Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00037
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 2000
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: VID00037
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

Morales-Ramos et al.: Two populations of C. grandis


Southern Regional Research Center
New Orleans, Louisiana


A new colony of the boll weevil ectoparasitoid Catolaccus grandis was introduced
from Guasave, Sinaloa, Mexico to improve vigor of a 12-year-old laboratory reared
stock in Weslaco, Texas. The biological characteristics of the introduced colony were
compared to those of the Weslaco colony and a crossbreed of these 2 colonies. Devel-
opmental time was not significantly different among the 3 colonies, but the preovipo-
sitional period of the Sinaloa females was 3 times as long compared to the other 2
colonies. The fecundity, net reproductive rate (R.), and intrinsic rate of increase (r.) of
females from Sinaloa were significantly lower than those of females from Weslaco and
the hybrid colony. Generation time (G) and doubling time (DT) were significantly
longer in the Sinaloa colony. These characteristics make the Sinaloa population less
desirable for mass propagation and release to control boll weevil populations than the
Weslaco colony. The biological and population parameters of the hybrid colony were
not significantly different from those of the Weslaco population. The implications of
the observed results on the mass propagation and release strategies against the boll
weevil are discussed and recommendations are presented.

Key Words: Boll weevil, parasitoid, population, variability, biological parameters, bi-
ological control


Una nueva colonia del parasitoide del picudo del algodonero Catolaccus grandis de
Guasave, Sinaloa, M6xico, fue introducida a Texas con el objetivo de mejorar el vigor
de la colonia madre de este parasitoide. Las caracteristicas biol6gicas de la colonia in-
troducida fueron comparadas con aquellas de la colonia de Weslaco, Texas y el hibrido
entire estas dos colonies. No hubo diferencias significativas en el tiempo de desarrollo
entire las tres colonies, pero el period de preoviposici6n de las hembras de Sinaloa fue
el triple del observado en las otras 2 colonies. La fecundidad, tasa reproductive neta
(Ro) y tasa intrinseca de crecimiento (r.) de las hembras de Sinaloa, fueron significan-
temente bajas comparadas con las de las hembras hibridas y de Weslaco. Los tiempos
de generaci6n (G) y de duplicaci6n (DT) fueron significantemente mas largos en la co-
lonia de Sinaloa. Estas caracteristicas indican que el uso de la colonia de Sinaloa en
un program de propagaci6n en masa y liberaci6n contra el picudo del algodonero po-
dria ser desventajoso. Sin embargo, las caracteristicas biol6gicas de la colonia hibrida
no fueron significativamente diferentes a las observadas en la colonia de Weslaco. Se
discuten las implicaciones de estos resultados en las estrategias de propagaci6n en
masa y liberaci6n contra el picudo del algodonero y se presentan recomendaciones.

Augmentative releases of the exotic ectoparasitoid Catolaccus grandis (Burks)
(Hymenoptera: Pteromalidae) have proven effective at biologically controlling the boll

Florida Entomologist 83(2)

June, 2000

weevil, Anthonomus grandis grandis Boheman (Coleman et al. 1996, King et al. 1995,
Morales-Ramos & King 1991, Morales-Ramos et al. 1994, 1995, Summy et al. 1992,
1993, 1994, 1995, Vargas-Camplis et al. 1997). Rearing methods for this parasitoid
have been improved over the years (Cate 1987, Morales-Ramos et al. 1992, 1994,
1997, Palamara 1995, Roberson & Harsh 1993). The development of an artificial diet
(Rojas et al. 1996, 1997) significantly advanced the mass propagation technology for
this parasitoid. The effectiveness of diet-reared C. grandis has been established by
laboratory and field evaluations (Morales-Ramos et al. 1995, 1998, R. J. Coleman, un-
published). All these advances have made the mass propagation of C. grandis econom-
ically feasible according to an economic analysis done by Ellis et al. (1997).
The process of colonization and mass rearing over long periods of time may have
detrimental effects in reared species (Bartlett 1984). Initial loss of genetic variability
and subsequent selection tend to induce domestication and adaptation to laboratory
environments (Bartlett 1984, van Lenteren 1991, Wajnberg 1991, Lappla 1993). After
12 years of constant laboratory rearing of C. grandis, there is no evidence of loss of
searching capacity or field adaptation in this parasitoid (Morales-Ramos et al. 1995,
Vargas-Camplis et al. 1997); however, the potential for loss of genetic variability and
subsequent adaptation to laboratory environments is a constant concern. Some meth-
ods for avoiding such problems include 1) pooling multiple-founder colonies, 2) main-
taining variable laboratory environments, and 3) regular infusion of wild genetic
stock (Joslyn 1984).
The first method has been used with the colony of C. grandis that is currently
maintained in Weslaco, TX. This colony is the result of systematic cross breeding
among 3 distinct populations of C. grandis from Tabasco and Oaxaca Mexico, and from
El Salvador (J. A. M., unpublished data) (see Materials and Methods). The second
method is difficult to apply because it requires large supplies of equipment, space, and
personnel. These resources have not been available to the present research program.
This study is an attempt to apply the third method by introducing wild C. grandis
from Sinaloa, Mexico to prevent the loss of genetic variability from the Weslaco colony.
The objectives were to evaluate the reproductive potential of the newly introduced
parasitoids to determine the desirability of cross breeding the wild population from
Sinaloa with the Weslaco population.


Boll weevil larvae used in this study were reared on artificial diet at the USDA-
ARS, Biological Control and Mass Rearing Research Unit at Mississippi State, Mis-
sissippi (Roberson & Wright 1984). Rearing techniques for C. grandis were as re-
ported by Morales-Ramos et al. (1992) using the Parafilm encapsulation method
developed by Cate (1987). Parasitoid colonies and experiments were held at constant
27 + 1C, 50 + 10% R.H., and a photoperiod of 14:10 (L:D).

Biological Materials Origin

A C. grandis colony was established in Guasave, Sinaloa, Mexico in early June,
1996 using parasitoids isolated from boll weevil infested cotton squares and bolls and
from net captures of adult C. grandis females. The infested cotton material and adult
parasitoids were collected from a commercial cotton field located near Guasave. The
colony was maintained and increased using Cate's (1987) method of encapsulation.
Two shipments of C. grandis were received at the APHIS quarantine facility located
at the Biological Control Center in Mission, TX (importation permit No. 31352). The ship-

Morales-Ramos et al.: Two populations of C. grandis

ments consisted of 120 females and 136 males on August 1st and 144 females and 106
males onAugust 8, 1996. The C.grandis colony was held in quarantine for one generation
while being screened for purity and microbial contamination. The Sinaloa colony of C.
grandis (Sinaloa population) was released from quarantine on October 25, 1996 (permit
No. 31669) to the USDA-ARS Subtropical Agricultural Research Center in Weslaco, TX.
The parasitoids used for this study were from the 2nd generation after introduction.
The mother colony maintained at Weslaco, TX (Weslaco population) originated
from 2 different localities in Southeast Mexico and one locality from El Salvador. The
localities in Mexico were Cardenas, Tabasco from Hampea nutricia (29 males and 13
females); La Ventosa, Oaxaca from Cienfuegosia rosei (1 male and 7 females); and in
El Salvador, The University of El Salvador, from cultivated and wild cotton (2 males
and 2 females). Samples of plant buds and fruits infested by boll weevil were hand-
carried to the quarantine facility at Texas A&M University in College Station, TX.
These colonies were reared independently for 3 years at the Department of Entomol-
ogy, Texas A&M University, from 1985 to 1988.
A crossbred colony was produced in 1989 from a systematic cross between the 3 pop-
ulations. This was accomplished by placing 300 newly emerged females from one colony
and 300 newly-emerged males from another in a new cage. A total of 6 combinations
were initially created. The progeny of all the 6 combinations were then mixed and
reared as a single colony. This colony was successfully transferred to Weslaco, TX in Feb-
ruary 1990 and it has been in constant culture since (approximately 153 generations).
Several permits have been issued to release this particular population in the state of
Texas. The most recent one was issued on September 21, 1994 (permit No. 944483).
The third C. grandis colony tested was a crossbreed (hybrid) of the Sinaloa (12th
generation) and Weslaco (generation 169) populations. The hybridization was accom-
plished by placing 100 female pupae of each population in separated cages made of
2.8-liter Rubbermaid containers (No. 6 Rubbermaid, Wooster, OH). The pupae were
allowed to emerge at 27 + 1C, 60 + 10% RH, and a photoperiod of 14:10 (L:D). One d
after emergence, the 100 females from the Weslaco population received 50 newly-
emerged males from the Sinaloa population. Similarly, the 100 females from the Si-
naloa population were provided with 50 newly emerged males from the Weslaco pop-
ulation. Each cage was provided with distilled water, honey and 120 encapsulated boll
weevil larvae, which were replaced daily for a period of 20 d. The parasitized boll wee-
vils were maintained at the conditions described above for 10 d to allow the parasi-
toids to develop to mid-age pupae. Eighteen female parasitoid pupae were randomly
selected from weevils parasitized by each of the 2 groups when the females were 8 d
old. This procedure was repeated again using weevils parasitized when the females
were 12 d old to obtain a total of 72 hybrid C. grandis pupae.

Pupal Weight

A total of 72 female parasitoid pupae of each population (3 d-old) were weighed in-
dividually using a Mettler H51 precision balance. The weights of the different groups
were analyzed by Analysis of variance (ANOVA) and the means were compared by
Tukey's test using SigmaStat' software (Jandel Corporation 1995). The female para-
sitoid pupae were placed individually in plastic square Petri dishes (9 x 9 cm) where
they completed development at the conditions described above.

Fecundity and Progeny Sex Ratio

Once the female parasitoids completed development and emerged, two males from
the same population were placed in each of the Petri dishes to ensure fertilization.

Florida Entomologist 83(2)

June, 2000

Each female was provided daily with 12 encapsulated boll weevils, water and honey
according to evaluation methods reported by Morales-Ramos and Cate (1992). Dead
females were not replaced, but dead males were replaced during the first 15 d. Each
day, the Parafilm capsules enclosing the parasitized weevils were opened to count
the number of eggs oviposited per female. They were resealed and returned to the en-
vironmental chamber for parasitoid development. Nine d later, the Parafilm cap-
sules were reopened to count and sex the parasitoid pupae. The number of eggs
oviposited per female per day and the number and sex of developing progeny were re-
corded over a 45-d period.
The sample size needed to estimate the population mean (i) of eggs/female and
eggs/female/d for confidence intervals (E) of + 20 and 1.5, respectively, with a = 0.05,
was determined by using the equation:

n ((Za/2)2 O2)/E2

where n is the sample size, Z,2 = 1.96 (from tables), and o is the population standard
deviation (estimated from sample 's'). This equation determines the adequate sample
size for a given value of E based on the standard deviation of a preliminary samples
from each population (Ott 1984).
The total number of eggs oviposited by each female during the 45-d period and the
mean number of eggs oviposited per day during the fecundity plateau period (as de-
fined by Morales-Ramos and Cate 1992) were used to compare the fecundity of fe-
males from each of the 3 populations studied. The starting age of the fecundity
plateau period was determined according to the criteria used by Morales-Ramos and
Cate (1992). The sex of each of the female's progeny was recorded and the sex ratio of
the progeny was calculated. The Analysis of variance was used to compare fecundity
and progeny sex ratio between the three populations studied and Tukey's test was
used to compare the means using SigmaStat software.


