C. Lofgren and the Associate Editors of
Florida Entomologist have done an exceptional job in maintaining the highest stan
dards for the Society's journal, AND WHEREAS L. Wood and P. Parkman have ex
celled in the production of the informative and timely newsletter of the Society, AND
WHEREAS members of other committees and individuals have generously contrib
uted their time and efforts to the Society this past year, THEREFORE BE IT RE
SOLVED that the Society commends these individuals and expresses its appreciation
for their service to the Society and to the Science of Entomology.
Resolution No. 4: WHEREAS members of the industry continue to provide much
needed financial support to the Society by way of Sustaining Memberships, advertis
ing in the program, support for the journal and numerous other Society functions,
THEREFORE BE IT RESOLVED that the Society hereby expresses its appreciation
to these groups.


The Sustaining Membership Committee is happy to report that five new Sustain
ing Members joined this year. They are:

AHO Enterprises, Inc. Mr. Tom Aho
Nozzle Nolen, Inc. Mr. Mickey Nolen
The Scotts Company -Dr. Wayne Mixson
Southern Agricultural, Inc. Mr. John Diem
Lemont Entomology Services -Mr. Brian Lemont

There is currently a total of 52 Sustaining Members in the Society.


The Florida Entomological Society gained 40 new regular members and 20 new
student members during 1994-95.


Volume 77 of the Florida Entomologist consisted of 532 pages including 33 re
search papers, 18 scientific notes, 12 symposia papers, 8 book reviews, 4 in memorial,
and the minutes of the 1994 Annual Meeting.

Florida Entomologist 79(1)

Seventy-six research papers and scientific notes were received in 1994, not include
ing symposia papers. Of these, 56 have been published, 3 accepted for publication, and
two rejected or withdrawn. The remaining 15 papers remain in the review process. A
review of the time required to process these papers revealed that the median time
from my receipt of the paper to actual publication was 10 months. Six of the papers
were published in 5-to-6 months, 14 in 7-to-8 months, 20 in 9-to-10 months, 10 in 11
to 12 months, and 9 in 13-to 16 months. Five of these papers were from scientists from
South or Central America and took 11 to 15 months to publish. Thus far, for the year
1995, I have received 39 papers; seven have been published and one rejected.
Our new billing procedure that requires scientists to provide accurate information
on billing addresses has been a big success. The Painter Printing Company, who does
all the billing for page charges, has informed me that all page-charge bills have been
paid through the March, 1995 issue, as well as, almost all the page charges for the
June issue.


The following members were nominated for the following offices:
Jerry Hogsette and David Hall for Vice-President
Susan Broda-Hydorn and Philip Stansly for Member-at-Large

Fifty-two ballots were received. The selection went as follows:
Vice President: David Hall
Member-at-Large: Philip Stansly


A petition from 10 FES members was presented to the Executive Committee re
questing four changes to FES Bylaws. The general membership was notified of the
proposed changes several times during the year via the FES Newsletter. A formal an
nouncement of the proposed changes issued by Secretary Petitt was carried in the
June issue of the Newsletter. The proposed changes were as follows:

1. Designate the President-Elect as the Chairperson of the Annual Meeting
(Bylaws Chapter II, Section 2).
2. Designate the Vice President as the Assistant Chairperson of the Annual
Meeting (Bylaws Chapter II, Section 3).
3. Remove from the Secretary the responsibility of maintaining a current list
of members and addresses (Bylaws Chapter II, Section 4).
4. Drop the Rules of Order Committee (Bylaws Chapter IV, Section 9).

Specific revisions to the Bylaws needed to meet the above four changes were pre
sented to the Executive Committee on February 21, 1995. A two thirds affirmative
vote at the 1995 Annual Business Meeting was required to sanction the amendments.
The proposed changes were reviewed at the Annual Meeting and a quorum was
present. Joe Eger moved to accept the changes; seconded by Ellen Thoms; unani
mously passed. The revised Bylaws will be published in an upcoming issue of the Flor
ida Entomologist.

March, 1996

Edited Minutes-78th Annual Meeting


The Seventy-Ninth Annual Meeting of the Society will be held from August 5-8,
1996, at the Sheraton Sand Key near Clearwater.
These minutes of the 78th Annual Meeting of the Florida Entomological Society
were reviewed by the 1995-96 Executive Committee on January 17, 1996.


H. A. Denmark
W. G. Eden
Eugene Gerberg
L. A. Hetrick
L. C. Kuitert
F. Mead
A. J. Rogers
A. G. Selhime
H. V. Weems
D. 0. Wolfenbarger


27 September 1994, Gainesville
29 November 1994, Lake Alfred
21 February 1995, Gainesville
9 May 1995, Lake Alfred
6 August 1995, San Jose, Costa Rica

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