Biological Parameters

Developmental time of C. grandis females was not significantly different among
the Sinaloa and Weslaco populations and their hybrid (Table 1). The preovipositional
period, on the other hand, was significantly longer (16 days) in the Sinaloa population
than in the Weslaco and hybrid populations (4.3 and 3.6 d respectively) (F = 64.4, df
= 2, 126, P < 0.001) (Table 1). The difference observed between the Weslaco and hybrid
populations was not significant, showing that the hybrids tend to resemble more the
Weslaco population for this parameter instead of showing an intermediate value be-
tween the Weslaco and Sinaloa populations as expected.
A longer preovipositional period is considered to be an undesirable trait. In many
parasitoid species, the females do not respond to host cues during this period (Vinson
1981, 1984). Evidence from field studies on searching capacity indicated that this may
be the case in C. grandis; no parasitism of boll weevil larvae was observed in an ex-
perimental field in Ricardo, TX up to 5 d after the release of newly emerged females
(J. A. M. unpublished data). However, a release of 5-d old parasitoids produced high
rates of parasitism during the same period of time in the same experimental field
(J. A. M., unpublished data).
No significant differences in female longevity and pupal weight were observed
among the Sinaloa and Weslaco populations and their hybrids (Table 1). Parasitoid

Morales-Ramos et al.: Two populations of C. grandis



Parameter Weslaco Hybrid Sinaloa

Developmental time 13.4 + 0.6 13.5 + 0.7 13.8 + 1.3
Preovipositional period' 4.3 + 3.4b 3.6 + 4.3b 16.0+ 10.1a
Longevity' 43.2 + 20.7a 44.1 + 16.4a 35.9 + 22.6a


Total eggs 554.0 247a 600.1 253a 242.0 300b
Eggs/F./d2 22.0 + 8.8a 21.6 + 9.8a 14.5 + 13.5b
Pupal weight3 6.4 + 1.1a 6.3 + 1.0a 6.6 + 1.5a
Progeny sex ratio4 5.1 + 4.4a 4.5 + 3.6a 4.6 + 4.3a

X + SD, means with the same letter are not significantly different after ANOVA Tukey test a = 0.05.
In days.
During the fecundity plateau period.
In mg.
In females per male.

females from the Weslaco and hybrid populations oviposited a significantly higher
number of eggs (554 and 600 respectively) than that of females from the Sinaloa pop-
ulation (242) (F = 37.6, df= 2, 213, P = 0.001) (Table 1). Total fecundity was not sig-
nificantly different between the Weslaco and hybrid populations, again, showing
resemblance of the hybrids to the Weslaco population.
Parasitoids from the Weslaco and hybrid populations produced most of their prog-
eny earlier in their life cycle than those of the Sinaloa population (Fig. 1A). The fecun-
dity plateau period (period of highest fecundity) started earlier in the Weslaco and
hybrid populations (9 and 8 d of age, respectively) than in the Sinaloa population (23 d
of age). The daily oviposition of Weslaco and hybrid parasitoids during the fecundity
plateau period was significantly higher (22.0 and 21.6 eggs/d, respectively) than that
of parasitoids from the Sinaloa population (14.5 eggs/d) (F = 95.3, df = 2, 2103, P <
0.001) (Table 1). The daily oviposition rate during this period was not significantly dif-
ferent between the Weslaco and hybrid populations.
A reduced oviposition rate is another undesirable trait. The effectiveness of C. gran-
dis in controlling the boll weevil depends on the rate at which the females find and par-
asitize host larvae. Females from the Sinaloa population seem to parasitize at nearly
half the rate as females from the Weslaco population, and at more than twice the age.
There was no significant difference in progeny sex ratio among the 3 populations
(Table 1). This indicates that the rates of fertilization and host acceptance were not af-
fected by hybridization.
These results indicate that the undesirable characteristics of the Sinaloa popula-
tion were not manifested after hybridization with the Weslaco population. However,
we recommend caution in introducing the Sinaloa strain to the mother colony
(Weslaco). The detrimental characteristics observed in the Sinaloa population may re-
surface in subsequent generations in the hybrid population. Since the only apparent
potential advantages of introducing the Sinaloa strain to the mother colony is gain of
genetic variability, we recommend the use of a different strain to achieve this purpose.

Florida Entomologist 83(2)

2 2

300 -

300-*- Weslaco

2*40 Sinaloa

180 -

0 10 20 30 40

Age in Days

Fig. 1. Age-dependent fecundity and survival of females of 2 C. grandis popula-
tions and their hybrid. A) Age-dependent fecundity expressed as eggs oviposited per
female per day; B) age-dependent survival expressed as 1, which is the proportion of
individuals surviving from eclosion to the beginning of age x (in days); and C) Age-de-
pendent reproductive value (Vx), which is the potential female progeny of a female of
age x (in days).
age x (in days).

June, 2000

Morales-Ramos et al.: Two populations of C. grandis


We thank Ma. Teresa Chavez technical director ofAsesoria Biol6gica Integral from
Guasave, Sinaloa, M6xico for her assistance in collecting, rearing, and shipping of the
Sinaloa specimens of C. grandis. We also thank Nina M. Barcenas from the Colegio de
Postgraduados, Montecillo, M6xico for her help in providing training and materials to
the contacts in Sinaloa. This was accomplished with financial support of USDA-
RSED-FAS Grant No. FG-Mx-101 Project No. MX-ARS-2.


BARTLETT, A. C. 1984. Genetic changes during insect domestication, pp. 2-8 in E. G.
King and N. C. Leppla (eds.), Advances and challenges in insect rearing. USDA,
ARS, New Orleans, LA.
CATE, J. R. 1987. A method of rearing parasitoids of boll weevil without the host plant.
Southwest. Entomol. 12: 211-215.
COLEMAN, R. J., J. A. MORALES-RAMOS, E. G. KING, AND L. A. WOOD. 1996. Suppres-
sion of the boll weevil in organic cotton by release of Catolaccus grandis as part
of the Southern Rolling Plains boll weevil eradication program. p. 1094 in Proc.
Beltwide Cotton Conferences 1996 Vol. 2. National Cotton Council of America,
Memphis, TN.
timated economical feasibility of Catolaccus grandis in control of the boll wee-
vil. Environmental Issues/Sustainability Team Technical Report 97-1, Texas
A&M University Agricultural Program. College Station, TX.
JANDEL CORPORATION. 1995. SigmaStat statistical software version 2.0 for Windows
95, NT & 3.1 user's manual. Jandel Scientific, San Rafael, CA.
JOSLYN, D. J. 1984. Maintenance of genetic variability in reared insects, pp. 20-29 in
E. G. King and N. C. Leppla (eds.), Advances and challenges in insect rearing.
USDA, ARS, New Orleans, LA.
1995. Suppression of the boll weevil in commercial cotton by augmentative re-
leases of the wasp parasite, Catolaccus grandis, pp. 26-30 in Addendum to the
Proc. Beltwide Cotton Conferences 1995. National Cotton Council of America,
Memphis, TN.
LEPPLA, N. C. 1993. Principles of quality control in mass-reared arthropods, pp. 1-11
in G. Nicoli, M. Bebuzzi and N. C. Leppla (eds.), Proceedings of the Seventh
Workshop of the IOBC Global Working Group "Quality Control of Mass Reared
Arthropods". International Organization for Biological Control of Noxious An-
imals and Plants, Rimini, Italy.
LOTKA, A. J. 1907. Studies on the mode of growth of material aggregates. Am. J. Sci.
24: 199-216.
MORALES-RAMOS, J. A. 1997. Caracterizaci6n Biol6gica de Catolaccus grandis (Hy-
menoptera: Pteromalidae) y el Uso de Modelos en el Desarrollo de un Programa
de Control Biol6gico del Picudo del Algodonero (Coleoptera: Curculionidae).
Vedalia 4: 35-43.
MORALES-RAMOS, J. A., AND E. G. KING. 1991. Evaluation of Catolaccus grandis
(Burks) as a biological control agent against the cotton boll weevil, P. 724 in
Proc. Beltwide Cotton Conferences 1991 Vol. 2. National Cotton Council of
America, Memphis, TN.
MORALES-RAMOS, J. A., AND J. R. CATE. 1992. Laboratory determination of age-depen-
dent fecundity, development, and rate of increase of Catolaccus grandis (Burks)
(Hymenoptera: Pteromalidae). Ann. Entomol. Soc. Am. 85: 469-476.
1992. Feasibility of mass rearing Catolaccus grandis, a parasitoid of the boll

Florida Entomologist 83(2)

June, 2000

weevil, pp. 723-726 in Proc. Beltwide Cotton Conferences 1992 Vol. 2. National
Cotton Council of America, Memphis, TN.
SUMMY, AND J. R. BRAZZEL. 1994. Suppression of boll weevil first generation by
augmentative releases of Catolaccus grandis in Aliceville, Alabama, pp. 958-
964 in Proc. Beltwide Cotton Conferences 1994 Vol. 2. National Cotton Council
of America, Memphis, TN.
MORALES-RAMOS, J. A., K. R. SUMMY, AND E. G. KING. 1995. Estimating parasitism by
Catolaccus grandis (Hymenoptera: Pteromalidae) after inundative releases
against the boll weevil (Coleoptera: Curculionidae). Environ. Entomol. 24:
MORALES-RAMOS, J. A., K. R. SUMMY, AND E. G. KING. 1996. ARCASIM, a Model to
Evaluate Augmentation Strategies of the Parasitoid Catolaccus grandis
Against Boll Weevil Populations. Ecological Modelling 93: 221-235.
MORALES-RAMOS, J. A., M. G. ROJAS, AND E. G. KING. 1997. Mass propagation of Cato-
laccus grandis in support of large-scale area suppression of the boll weevil in
south Texas cotton, pp. 1199-1200 in Proc. Beltwide Cotton Conferences 1997
Vol. 2. National Cotton Council of America, Memphis, TN.
MORALES-RAMOS, J. A., M. G. ROJAS, R. J. COLEMAN, AND E. G. KING. 1998. Potential
use of in vitro-reared Catolaccus grandis (Hymenoptera: Pteromalidae) for bi-
ological control of the boll weevil (Coleoptera: Curculionidae). J. Econ. Entomol.
91: 101-109.
OTT, L. 1984. An introduction to statistical methods and data analysis, second Ed.
Duxbury Press, Boston, MA.
PALAMARA, K. J. 1995. Mass propagation of the boll weevil parasite, Catolaccus gran-
dis, pp. 30-34 in Addendum to the Proc. Beltwide Cotton Conferences 1995. Na-
tional Cotton Council of America, Memphis, TN.
ROBERSON, J. L., AND D. K. HARSH. 1993. Mechanized production processes to encap-
sulate boll weevil larvae (Anthonomus grandis) for mass production of Catolac-
cus grandis (Burks) pp. 922-923 in Proc. Beltwide Cotton Conferences 1993 Vol.
2. National Cotton Council of America, Memphis, TN.
ROBERSON, J. L., AND J. E. WRIGHT. 1984. Production of boll weevils, Anthonomus
grandisgrandis, pp. 188-192 in E. G. King and N. C. Leppla [eds.] advances and
challenges in insect rearing. USDA, ARS, New Orleans, LA.
ROJAS, M. G., J. A. MORALES-RAMOS, AND E. G. KING. 1996. In vitro rearing of the boll
weevil (Coleoptera: Curculionidae) ectoparasitoid Catolaccus grandis (Hy-
menoptera: Pteromalidae) on meridic diets. J. Econ. Entomol. 89: 1095-1104.
ROJAS, M. G., J. A. MORALES-RAMOS, AND E. G. KING. 1997. Reduction of cost for mass
propagating Catolaccus grandis by the use of artificial diet, pp. 1197-1199 in
Proc. Beltwide Cotton Conferences 1997 Vol. 2. National Cotton Council of
America, Memphis, TN.
SUMMY, K. R., J. A. MORALES-RAMOS, AND E. G. KING. 1992. Ecology and potential im-
pact of Catolaccus grandis (Burks) on boll weevil infestations in the Lower Rio
Grande Valley. Southwest. Entomol. 17: 279-288.
SUMMY, K. R., J. A. MORALES-RAMOS, AND E. G. KING. 1993. Suppression of boll wee-
vil infestations by augmentative release of Catolaccus grandis, pp. 908-909 in
Proc. Beltwide Cotton Conferences 1993 Vol. 2. National Cotton Council of
America, Memphis, TN.
1994. Integration of boll weevil parasite augmentation into the short-season
production system of the Lower Rio Grande Valley, pp. 953-957 in Proc. Belt-
wide Cotton Conferences 1994 Vol. 2. National Cotton Council of America,
Memphis, TN.
SUMMY, K. R., J. A. MORALES-RAMOS, AND E. G. KING. 1995. Suppression of boll wee-
vil infestations on South Texas cotton by augmentative releases of the exotic
parasite Catolaccus grandis (Hymenoptera: Pteromalidae). Biological Control
5: 523-529.

Smith & McSorley: Trap and Barrier Crops for Whitefly 145

VAN LENTEREN, J. C. 1991. Quality control of natural enemies: hope or illusion?, pp.
1-14 in F. Bigler (ed.), Proceedings of the Fifth Workshop of the IOBC Global
Working Group "Quality Control of Mass Reared Arthropods". International
Organization for Biological Control of Noxious Animals and Plants, Wagenin-
gen, Netherlands.
Life table analysis of cotton boll weevil in the tropics of Tamaulipas, Mexico af-
ter Catolaccus grandis releases, pp. 1194-1197 in Proc. Beltwide Cotton Con-
ferences 1997 Vol. 2. National Cotton Council of America, Memphis, TN.
VINSON, S. B. 1981. Habitat location, pp. 51-77 in D. A. Norlund, R. L. Jones, and W.
J. Lewis (eds.), Semiochemicals their role in pest control. John Wiley & Sons,
New York, NY.
VINSON, S. B. 1984. Parasitoid-host relationship, pp. 205-233 in W. J. Bell and R. T.
Card6 (eds.), Chemical Ecology of Insects. Sinauer Associates Inc. Sunderland,
WAJNBERG, E. 1991. Quality control of mass-reared arthropods: a genetical and sta-
tistical approach, pp. 15-25 in F. Bigler (ed.), Proceedings of the Fifth Workshop
of the IOBC Global Working Group "Quality Control of Mass Reared Arthro-
pods". International Organization for Biological Control of Noxious Animals
and Plants, Wageningen, Netherlands.

Smith & McSorley: Trap and Barrier Crops for Whitefly


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


Trap crops and barrier crops are among the cultural control methods promoted for
management of Bemisia argentifolii Bellows & Perring, particularly for small farmers
in the tropics. In 1996 eggplant, Solanum melongena L., was tested as a trap crop, and
in 1996 and 1997 corn, Zea mays L., was tested as a barrier crop for management of
B. argentifolii on bean, Phaseolus vulgaris L. In 1996 treatments were compared by
sampling immature B. argentifolii on bean leaves. Neither egg nor nymphal densities
were reduced by eggplant or corn treatments in 1996. In the 1997 corn barrier trial
plot size was increased and the orientation of barrier row to wind direction was eval-
uated. A dust-and-release procedure was used to measure entry of greenhouse-reared
adult B. argentifolii into experimental plots. Counts from yellow sticky traps in 1997
indicated that migration by adult whiteflies into plots was determined primarily by
air currents and was only marginally influenced by the presence of a corn barrier. The
results indicate that barrier crops and certain trap crops may have limited value for
whitefly management.

Key Words: Intercropping, polyculture, vector management, wind dispersal, pest

Florida Entomologist 83(2)

June, 2000


El uso de cultivos trampa y cultivos de barrera se esta promoviendo como media
de control de la mosquita blanca (Bemissia argentifolii Bellows & Perring), principal-
mente entire pequenos agricultores de los tr6picos. En 1996 se prob6 el cultivo de be-
renjena (Solanum melongena L.) como plant trampa y en 1996 y 1997 se utilize maiz
(Zea mays L.) como cultivo barrera para el control de B. argentifolii en un campo de fri-
jol, Phaseolus vulgaris L. En 1996 se colectaron muestras de B. argentifolii en fase in-
madura de hojas de frijol para comparar el efecto de los tratamientos. En 1996 no se
logr6 reducir la densidad de huevecillos o ninfas en el frijol al emplear berenjena o
maiz. En 1997 se aument6 el tamano de la parcela experimental de maiz y se evalu6 el
efecto de la orientaci6n del surco barrera en relaci6n a la direcci6n del viento. La can-
tidad de adults de B. argentifolii que entraron a los lotes experimentales se cuantifico6
mediante un procedimiento de espolvoreo y liberaci6n (dust-and-release). En 1997 se
encontr6 que la migraci6n de adults de mosquita blanca hacia los lotes experimenta-
les fue determinada principalmente por corrientes de aire y que la presencia de maiz
barrera tuvo muy poco efecto en controlar su entrada. Los resultados indican que los
cultivos barrera y trampa fueron poco efectivos en el control de la mosquita blanca.

Bemisia argentifolii Bellows & Perring, also known as the B strain of B. tabaci
(Gennadius), causes significant economic damage to agronomic and horticultural
crops throughout warm regions of the world (Brown et al. 1995). Bemisia argentifolii
is a phloem-feeder which vectors numerous geminiviruses and inflicts a variety of
plant disorders as well as mechanical damage (Byrne et al. 1990, Hiebert et al. 1996,
Shapiro 1996). It has demonstrated resistance to most classes of pesticides (Denholm
et al. 1996), forcing growers and researchers to evaluate alternative methods of con-
trol. Attempts to manage whiteflies by cultural means have included the use of trap
crops (Al-Musa 1982, Ellsworth et al. 1994, McAuslane et al. 1995, Schuster et al.
1996) and barrier crops (Sharma & Varma 1984, Fargette & Fauquet 1988, Rataul et
al. 1989, Morales et al. 1993).
Trap crops are preferred host plants which are used to draw an herbivore away
from a less-preferred main crop (Vandermeer 1989). Bemisia argentifolii has been ob-
served to oviposit heavily on eggplant, Solanum melongena L. (Tsai & Wang 1996),
leading researchers to suggest eggplant as a promising trap crop (Faust 1992).
Whiteflies are weak fliers, relying on air currents for both short and long distance
migration (Byrne & Bellows 1991, Byrne et al. 1996). Several tall-growing non-host
plants, primarily in the family Gramineae, have been tested as barrier crops or inter-
crops to reduce whitefly colonization and virus transmission among main crops. Re-
sults have been mixed. Morales et al. (1993) reported that a sorghum, Sorghum
bicolor (L.) Moench, barrier reduced B. tabaci densities and transmission of virus, on
tomatoes, Lycopersicon esculentum Mill. A pearl millet, Pennisetum typhoides (Burm.
f.) Stapf & Hubbard, barrier reduced whitefly virus transmission on cowpea, Vigna
unguiculata (L.) Walp. (Sharma & Varma 1984) and on soybean, Glycine max (L.) Mer-
rill (Rataul et al. 1989). Gold et al. (1990) found reduced densities of Aleurotrachelis
socialis Bondar and Trialeurodes variabilis (Quaintance) on cassava, Manihot escu-
lenta Crantz, intercropped with maize, Zea mays L., and cowpea, but attributed this
in part to reduced host quality due to intercrop competition. Fargette & Fauquet
(1988), whose study included the effect of wind direction, found densities of B. tabaci
and virus incidence were sometimes higher on cassava intercropped with maize than
on monocropped cassava.

Smith & McSorley: Trap and Barrier Crops for Whitefly 147

These studies have been carried out primarily in the tropics, where safe, inexpen-
sive cultural control measures are a priority for low resource farmers. Extension ma-
terial from Central America promotes the use of crop barriers as a component of
whitefly management programs (Salguero 1993). The present study was undertaken
in 1996 to test the usefulness of eggplant as a trap crop and field corn as a barrier crop
for management of B. argentifolii on common bean, Phaseolus vulgaris L. It was con-
tinued in 1997 focusing only on the barrier crop treatment and including the effects
of wind direction and barrier row orientation.


Research Design and Plot Management, 1996

The experiment was carried out at the University of Florida Green Acres Agronomy
Research Farm, northwest of Gainesville, FL (29040'N, 8230'W). Four treatments
were compared: 1) bean planted in monoculture, 2) bean intercropped with eggplant,
3) bean intercropped with field corn, and 4) bean monoculture treated with imidaclo-
prid (Provado 1.6F, Bayer, Kansas City, MO), a systemic insecticide. The imidacloprid
treatment was included for yield comparison only. It was not sampled for whiteflies.
'Espada' bean (Harris Seed, Rochester, NY) was used in the monoculture and inter-
crop treatments. 'Black Beauty' eggplant (Ferry-Morse Seed, Fulton, KY) was tested
as a trap crop in one intercrop treatment. The subtropical field corn hybrid Howard II-
IST (Gallaher et al. 1998) was tested as a barrier crop in the other intercrop treat-
ment. Plant spacing within the row was 10 cm for bean, 15 cm for corn, and 46 cm for
eggplant. Each plot contained 14 rows which were 6.1 m in length with 0.9 m between
rows. Intercropped plots were planted in a 2:4:2:4:2 pattern, with corn or eggplant in
the outermost and central 2 rows, surrounding 2 four-row patches of bean. Each treat-
ment was replicated 5 times and arranged in a randomized complete block design.
Corn was planted on 26 July and fertilized with 0.68 kg 15-0-14 (N-P205-K20) per
row. Corn received 0.3 kg 15-0-14 per row on 9 August. Heavy Spodoptera frugiperda
(J. E. Smith) damage threatened the barrier crop treatment in August. Corn was
treated with 1.74 liter/ha methomyl (Lannate, DuPont Corp., Wilmington, DE) on 9
August and 29 August. Eggplant was transplanted on 22 August when 3 wks old. Egg-
plant received 0.23 kg per row 15-0-14 fertilizer 27 August, and 0.8 kg on 27 Septem-
ber and 10 October. Beans were planted on 15 September and fertilized with 0.37 kg
15-0-14 per row on 23 September and 12 October.
The experimental area was treated with 0.19 liter/ha paraquat (Gramoxone, Zen-
eca) on 26 July. Subsequent weed control was mechanical or by hand. The imidaclo-
prid-treated beans received 52.6 g/ha ai imidacloprid (Provado 1.6 F, Bayer) on 4
October and 12 October. This is the rate recommended on the label for most vegetables
(Bayer, Kansas City, MO). Provado 1.6F was applied with a backpack sprayer. Imida-
cloprid is not registered for use on beans but was included so that yield from inter-
cropping treatments could be compared with yield from chemically-protected beans.


Whole plant examinations were made of 1 or 2 bean plants per plot each week from
22 September through 11 November except for 29 September. Only the underside of
the leaf was examined. The area of each leaf was recorded using a LI-COR portable
leaf area meter (model LI-3000A, LI-COR, Lincoln, NE). Bean treatment comparisons
were made on the basis of whole plant counts. Leaf counts from upper, middle, and

Florida Entomologist 83(2)

June, 2000

lower plant strata were used for comparison with eggplant on 21 October and 4 No-
vember. On 29 September bean and eggplant comparisons were based on the average
of counts taken from one 3.35 cm2 disc from a leaf in the upper and lower strata of two
plants per plot (McAuslane et al. 1995).
Whole plant examinations were made of 1 to 3 eggplants per block each week from
25 August through 8 October. After that time, plants became too large for whole plant
examinations. Whole leaf counts from upper, middle, and lower strata were made of
eggplant on 21 October and 4 November.
Leaves were examined using a stereoscope and fiber-optic light. Total number of B.
argentifolii eggs, nymphs, parasitized nymphs, and red-eyed nymphs (also called 'pu-
pae') was recorded for each leaf. Leaves with nymphs showing symptoms of parasit-
ism were placed in unwaxed cylindrical 0.95 liter cardboard cartons (Fonda Group,
Union, NJ) to allow parasitoids to emerge.
The height of five corn plants per row was measured on 4 October to assess the bar-
rier effect. Beans were harvested from two 1.8-m sections from each plot on 22 No-
vember and fresh weight was recorded.

Research Design and Plot Management, 1997

In 1997 the corn barrier treatment was repeated on a larger scale. Three treat-
ments were compared to evaluate the influence of the barrier crop and the effect of
barrier row orientation to wind direction on adult whitefly movement. Prevailing winds
in August in the area tend to be from the east. The treatments were 1) bean planted
in monoculture (bean alone), 2) alternating rows of bean and corn planted north to
south (barrier), and 3) alternating rows of bean and corn planted east to west (open).
Treatments were arranged in a randomized complete block strip split plot design.
Each treatment was replicated four times. The four blocks were arranged in pairs on
either side of a 12 m-wide path running north to south. Treatment plots were 15.25 m
x 30.5 m, with the shorter side parallel to the central path. This design was used to
allow for a release of whitefly adults from points spaced evenly along the central path.
Corn was planted on 25 March. It was fertilized with 67 kg/ha 15-0-14 (N-P20s-
K20) on 1 April, 26 April, and 14 May. Bean was planted on 1 July and fertilized with
33 kg/ha 15-0-14 (N-P20,-K20) at planting, 10 July, and 20 July Overhead irrigation
was used to supplement rainfall. Plots were weeded mechanically and by hand.

Mass-rearing of B. argentifolii

About 30 senescing broccoli, Brassica olerecea L., plants infested with B. argenti-
folii were removed from an organic farm near Gainesville between 1-6 June. They
were potted and placed with 36 flowering hibiscus, Hibiscus rosa-sinensis L., plants in
a greenhouse at the Department of Entomology and Nematology at the University of
Florida. Hibiscus plants were watered regularly and fertilized with Purcell's Sta-
Green plant food (18-6-12 N-P20,-K20, Purcell Industries, Sylacauga, AL). By early
August, the hibiscus plants were heavily infested with whiteflies.

Trap Preparation

Yellow sticky traps have been used in many instances to monitor and sample
whitefly adults (Ekbom & Xu 1990). In the evening of 7 August 180 plastic yellow 710-
ml cups (Solo Cup Company, Urbana, IL) were coated with an aerosol adhesive (prod-
uct 95010, Tanglefoot Company, Grand Rapids, MI) for use as whitefly traps. The
traps were arranged in 5 rows within each plot at 1.5, 7.6, 14, 20, and 26 m from the

Smith & McSorley: Trap and Barrier Crops for Whitefly 149

edge of the plot bordering the central path. Three traps were placed in each row. One
trap was placed 3.8 m in from either side of the plot, and one was placed 7.6 m within
the plot, at the center of the row.

Dust-and-release Procedure

Byrne et al. (1996) developed a method of dusting whitefly adults with a fluores-
cent pigment in the field and trapping them at a distance as a means to monitor move-
ment. We modified this method to distinguish the released whitefly adults from the
trapped field population.
Before dawn on 8 August the infested hibiscus plants were enclosed in 113.5 liter
plastic leaf litter bags. The nozzle of a technical duster (product 1964, Lesco, Cleve-
land, OH) was forced through the plastic and approximately 8.5-14 g orange fluores-
cent AX-14-N pigment (Fire Orange, Day-Glo Color, Cleveland, OH) was puffed from
the duster into the bag onto the infested plants. The hibiscus plants were transported
to the experimental area enclosed in plastic bags and arranged in 6 clusters of 6 plants
along the central path and between pairs of treatment plots. The plastic bags were re-
moved between 7:30 and 7:50 AM to allow a unified release of dyed whitefly adults.
The traps were removed and replaced at dusk. The second set of traps was removed
at dusk on 9 August. After removal, traps were kept refrigerated until examined.
On 10 August, the hibiscus plants were returned to the greenhouse. Traps were
placed in the plots from 8:00 AM to 5:00 PM on 14 August to determine that whitefly
adults from the first release were no longer measurably present in the area. On 24
August the dust-and-release procedure was repeated. Traps were set out from 8:00 AM
to 8:00 PM on 24 August, and replaced with traps that were recovered at dusk on 25
August. Hibiscus plants were removed after the second set of traps had been retrieved.
Traps were examined using a Spectroline 365 nm black light (model B-14N, Spec-
tronics, Westbury, NY). The number of fluorescing whitefly adults on each trap was re-
corded. The height of 15 corn plants per plot was measured on 27 August to evaluate
the barrier effect.

Statistical Analysis

In 1996, densities of B. argentifolii eggs, nymphs, parasitized nymphs, and red-
eyed nymphs were compared among bean treatments using analysis of variance
(PROC GLM, SAS Institute 1996). Densities of whitefly immatures on bean and egg-
plant in the trap crop test were compared using the same test, as was bean yield. For
the 1997 study, the effect of treatment, block, and trap position on trap count was an-
alyzed using analysis of variance. Orthogonal contrasts were then used to compare
trap counts in the same treatment east and west (upwind and downwind) of the re-
lease point, and to compare trap counts among treatments in blocks west of the re-
lease point. Wind direction data collected at the site were provided by Dr. E. B. Whitty,
Agronomy Department, University of Florida, Gainesville, FL.


Whitefly Densities, 1996

Densities of eggs were highest on bean when sampling began and declined over
subsequent weeks (Table 1). Nymphal densities were highest during weeks 3 and 4.
Observations of parasitized nymphs and red-eyed nymphs were low throughout, al-
though parasitism increased slightly over time.


Date Treatment Egg Nymph Para. nymph' REN2

22 Sept. Bean alone 0.79 -- 0.58 0 0 0
Bean w/ corn 1.04 -- 0.73 0 0 0
Bean w/eggplant 1.27 0.68 0 0 0
8 Oct. Bean alone 0.62 -- 0.40 0.64 -- 0.29 0 0
Bean w/ corn 0.93 -- 0.26 0.86 -- 0.33 0.002 -- 0.004 0
Bean w/eggplant 1.00 -- 0.58 1.31 -- 0.87 0.006 -- 0.01 0.004 -- 0.008
14 Oct. Bean alone 0.40 -- 0.30 0.79 -- 0.30 0.010 -- 0.008 0.010 -- 0.02
Bean w/ corn 0.67 -- 0.49 1.10 -- 0.65 0.010 -- 0.004 0.002 -- 0.004
Bean w/ eggplant 0.60 -- 0.27 0.80 -- 0.25 0.004 -- 0.005 0
21 Oct. Bean alone 0.36 -- 0.20 0.48 -- 0.30 0.006 -- 0.005 0.006 -- 0.005 "
Bean w/ corn 0.39 -- 0.10 0.80 -- 0.51 0.016 -- 0.015 0.008 -- 0.013
Bean w/ eggplant 0.44 -- 0.15 0.61 -- 0.33 0.006 0.008 0.004 0.005
28 Oct. Bean alone 0.41 -- 0.34 0.46 -- 0.23 0.004 -- 0.005 0.012 -- 0.011
Bean w/ corn 0.43 -- 0.16 0.58 -- 0.22 0.020 -- 0.015 0.018 -- 0.016
Bean w/ eggplant 0.46 -- 0.14 0.67 0.26 0.010 0.007 0.016 0.011 0
4 Nov. Bean alone 0.54 -- 0.58 0.51 -- 0.26 0.010 -- 0.01 0.010 -- 0.010
Bean w/ corn 0.22 -- 0.18 0.44 -- 0.15 0.016 -- 0.015 0.016 -- 0.013
Bean w/ eggplant 0.34 -- 0.32 0.41 -- 0.22 0.036 -- 0.027 0.014 -- 0.008
11 Nov. Bean alone 0.26 -- 0.06 0.45 -- 0.35 0.046 -- 0.049 0.006 -- 0.008
Bean w/ corn 0.06 -- 0.04 0.31 -- 0.20 0.052 -- 0.043 0.008 -- 0.008
Bean w/ eggplant 0.11 -- 0.16 0.33 -- 0.24 0.024 -- 0.018 0.002 -- 0.004 '

'Parasitized nymphs.
Red-eyed nymphs.

Smith & McSorley: Trap and Barrier Crops for Whitefly 151

There were no differences (p > 0.10) in egg density among treatments during the
first six weeks of sampling. Egg densities on bean alone were higher (p < 0.05) than
on bean intercropped with corn or eggplant during weeks 7 and 8. No differences (p >
0.10) in nymphal densities occurred among treatments. Densities of red-eyed nymphs
were higher (p < 0.05) on bean alone than on the corn and eggplant treatments during
week 4. During week 7, parasitism was more than twice as high in the eggplant treat-
ment as in the other two treatments.
Whitefly adults were observed on eggplant the day following transplanting on 22
August. Egg densities on eggplant foliage ranged from 0.66 + 0.46/cm2 on 25 August
to 3.53 + 0.72/cm2 on 16 September, and declined over the following weeks. Nymphal
densities on eggplant foliage were 1.31 + 1.60/cm2 on 1 September and peaked at 2.39
+ 0.33 on 16 September, declining on subsequent sampling dates. When bean plants
were emerging, eggplants were quite large; they had an average of 7.0 + 1.3 branches,
a mean height of 17.33 + 0.28 cm, and mean leaf area of 485 + 156 cm2 (n = 5).

Bean vs. Eggplant

On all sampling dates after the first week, egg densities were higher (p < 0.05) on
bean than on eggplant (Table 2). During the week that nymphs were first observed on
bean, densities were lower (p < 0.05) on bean than on eggplant. During subsequent
sampling dates, nymphal densities were either higher (p < 0.05) on bean or not statis-
tically different. Observations of parasitized and red-eyed nymphs were either higher
on eggplant than on bean or not statistically different on the two hosts.

Parasitoid Species

All parasitoids reared from bean and eggplant were hymenopterans from the family
Aphelinidae. Thirty-nine parasitoid individuals were recovered from bean leaves. Thirty-
two of these were Encarsia nigricephala Dozier (82%), 4 were Eretmocerus sp. (10.3%),
and 3 were Encarsia pergandiella Howard (7.7%). Among the 121 parasitoid individuals
reared from eggplant leaves, 51 were E. pergandiella (42.1%), 48 were E. nigricephala
(39.7%), 13 were Eretmocerus sp. (10.7%), 6 were E. transuena (Timberlake) (5%), and 3
were Encarsia sp. (2.5%). The greater parasitism and variety of parasitoid species on egg-
plant may be due to the greater number of weeks that eggplant was in the field.

Bean Yield

There was an average of 37.30 + 5.88 bean plants per 3.6 m of row in all treat-
ments. Bean yield per 3.6 m of row was not different among the three treatments and
the imidacloprid-treated bean plants (imidacloprid: 0.95 kg + 0.71; bean: 0.87 kg +
0.58; corn: 0.47 kg + 0.28; eggplant: 1.14 kg + 0.77).

Eggplant as a Trap Crop

Eggplant did not reduce oviposition on adjacent bean early in the season, and so
did not function as a trap crop. Oviposition was not consistently higher on eggplant
than on bean as reported elsewhere (Tsai & Wang 1996). Eggplant leaves may have
been less suitable for oviposition because they were several weeks older than the bean
leaves. A concurrent test of squash, Cucurbita pepo L., as a trap crop for whiteflies
also produced negative results (Smith 2000).
It is possible that host-finding mechanisms used by whitefly adults prevent them
from being drawn away from one host plant by the presence of another. Bemisia tabaci


Egg Nymph Parasitized nymph Red-eyed nymph

Date Bean Eggplant Bean Eggplant Bean Eggplant Bean Eggplant

22 Sept. 1.66 1.67 2.74 + 1.72 0 1.84 + 1.72* 0 0 0 0
29 Sept. 5.52 + 3.44 1.68 1.72*' 0.88 0.62 2.13 1.78* 0 0 0 0.031 0.104

8 Oct. 0.65 + 0.31 0.24 + 0.41* 1.59 0.83 0.29 0.19* 0.005 0.016 0.035 0.037* 0.005 0.016 0.012 + 0.015

21 Oct. 0.64 + 0.54 0.23 0.22* 0.45 0.35 0.49 0.65 0.009 0.014 0.025 0.038 0.003 0.009 0.046 0.078
4 Nov. 0.26 0.26 0.02 0.03* 0.28 0.19 0.11 0.10* 0.024 0.043 0.069 0.067 0.006 0.012 0.048 + 0.043* c

1*Indicates that numbers on bean and eggplant are significantly different on a given date according to analysis of variance at at = 0.05.




Smith & McSorley: Trap and Barrier Crops for Whitefly 153

apparently does not respond to host-specific visual or olfactory cues (Mound 1962).
Elucidation of the precibarial and cibarial chemosensilla of B. tabaci by Hunter et al.
(1996) indicates that B. tabaci may be able to evaluate plant sap before ingesting it.
It has been demonstrated that Trialeurodes uaporariorum (Westwood) relies on gus-
tatory information to accept or reject a host (van Lenteren & Noldus 1990), and this
may also be true for B. argentifolii. In addition, whitefly adults tend to leave some host
plant species more quickly than others (Costa et al. 1991, Verschoor-van der Poel
1978). The observed differences in host-specific oviposition density by B. argentifolii
may be due in part to length of tenure on the plant rather than to some preference ex-
pressed in the host-finding stage (Bernays 1999).
Many trap crop studies have not resulted in consistent reductions of whitefly den-
sities on the main crop (Ellsworth et al. 1994, McAuslane et al. 1995, Perring et al. 1995,
Puri et al. 1996, Schuster et al. 1996). However, Al-Musa (1982) and Schuster et al.
(1996) reported a reduction in virus incidence on tomato using cucumber, Cucumis sa-
tivus L., and squash, respectively, as trap crops. Power (1990) suggests that crop com-
binations which cause virus vectors to probe for briefer periods may reduce the
incidence of persistent viruses such as geminiviruses. Bernays (1999) demonstrated
that B. tabaci tends to move more often and spend less time on certain plants when
they were grown in combination than when they were grown in pure stands. The crop
combinations and densities employed by Al-Musa (1982) and Schuster et al. (1996)
may have led to reduced probing by the vector, and so reduced incidence of virus.

Corn as a Barrier Crop

The corn did not grow well in 1996 due to insufficient fertilizer. It attained a mean
height of 1.18 m + 0.34 (n = 150) and a density of 27 + 7 plants per 6.1m row (n = 30).
We re-evaluated the barrier effect in 1997 with larger, properly fertilized plots. Egg-
plant did not appear to be a promising trap crop, and so was not included in the field
experiment the following year.

Release of Adult Whiteflies, 1997

Average corn height was 2.45 + 1.97 m when whitefly releases were made. The ef-
fect of treatment on trap count was not significant (p > 0.10) on any of the four collec-
tion dates. Wind direction was from the east or northeast during the 4 days that
collections were made (Table 3). Trap counts in plots to the west of the release point
were significantly higher than trap counts in plots to the east of the release point for
each treatment on each collection date (Table 3). When treatments were compared on
the basis of downwind plots only, counts were lower (p < 0.05) in the barrier treatment
than the monocropped bean treatment on 9 August and 25 August. Trap counts were
lower in the downwind barrier plots than in the downwind 'open' plots on 24 August
(p < 0.05) and 25 August (p < 0.1) (Table 3).
Wind direction appeared to be the primary factor determining where whitefly
adults were trapped. This is consistent with observations that whitefly adults move
passively with wind currents as 'aerial plankton' (Byrne & Bellows 1991). Among
downwind plots, the barrier treatment tended to have the lowest counts, indicating
that the arrangement of corn rows perpendicular to the prevailing wind direction did
have some effect on the movement of adults within the plot. However the overall trap
counts in this study were low. The contribution made by corn barriers to reducing
whiteflies may depend on the density of the whitefly population. Crop barriers such as
corn may be more effective when used with other control measures. Short of employ-


Bean alone Corn: barrier to wind Corn: open to wind

Date Row' Downwind Upwind Downwind Upwind Downwind Upwind

Release 12

8 Aug. 1 1.67 + 2.25 0.33 + 0.52 2.33 + 1.03 0.33 + 0.52 2.33 + 1.21 0.50 + 0.84
2 1.33 1.97 0 1.00 + 0.63 0 0.33 + 0.52 0.17 + 0.41
3 0.67 + 0.52 0 1.33 + 1.51 0 0.17 + 0.41 0.17 + 0.41
4 0.50 + 0.55 0.33 + 0.33 + 0.52 0 0.67 + 0.82 0
5 0.33 + 0.52 0 0.17 + 0.41 0.16 + 0.41 0.67 + 1.03 0
E' 0.90 + 1.40 0.13 + 0.35*3 1.03 + 1.16 0.10 + 0.31* 0.83 + 1.12 0.17 + 0.46*

9 Aug. 1 2.00 + 1.79 0.17 + 0.41 1.17 + 0.75 0 1.83 + 1.47 0.33 + 0.52
2 1.67 + 0.82 0.33 + 0.52 1.00 + 0.89 0.17 + 0.41 1.17 + 1.17 0
3 1.50 + 1.22 0 0.83 + 0.98 0 0.67 + 1.21 0
4 0.50 + 0.55 0 0.50 + 0.84 0 1.00 + 0.89 0
5 0.50 + 0.84 0.17 + 0.41 0.67 + 0.82 0 0.50 + 0.84 0
E' 1.23 + 1.22a4 0.13 + 0.35* 0.83 + 0.83b 0.03 + 0.18* 1.03 + 1.16ab 0.07 + 0.25*

Release 22

24 Aug. 1 3.00 + 2.00 0.33 + 0.52 2.83 + 3.25 0 3.50 + 2.17 0.17 + 0.41
2 1.67 + 1.21 0.17 + 0.41 1.50 + 1.05 0.17 + 0.41 2.50 + 1.05 0.17 + 0.41
3 0.83 + 0.75 0.17 + 0.41 0.50 + 0.84 0 1.83 + 1.33 0
4 0.50 + 0.55 0.17 + 0.41 0.17 + 0.41 0 1.17 + 0.41 0

Row refers to trap location (1 = nearest, 5 farthest from release point; see text). (x = mean across all 5 row locations.)
Wind direction on release dates: 8 Aug.: 75; 9 Aug.: 97; 24 Aug.: 61; 25 Aug.: 55.
*Indicates mean trap counts in the same treatment upwind and downwind of the release point are significantly different at p < 0.05 according to F-test for contrasts.
Different letters indicate that mean trap counts among treatments downwind of release point are significantly different at p < 0.05 according to F-test for contrasts.


Bean alone Corn: barrier to wind Corn: open to wind

Date Row' Downwind Upwind Downwind Upwind Downwind Upwind

5 0.33 + 0.52 0.17 + 0.41 0 0 1.33 1.21 0
1' 1.27 + 1.46b 0.20 + 0.41* 1.00 1.82b 0.03 0.18* 2.10 1.52a 0.07 0.25*

25 Aug. 1 3.33 + 1.97 0.17 + 0.41 0.33 0.52 0 1.17 + 1.17 0
2 1.00 + 1.10 0 1.00 + 1.10 0 1.00 + 0.89 0
3 1.17 + 0.98 0 0.17 + 0.41 0 0.50 + 0.55 0
4 0.67 + 0.82 0 0.33 + 0.82 0 0.83 + 1.17 0
5 0.33 + 0.52 0 0.17 + 0.41 0 1.00 + 0.89 0
K' 1.30 1.53a 0.03 + 0.18* 0.40 0.72b 0 0.90 + 0.92 0*

Row refers to trap location (1 = nearest, 5 farthest from release point; see text). (x = mean across all 5 row locations.)
Wind direction on release dates: 8 Aug.: 75; 9 Aug.: 97; 24 Aug.: 61; 25 Aug.: 55.
*Indicates mean trap counts in the same treatment upwind and downwind of the release point are significantly different at p < 0.05 according to F-test for contrasts.
Different letters indicate that mean trap counts among treatments downwind of release point are significantly different at p < 0.05 according to F-test for contrasts.

Florida Entomologist 83(2)

June, 2000

ing manufactured barriers such as floating row covers or fine mesh screens, whitefly
adults probably cannot be excluded from a cropped area (Norman et al. 1993).
Trap position had a significant effect on trap count (ANOVA, d.f. = 4; 8 August, F =
4.67, p < 0.05; 9 August, F = 4.65, p < 0.05; 24 August, F = 2.99, p < 0.1; 25 August,
F = 2.86, p < 0.1). The number of whiteflies caught decreased as trap distance from the
release point increased. The interaction of treatment and trap position interaction
was not significant, suggesting that this decline was not different among treatments.
Data derived from attractive traps may be ambiguous. A gravid or hungry whitefly
adult which is surrounded by non-hosts, such as corn, may be more sensitive to a dis-
tant patch of bright yellow than an adult in similar condition surrounded by accept-
able hosts, such as bean. It is conceivable that the whitefly adults in the corn
treatments spent more time searching and so were drawn from a greater area than
the whitefly adults trapped in the monocropped bean treatments. It is possible that
fewer whitefly adults entered the corn treatments than the monocropped bean, but
that a higher proportion of those entering the corn treatments were trapped. How-
ever, these considerations do not alter the overall impression that where air currents
can enter, whitefly adults can follow.


The authors would like to thank to R. Wilcox, without whom this research would
not have been possible. We are grateful to D. C. Beitrusten, D. W. Dickson, R. N. Gal-
laher, and C. W. Scherer for field assistance, and to J. L. Stimac for sampling advice.
G. A. Evans kindly identified whitefly parasitoids, and J. M. Harrison gave invaluable
statistical guidance. We thank E. B. Whitty for wind direction data. Thanks to R. N.
Gallaher and H. J. McAuslane for improving earlier versions of this manuscript. This
research was made possible by support from the Ruth Freeman and Anselm Fisher
Foundation. Florida Agricultural Experiment Station Journal Series No. R-06937. No
endorsements or registrations are implied herein.


AL-MUSA, A. 1982. Incidence, economic importance, and control of tomato yellow leaf
curl in Jordan. Plant Dis. 66: 561-563.
BERNAYS, E. A. 1999. When host choice is a problem for a generalist herbivore: exper-
iments with the whitefly, Bemisia tabaci. Ecolog. Entomol. 24: 260-267.
BROWN, J. K., D. R. FROHLICH, AND R. C. ROSELL. 1995. The sweetpotato or silverleaf
whiteflies: biotypes of Bemisia tabaci or a species complex? Ann. Rev. Entomol.
40: 511-534.
BYRNE, D. N., T. S. BELLOWS, JR., AND M. P. PARRELLA. 1990. Whiteflies in agricul-
tural systems, pp. 227-261 in D. Gerling (ed.), Whiteflies: their bionomics, pest
status and management. Intercept Ltd. Andover, Hants, U.K. 348 pp.
BYRNE, D. N., AND T. S. BELLOWS, JR. 1991. Whitefly biology. Ann. Rev. Entomol. 36:
BYRNE, D. N., R. J. RATHMAN, T. V. ORUM, AND J. C. PALUMBO. 1996. Localized mi-
gration and dispersal by the sweet potato whitefly, Bemisia tabaci. Oecologia
105: 320-328.
COSTA, H. S., J. K. BROWN, AND D. N. BYRNE. 1991. Host plant selection by the white-
fly, Bemisia tabaci, under greenhouse conditions. J. Appl. Ent. 112: 146-152.
DENHOLM, I., F. J. BYRNE, M. CAHILL, AND A. L. DEVONSHIRE. 1996. Progress with
documenting and combating insecticide resistance in Bemisia, pp. 507-603 in
D. Gerling and R. T. Mayer (eds.), Bemisia 1995: taxonomy, biology, damage,
control, and management. Intercept Ltd. Andover. Hants, U.K. 702 pp.

Smith & McSorley: Trap and Barrier Crops for Whitefly 157

EKBOM, B. S., AND R. Xu. 1990. Sampling and spatial patterns of whiteflies, pp. 107-
121 in D. Gerling (ed.), Whiteflies: their bionomics, pest status and manage-
ment. Intercept Ltd. Andover, Hants, U.K. 348 pp.
1994. Progress on the use of trap crops for whitefly suppression, p. 160 in T. J.
Henneberry, N. C. Toscano, R. M. Faust, and J. R. Coppedge, (eds.), Silverleaf
whitefly: 1994 supplement to the five-year national research and action plan.
ARS-125, United States Department of Agriculture, Washington, D.C.
FARGETTE, D., AND C. FAUQUET. 1988. A preliminary study on the influence of inter-
cropping maize and cassava on the spread of African cassava mosaic virus by
whiteflies. Aspects Appl. Biol. 17: 195-202.
FAUST, R. M. 1992. Conference report and five-year national research and action plan
for development of management and control methodology for the sweetpotato
whitefly, p. 58. ARS-107, United States Department of Agriculture, Washing-
ton, D.C.
IST and Howard IIST subtropical corn hybrids. Agronomy research report AY-
98-02. Agronomy Department, Institute of Food and Agricultural Sciences,
University of Florida.
GOLD, C. S., M. A. ALTIERI, AND A. C. BELLOTTI. 1990. Direct and residual effects of
short duration intercrops on the cassava whitefliesAleurotrachelus socialis and
Trialeurodes variabilis in Colombia. Agric. Ecosystems Environ. 32: 57-68.
HIEBERT, E., A. M. ABOUZID, AND J. E. POLSTON. 1996. Whitefly-transmitted gemini-
viruses, pp. 277-288 in D. Gerling and R. T. Mayer (eds.), Bemisia 1995: taxon-
omy, biology, damage, control, and management. Intercept Ltd. Andover,
Hants, U.K. 703 pp.
cibarial and cibarial chemosensilla in the whitefly, Bemisia tabaci (Gennadius)
(Homoptera: Aleyrodidae). Internat. J. Insect Morphol. Embryol. 25: 295-304.
McAUSLANE, H. J., F. A. JOHNSON, D. L. COLVIN, AND B. SOJACK. 1995. Influence of
foliar pubescence on abundance and parasitism of Bemisia argentifolii on soy-
bean and peanut. Environ. Entomol. 24: 1135-1143.
MORALES, J. R., D. E. DARDON, AND V. E. SALGUERO. 1993. Parcela MIP de validaci6n
y transferencia en tomate, p. 130 in Manejo integrado de plagas en tomate; fase
1: 1991-1992. Ministerio de Agricultura, Ganaderia y Alimentacion, Guate-
mala City, Guatemala.
MOUND, L. A. 1962. Studies on the olfaction and colour sensitivity of Bemisia tabaci.
Entomol. Experiment. Appl. 5: 99-104.
1993. Management of silverleaf whitefly: a comprehensive manual on the biol-
ogy, economic impact and control tactics. United States Department of Agricul-
ture, Washington, D.C.
PERRING, T. M., K. S. MAYBERRY, AND E. T. NATWICK. 1995. Silverleaf whitefly man-
agement in cauliflower using a trap crop, p. 151 in T. J. Henneberry, N.C.
Toscano, R. M. Faust, and J. R. Coppedge (eds.), Silverleaf whitefly: 1995
(USDA), Supplement to the five-year national research and action plan. ARS
POWER, A. G. 1990. Cropping systems, insect movement, and the spread of insect-
transmitted diseases in crops, pp. 47-69 in S. R. Gliessman (ed.), Agroecology:
researching the ecological basis for sustainable agriculture, vol. 78. Springer-
Verlag, New York.
R. KAWTHEKAR, G. D. BUTLER, AND T. J. HENNEBERY. 1996. Wild brinjal
Solanum khasianum Clarke as a potential trap crop management tool for Be-
misia in cotton, pp. 237-240 in D. Gerling and R. T. Mayer (eds.), Bemisia 1995:
taxonomy, biology, damage, control and management. Intercept, Ltd. Andover,
Hants, U.K. 703 pp.

Florida Entomologist 83(2)

June, 2000

RATAUL, H. S., C. K. GILL, AND S. BRAR. 1989. Use of barrier crop and some cultural
measures in the management of yellow mosaic virus on soybean. J. Res. Punjab
Ag. Univ. 26: 227-230.
SALGUERO, V. E. 1993. Manejo de mosca blanca y acolochamiento en tomate. Ministe-
rio de Agricultura Ganaderia y Alimentacion, Guatemala City, Guatemala.
SAS INSTITUTE INC. 1996. SAS/STAT@ Software: changes and enhancements through
release 6.11, SAS Institute. Cary, NC.
SCHUSTER, D. J., P. A. STANSLY, D. E. DEAN, AND J. E. POLSTON. 1996. Potential of
companion plantings for managing silverleaf whitefly and tomato mottle gem-
inivirus on tomato, pp. 168 in T. J. Henneberry, N.C. Toscano, R. M. Faust, and
J. R. Coppedge (eds.), Silverleaf whitefly: 1996 supplement to the five-year na-
tional research and action plan. ARS 1996-01, United States Department of Ag-
riculture. Washington, D.C.
SHAPIRO, J. P. 1996. Insect-plant interactions and expression of disorders induced by
the silverleaf whitefly, Bemisia argentifolii, pp. 167-177 in D. Gerling and R. T.
Mayer (eds.), Bemisia 1995: taxonomy, biology, damage, control, and manage-
ment. Intercept Ltd. Andover, Hants, U.K.
SHARMA, S. R., AND A. VARMA. 1984. Effect of cultural practises on virus infection in
cowpea. Zeitschrift fuir Acker- und Pflanzenbau 153: 23-31.
SMITH, H. A., R. L. Koening, H. J. McAuslane, and R. McSorley. 2000. Effect of silver
reflective mulch and a summer squash (Cucurbita pepo L.) trap crop on densi-
tites of immature Bemisia argentifolii (Homoptera: Aleyrodidae) on organic
bean (Phaseolus vulgaris L.) Journal of Economic Entomology 93: (in press).
TSAI, J. H., AND K. WANG. 1996. Development and reproduction ofBemisia argentifolii
on five host plants. Environ. Entomol. 25: 810-816.
VANDERMEER, J. 1989. The Ecology of Intercropping. Cambridge University Press.
Cambridge, UK.
VAN LENTEREN, J. C. AND L. P. J. J. NOLDUS. 1990. Whitefly-plant relationships: be-
havioral and ecological aspects, pp. 47-87 in D. Gerling (ed.), Whiteflies: their
bionomics, pest status and management. Intercept Ltd. Andover, Hants, U.K.
348 pp.
VERSCHOOR-VAN DER POEL, P. J. G. 1978. Host-plant selection by the greenhouse
whitefly, Trialeurodes vaporariorum (Westwood) (Homoptera: Aleyrodidae) (in
Dutch). M.Sc. Thesis, University of Leiden, 39 pp.

Davidson et al.: Bacteria in Bemisia argentifolii 159


1Department of Biology, Arizona State University, Tempe, AZ 85287-1501

2Department of Biology, University of St. Thomas, Houston, TX 77006

'USDA-ARS, Western Cotton Research Laboratory, Phoenix, AZ 85040


Several different types of bacteria were cultured from surface-sterilized Bemisia
argentifolii Bellows, Perring, Gill and Hedrick 1994 (Homoptera: Aleyrodidae) adults
and nymphs, including Bacillus spp., Gram-variable pleomorphic rods and Gram-pos-
itive cocci. Two of the isolates were capable of being ingested by adults and passed into
the honeydew. One of these, Enterobacter cloacae, was found within the gut cells of
adult whiteflies and was mildly pathogenic. This isolate represents the first bacte-
rium with potential as a pathogen of whiteflies. Bacteria which were not capable of be-
ing ingested, may have been located in structures which were protected from surface
sterilization, such as the lingula or the female reproductive tract.

Key Words: Bemisia tabaci B-biotype, Enterobacter cloacae, Bacillus sp., symbiotic


Diferentes tipos de bacteria fueron aisladas de adults y ninfas de Bemisia argen-
tifolii Bellows et al., 1994 (Homoptera: Aleyrodidae) esterilizados superficialmente.
Entre las bacteria aisladas se encontr6 Bacillus spp., bacilos pleom6rficos Gram-va-
riables y cocos Gram-positivos. Los adults fueron capaces de ingerir a dos de los or-
ganismos aislados y de transferirlos a la mielecilla que secretan. Uno de 6stos,
Enterobacter cloacae, fu6 encontrado dentro de c61lulas intestinales de mosquitas
blancas adults y fu6 moderadamente patog6nico. Esta bacteria represent el primer
organismo aislado que posee potential patog6nico para el control de la mosquita
blanca. Otras bacteria aisladas no fueron capaces de ser ingeridas, lo cual se atribuy6
a que se ubicaron en estructuras protegidas de la esterilizaci6n superficial, como la li-
gula o el tracto reproductive femenino.

The silverleaf i.,i.. i1. D. -- .. P .. - ... (= B. tabaci B biotype) (Bellows, Perring,
Gill and Hedrick, 1994), is one of the most damaging insects to agriculture in the south-
ern United States and in warmer regions of many other countries. B. argentifolii feeds
on a wide variety of plant species, including cotton, melons, brassicas, tomato, peppers,
and ornamentals such as poinsettia (Cock 1986, De Barro 1995). This insect causes eco-
nomic damage in four ways: by removing photosynthetic products during phloem feed-
ing; by contaminating cotton fiber and ornamental plants with sticky excreted
honeydew; by inducing physiological responses in the host such as squash silverleaf and
tomato irregular ripening; and by vectoring viruses (Byrne & Bellows 1991).
Bemisia argentifolii possesses pleomorphic and coccoid obligate intracellular bac-
terial endosymbionts, housed in mycetomes, which are transmitted from the female to

Florida Entomologist 83(2)

June, 2000

her eggs (Costa et al. 1993a). Growth and development of the nymph and induction of
the squash silverleaf disorder, for which the species is named, are retarded if the sym-
biotic bacteria are reduced or eliminated by feeding antibiotics to the adult female be-
fore oviposition or if fed to the nymph via the leaf (Costa et al. 1993b, 1997). These
endosymbionts have not reportedly been cultured on laboratory media.
The objectives of this study were to investigate whether other bacteria are present
in whiteflies, and modes of entry of these bacteria into the whiteflies. We have isolated
a variety of bacteria from surface-sterilized B. argentifolii adults and nymphs, some
of which have previously been implicated in the production of medium-length oli-
gosaccharides in the honeydew (Davidson et al. 1994). Two of the isolated bacteria
were shown to be ingested by the whitefly and one of these was mildly pathogenic.



Whitefly adults, nymphs and eggs were collected from cotton, cabbage, cucumber,
squash, lantana, pepper or melon plants and surface sterilized with ethanol and
household chlorine bleach as described previously (Davidson et al. 1994). Because of
the small size of these insects, samples were processed in groups of ca. 50-200. Adults,
nymphs and eggs were processed separately. Surface sterilized insects were inocu-
lated into liquid nutrient broth-yeast extract-salts medium (Myers and Yousten 1978)
or brain heart infusion broth, either whole or homogenized in sterile 0.9% saline. Ho-
mogenized insects were also plated directly on microbiological media including nutri-
ent agar, brain heart infusion agar, tryptose agar, Luria agar, purple agar, and
chocolate agar (Sigma, St. Louis, MO) and incubated aerobically at 25C.

Bacterial Cultures

When bacterial growth was observed in liquid or agar bacterial media, cultures
were streaked for purity on agar media of the same type on which positive growth had
been observed. Cultures were preserved by freezing at -70C in 20% glycerol in liquid
medium which best supported growth of each culture.
Bacteria were identified according to microscopic appearance, Gram's stain,
anaerobic growth, catalase production, colony morphology, and reactions on API 20E
and API CH identification strips (bioMerieux Vitek, Hazelwood, MO). Two isolates,
designated WFA73 and WFN29, were also identified by gas chromatography (GC)
fatty acid profiles using the MIDI identification system, by Dr. Joel Siegel, Illinois
Natural History Survey.

Whitefly Ingestion of Bacteria

Eight different bacterial strains, representative of the morphological and Gram-
stain groups isolated from whiteflies (Table 3), were suspended in 30% sucrose and
green food coloring at ca. 10' cells/ml, and fed to adult whiteflies through parafilm sa-
chets. Control whiteflies were fed on sucrose alone. After 48 hr, mortality was re-
corded, and all insects were surface sterilized, homogenized, and plated to
appropriate agar medium. Sterile petri dishes were used to collect honeydew from
control whiteflies and those fed bacteria. The dishes were rinsed with sterile saline
and the saline plated to agar medium. Bacterial colonies recovered were compared
microscopically and in colony morphology to those originally fed to the whiteflies. Bio-

Davidson et al.: Bacteria in Bemisia argentifolii 161

assays of strains WFA73 and WFN29 were repeated twice and % mortality reported
as the mean of two replicates; at least 100 insects were included in each treatment.

Electron Microscopy

Adult whiteflies were fed strain WFA73, which had been found to be ingested and
mildly pathogenic in experiments described above, in green sucrose solution at ca.
10' cells/ml in parafilm sachets. Control insects were fed on green sucrose only. After
24 hr, insects with green digestive tracts were prepared for electron microscopy. In-
sects were gently pierced in the thorax region while immersed in fixative, which con-
sisted of 4% glutaraldehyde in 0.05M cacodylate buffer. Insects were fixed for 4 hr,
postfixed in 0.5% osmium tetroxide in 0.05M cacodylate buffer for 1-2 hr, dehydrated
in ethanol, and embedded in Spurr's resin or LR White resin. Thin sections from 2 con-
trol and 4 experimental whiteflies were observed by transmission electron microscopy
using a Philips EM 200 TEM (Eindhoven, Netherlands).


Bacteria from Whiteflies

Bacteria were cultured from surface-sterilized whitefly adults and nymphs collected
from all host plant species and from all groups of whiteflies collected at 30 different
times over a period of five years. A greater variety of bacteria was recovered from adult
whiteflies than from nymphs, however none were cultured from surface-sterilized eggs.
A total of 80 isolates were preserved from adults and 29 from nymphs; representative
examples are shown in Tables 1 and 2. There was no relationship between the host
plant species and the type of bacteria isolated from the whiteflies (Tables 1 and 2).
Four major types of bacteria were cultured from B. argentifolii using standard mi-
crobiological media. These included: 1. Gram-positive sporeforming aerobic rods, Ba-
cillus spp.; 2. Gram-positive cocci; 3. Gram-variable short pleomorphic rods producing
very short rods to cocci in older cultures; 4. Gram-variable long, thin, highly pleomor-
phic rods forming cocci in older cultures (Tables 1 and 2).
There was a strong correlation between the insect life stage and the types of bac-
teria isolated. Most isolates of Gram-positive or Gram-variable rod-shaped bacteria
were obtained from adult whiteflies, whereas most isolates of cocci were produced
from nymphs (Tables 1 and 2).

Bacterial Identification

Sporeforming aerobic bacteria were identified as Bacillus licheniformis (Weig-
man), B. megaterium deBary, B. amyloliquefasciens Fukumoto, and B. subtilis
(Ehrenberg), based upon microscopic examination and the results of API diagnostic
tests. The B. subtilis, B. licheniformis and B. megaterium isolates were found to pro-
duce medium-length sugars from sucrose in an earlier study (Davidson et al. 1994).
Spherical cells, forming diads and tetrads and occasionally chains, were frequently
isolated from B. argentifolii nymphs. Isolates which were Gram-positive, catalase pos-
itive, facultatively anaerobic, and capable of growing in the presence of 10% NaC1,
were identified as Staphylococcus spp., close to S. aureus Rosenbach, S. sciuri Kloos,
Schleifer & Smith and S. epidermidis (Winslow & Winslow). Larger gram-positive
cocci forming diads and tetrads resembled Sporosarcina spp. (Orla-Jensen) although
spores were not observed in these cultures (Tables 1 and 2).

Florida Entomologist 83(2)

June, 2000


Designation Description Host Identification

WFA9, 10, 11, 12
WFA12, 14, 35,
36, 40, 45, 46


WFA56, 59


WFA69, 70
WFA 73

sporeforming rod
spheres in diads
sporeforming rod
sporeforming rod
sporeforming rod
sporeforming rod

sporeforming rod
short pleomorphic
rod, yellow colony
cocci, diads
small pleomorphic
rods, clear colonies
diplococci, tetrads
very short rods,
thick at one end
short rods, yellow
short pointed rods
short curved rods
sporeforming rod
short pleomorphic rods
short rod with inclusions
short pleomorphic rods


Bacillus amyloliquefasciens
B. licheniformis
B. megaterium
B. licheniformis
Bacillus spp.

cotton Bacillus spp.

cotton ND
cotton ND

cotton Chryseomonas luteola
cotton ND

cotton Acinetobacter lwoffsii


Citrobacter sp.
Flavomonas oryzihabitans
Acinetobacter baumanii?
Bacillus sp.
Enterobacter cloacae*
Flavomonas oryzihabitans

On the basis of API tests, short, Gram-negative or Gram-variable pleomorphic
rods were identified as Enterobacter cloacae (Jordan), Flavimonas oryzihabitans
Holmes et al., Citrobacter sp. Workman and Gillen, Cellulomonas sp. Bergey et al.,
Chryseomonas luteola Holmes et al., Acinetobacter lwoffsii Brisou & Prevot orA. bau-
manni (Deacon). Some strains produced no reaction on API tests and therefore re-
main unidentified (Tables 1 and 2). Two strains which were found to be ingested by
adult whiteflies (below) were further identified by MIDI-GC.

Whitefly Ingestion of Bacteria

To determine whether bacteria isolated from whiteflies could have entered the in-
sects during feeding, we fed eight different bacterial strains to adults (Table 3). Bac-
teria strains designated WFA73 and WFN29 were recovered in large quantity from
surface-sterilized adults and their honeydew following feeding on these bacteria. Bac-
terial colonies, morphologically and microscopically identical to WFA73 or WFN29,
were not found in homogenates or honeydew of control insects. WFA73 is a short ple-
omorphic Gram-variable rod, identified by GC analysis of fatty acids and API analysis

Davidson et al.: Bacteria in Bemisia argentifolii


Designation Description Host Identification

WFN1 spheres, diads cotton ND
WFN3 sporeforming rod cotton Bacillus sp.
WFN4 large diplococcus cotton Sporosarcina sp.?
WFN6 sporeforming rod cabbage Bacillus sp.
WFN7 coccus cabbage Staphylococcus epidermidis
WFN10, 11, 13 cocci cabbage Staphylococcus epidermidis
WFN12 cocci cabbage Staphylococcus aureus
WFN14 large cocci cabbage Sporosarcina sp.?
WFN15 short pleomorphic rods cabbage ND
WFN17A very thin rods cabbage Agromonas sp.
WFN28 sporeforming rod cabbage Bacillus licheniformis
WFN29 pleomorphic rods cabbage Cellulomonas turbata*
forming cocci

as E. cloacae (0.70 agreement). Isolate WFN29, which formed bent rods in young cul-
tures and cocci in older cultures, was identified by GC and API analysis as Cellulomo-
nas (Oerskovia) turbata (Erikson) (0.78 agreement). Bacillus spp., Gram-positive
cocci and the other Gram-negative or Gram-variable isolates were not recovered from
adults fed these isolates (Table 3).


Strain from insect
designation Morphology Identification or honeydew

WFA5 Gram-positive Bacillus Bacillus megaterium no
WFA11 Gram-positive Bacillus Bacillus subtilis no
WFN12 Gram-positive Coccus Staphylococcus aureus no
WFN7 Gram-positive Coccus Staphylococcus epidermidis no
WFA73 Short pleomorphic
Gram-variable rod Enterobacter cloacae yes
WFN29 Pleomorphic
Gram-negative rod Cellulomonas turbata yes
WFA69 Short curved or branched
rod Acinetobacter baumanii? no
WFA74 Short rod with inclusions No API reaction, Gram-
variable no

Florida Entomologist 83(2)

June, 2000

Bioassay Results

Strain WFA73 (E. cloacae) produced an average 34% mortality (s.d. = 1.41) of adult
whiteflies at ca. 10'bacterial cells/ml after 24 hr in two experiments (control mortality
= 4.0%; s.d. = 2.83). Mortality increased to 75% (s.d. = 0) after 48 hr in adult whiteflies
fed WFA73 (control mortality = 9.5%; s.d.= 6.36). Bacteria identical to E. cloacae were
also recovered from honeydew of adults fed these bacteria but not from the honeydew
of control adults fed only on sucrose. Strain WFN29 (C. turbata) produced only 4.5%
mortality (s.d. = 2.12) of adult B. argentifolii at 10' cells/ml after 48 hr, similar to con-
trol mortality (4.0%; s.d. = 1.41), but was recovered from homogenized adults fed this
strain and from their honeydew.

Electron Microscopy

Bacteria were not observed in the digestive system of control whiteflies in these
studies (not shown).
In adults fed strain WFA73 (E. cloacae), large numbers of bacteria were seen
throughout the digestive tract (Fig. 1). The lumen of the entire midgut was filled with
rod-shaped bacteria (WFA73). Bacteria adhered to the apical portion of the descend-

Fig. 1. After feeding on bacteria, both the ascending and descending portions of the
midgut of B. argentifolii contained high numbers of WFA73 (E. cloacae) bacteria (B)
in the lumina (L). Thorax appears to the left and abdomen to the right. Note the pres-
ence of engulfed bacteria (EB) in the apical portion of the epithelial cells. Nu, nucleus
of midgut cell; FB, fat body; E, egg. Bar = 10 pm.

Davidson et al.: Bacteria in Bemisia argentifolii 165

ing and ascending midgut epithelial cells and were also found within the epithelial
cell cytoplasm in membrane bound vesicles (Figs. 2 and 3). Cells in the descending
midgut appeared to be taking up the bacteria by phagocytosis as evidenced by bacte-
ria in various stages of engulfment (Figs. 3 and 4). Descending midgut cells of infected
insects also exhibited poorly developed microvilli, numerous spherical vesicles, each
with a small amount of electron dense material, many large electron dense lysosomal-
like vesicles, and small electron dense residual bodies (Fig. 4). Vacuolation of mito-
chondria was observed both in control and bacteria-fed whiteflies, probably a result of
slow fixative penetration.
We observed bacteria-like organisms in the reproductive tracts of two female
whiteflies (Fig. 5). The bacteria in the female reproductive tracts were readily distin-
guishable from sperm, which were observed in the spermatheca and appeared as elec-
tron-dense structures which lacked notable internal structure, and were aligned in
packets (not shown). The identity of the bacteria found in the female reproductive
tracts is not currently known.


Some of the bacteria cultured from whiteflies may be involved in a mutualistic re-
lationship with the insects, contributing to the digestion and nutrition of the insect,
while obtaining access to the high sugar content of phloem sap in the gut of the insect
and honeydew. As described earlier (Davidson et al. 1994), Bacillus spp. associated
with B. argentifolii may produce long-chain sugars which contribute to the stickiness
of the honeydew of this insect. During a study of digestive tract ultrastructure (Rosell
et al., unpubl.), bacteria were observed in the esophagus of adult Bemisia spp. taken
from a laboratory colony reared on cotton.
Similar gut bacteria have been isolated from other Homoptera. Staphylococcus sci-
uri and S. epidermidis, and Gram-negative rods, close to Pseudomonas fluorescens
Migula, have been isolated from the pea aphid, Acythosiphon pisum (Harris) (Grenier
et al. 1994). The flora in the aphid gut was assumed to have been acquired during
probing on the leaf surface, as is probably the case with most of the bacteria isolated
from whiteflies as well. Srivastata and Rouatt (1963) isolated Sarcina, Micrococcus,
Achromobacter and Flavobacterium from aphids. Bacteria have also been reported in
the hemocoel of aphids and leafhoppers (Grenier et al. 1994, Purcell et al. 1986). At
this time we cannot rule out the possibility that some of the bacteria isolated from B.
argentifolii are also occasional residents in the hemocoel.
We do not currently know the significance of bacteria present in the female repro-
ductive tract. Morphologically, they resemble the short rod-shaped bacteria commonly
isolated from adult whiteflies (Table 1). Following electron microscope observation of
bacteria in the female reproductive tract, two collections of whiteflies were separated
into males and females, surface sterilized and processed separately for bacterial iso-
lation. Bacteria were cultured only from females. Their presence suggests possible
transmission of these bacteria to offspring, however bacteria were not cultured from
surface-sterilized eggs.
The results presented here confirm that B. argentifolii adults and nymphs can in-
gest certain culturable bacteria and contain these bacteria in their digestive tracts. As
these bacteria were ingested through Parafilm, and were recovered from honeydew af-
ter feeding, the bacteria were truly ingested and not simply resident on the external
surface or mouthparts of the insect. Our results are in agreement with those of Zeidan
and Czosnek (1994) who found that B. tabaci could ingest Agrobacterium. The failure
to culture similar bacteria from surface sterilized whitefly eggs, suggests that these

Florida Entomologist 83(2)

June, 2000

Fig. 2. Descending midgut from whitefly fed WFA73 bacteria. Note bacterium (B)
present in the lumen (L) that is being engulfed and bacteria (EB) present in mem-
brane bound vesicles. M, mitochondrion. Bar = 1 pm.
Fig. 3. WFA73 bacterial cell (EB) actively being engulfed at apical surface of de-
scending midgut epithelia in this whitefly fed the bacteria. Extracellular bacteria (B)
are present in the lumen (L). M, mitochondrion. Bar = 1 pm.

Davidson et al.: Bacteria in Bemisia argentifolii

Fig. 4. Descending midgut from whitefly fed WFA73 showing electron lucent
spherical vesicles (Sv), electron dense lysosomal-like vesicles (Lv) and small dense re-
sidual bodies (Rb). Bacteria are present in the lumen (L). Bar = 1 (m.

bacteria were obtained from the leaf surface during probing prior to feeding. These
bacteria are not transovarially transmitted, as the same bacteria were not isolated
from all samples and bacteria were not isolated from homogenized surface-sterilized
eggs. These culturable bacteria are therefore not obligate symbionts.
Rosell et al. (1995) demonstrated ultrastructurally that the adult B. tabaci stylet
food canal is 0.65 pm in diameter. Using fluorescent beads, we have recently shown
that 0.2 pm beads are ingested and pass in honeydew, but 0.5 pm beads do not enter
the insect (Rosell et al., in prep). Therefore in order for bacteria to enter the food ca-
nal, they must be less than 0.5 pm in diameter, or be pleomorphic with some members
of the population smaller than 0.5 pm. The bacteria which successfully entered the
food canal, strains WFA73 and WFN29, fulfilled these requirements. In contrast,
Gram-positive bacteria including Bacillus spp. and Staphylococcus spp., which have
generally larger diameters, entered the food canal poorly or not at all. As several Ba-
cillus spp. and Staphylococcus spp. were isolated from surface-sterilized whiteflies,
these organisms may have been located under the lingula at the posterior of the gut,
where honeydew accumulates prior to excretion, or in the female reproductive tracts
as seen in electron microscopy (Fig. 5).

Florida Entomologist 83(2)

June, 2000

Fig. 5. Bacteria-like organisms (BL) are found in the female reproductive tract
which is convergent with the ovipositor canal (OV). C, cuticle; Mu, muscle; nu, nucleus
of epidermal cell. Posterior of the whitefly is to the right. Bar = 5 Pm.

Isolate WFA73 (E. cloacae) appears to be mildly pathogenic to B. argentifolii
adults, and represents the first bacterial pathogen reported from whiteflies. In speci-
mens fed this bacterium, midgut cells were massively invaded by bacteria, likely lead-
ing to loss of part or all of gut function (Fig. 1). Phagocytosis of bacteria by insect
midgut cells has been observed during the pathogenesis of both American foulbrood
disease of honey bees (Davidson 1973), and milky disease of beetles (Kawanishi et al.
1978, Splittstoesser et al. 1978). Enterobacter cloacae was described as a pathogen of
grasshoppers under its original name, Coccobacillus acridiorum. Enterobacter (Aero-
bacter) aerogenes is a pathogen of lepidoptera in association with Proteus mirabilis
(Tanada and Kaya 1993, Wysoki and Raccah 1980) and occurs in the gut flora of the
New Zealand grass grub (Stucki et al. 1984).
Isolate WFA73 (E. cloacae), which was readily ingested by adults, originated from
whiteflies with enhanced resistance to the insecticide Danitol (Valent, USA) and is ca-
pable of precipitating Danitol in vitro (E. Davidson, L. Williams and D. Alexander, unpubl.
results). Therefore culturable bacteria associated withB. argentifolii may be important to
insecticide degradation in the phylloplane, and perhaps in the insect as well.
The relationship of E. cloacae to B. argentifolii is similar in several aspects to the
bacterium designated BEV in leafhoppers. BEV can be cultured on bacteriological me-

Davidson et al.: Bacteria in Bemisia argentifolii 169

dia, is mildly pathogenic to its normal host, Euscelidius variegatus Kirshbaum, and
penetrates gut cells in a manner ultrastructurally similar to E. cloacae in the whitefly.
However BEV is both transmitted transovarially and acquired from the plant (Purcell
et al. 1984, Purcell and Suslow 1987, Cheung and Purcell 1993).
Clark et al. (1992) examined the endosymbionts of B. argentifolii and B. tabaci
using 16S rDNA analysis, and found the secondary endosymbiont of Bemisia is re-
lated to Enterobacteriaceae. Results presented here confirm that Enterobacteri-
aceae are commonly present in B. argentifolii, and E. cloacae can enter the cells of
the insect, suggesting that such bacteria could have been ancestors of whitefly endo-
symbiotic bacteria, as suggested by Harada et al. (1996) for an endosymbiont of the
pea aphid. Gut bacteria also present contaminating foreign bacterial DNA which
may confuse genetic analysis of endosymbionts, as pointed out by Grenier et al.
(1994) for aphids.
While it is clear that mycetome endosymbionts are critical to the development of
B. argentifolii (e.g. Costa et al. 1993b), the presence of other bacteria must be taken
into consideration in studies of the physiology of this insect. Although WFA73 (E. clo-
acae) is only mildly pathogenic to B. argentifolii, its ability to penetrate whitefly gut
cells suggests that this microorganism could be genetically modified to enhance its ef-
fectiveness as a biological control agent. Transformation of a cotton phyllosphere bac-
terium, B. megaterium, with Bacillus thuringiensis Berliner toxin genes for control of
lepidoptera (Bora et al. 1994) is an example of such manipulation. Finally, genetic
modification of gut symbionts ofRhodnius prolixus to interfere with vectoring of Cha-
gas disease has been reported (Beard et al. 1992, 1993). Similar modification of gut
bacteria in other insects, such as whiteflies, to alter their ability to vector plant vi-
ruses or other characteristics of these insects, is an intriguing possibility (Richards


Research at ASU was funded by a Cooperative Agreement with the USDA-ARS
Western Cotton Laboratory, Phoenix, AZ, and USDA CSREES 97-35316-5139. The
ASU Life Sciences Electron Microscopy Laboratory and the University of Arizona Di-
vision of Biotechnology Electron Microscopy facilities were used in this study, and we
are grateful to William Sharp, Gina Zhang, Leah Kennaga and David Bentley for as-
sistance. Identification of bacteria using fatty acid analysis was performed by Dr. Joel
Siegel, Illinois Natural History Survey, Urbana, IL.


BEARD, C. B., P. W. MASON, S. AKSOY, R. B. TESH, AND F. F. RICHARDS. 1992. Trans-
formation of an insect symbiont and expression of a foreign gene in the Chagas'
disease vector, Rhodnius prolixus. Am. J. Trop. Med. Hyg. 46: 195-200.
BEARD, C. B., S. L. O'NEILL, R. B. TESH, F. F. RICHARDS, AND F. AKSOY, 1993. Modifi-
cation of arthropod vector competence via symbiotic bacteria. Parasitol. Today
9: 179-183.
BELLOWS, T. S. JR., T. M. PERRING, R. J. GILL, AND D. H. HEADRICK. 1994. Description
of a species of Bemisia (Homoptera: Aleyrodidae). Ann. Entomol. Soc. Am. 87:
BORA, R. S., M. G. MURTY, R. SHENBAGARATHAI, AND V. SEKAR. 1994. Introduction of
a lepidopteran-specific insecticidal crystal protein gene of Bacillus thuringien-

Florida Entomologist 83(2)

June, 2000

sis subsp. kurstaki by conjugal transfer into a Bacillus metagerium strain that
persists in the cotton phyllosphere. Appl. Environ. Microbiol. 60: 214-222.
BYRNE, D. N., AND T. S. BELLOWS, JR. 1991. Whitefly biology. Annu. Rev. Entomol. 36:
CHEUNG, W. W. K., AND A. H. PURCELL, 1993. Ultrastructure of the digestive system
of the leafhopper Euscelidius variegatus Kirshbaum with and without congen-
ital bacterial infections. Int. J. Insect Morphol. & Embryol. 22: 49-61.
FUS, L. S. OSBORNE, AND N. A. MORAN. 1992. The Eubacterial endosymbionts
of whiteflies constitute a lineage distinct from the endosymbionts of aphids and
mealybugs. Curr. Microbiol. 25: 119-123.
COCK, M. J. W. 1986. Bemisia tabaci, a literature survey on the cotton whitefly with
an annotated bibliography. FAO/CAB, Ascot, UK.
COSTA, H. S., D. M. WESTCOT, D. E. ULLMAN, AND M. W. JOHNSON. 1993a. Ultrastruc-
ture of the endosymbionts of the whitefly, Bemisia tabaci and Trialeurodes va-
porariorum. Protoplasma 176: 106-115.
COSTA, H. S., D. E. ULLMAN, M. W. JOHNSON, AND B. E. TABASHNIK. 1993b. Antibiotic
oxytetracycline interferes with Bemisia tabaci oviposition, development and
ability to induce squash silverleaf. Ann. Entomol. Soc. Am. 86: 740-748.
COSTA, H. S., T. J. HENNEBERRY, AND N. C. TOSCA. 1997. Effects of antibacterial mater-
ials on Bemisia argentifolii oviposition, growth, survival and sex ratio. J. Econ.
Entomol. 90: 333-339.
DAVIDSON, E. W. 1973. Ultrastructure of American foulbrood disease pathogenesis in
larvae of the worker honey bee, Apis mellifera. J. Invertebr. Pathol. 21: 53-61.
DAVIDSON, E. W., B. J. SEGURA, T. STEELE, AND D. L. HENDRIX. 1994. Microorganisms
influence the composition of honeydew produced by the silverleaf whitefly, Be-
misia argentifolii. J. Insect Physiol. 40: 1069-1076.
DE BARRO, P. J. 1995. Bemisia tabaci biotype B: a review of its biology, distribution
and control. CSIRO Div. Entomol. Tech. Paper No. 33, Canberra, Australia.
GRENIER, A.-M., C. NARDON, AND Y. RAHBE. 1994. Observations on the micro-organ-
isms occurring in the gut of the pea aphidAcyrthosiphon pisum. Entomol. Exp.
Appl. 70: 91-96.
HARADA, H., H. OYAIZU, AND H. ISHIKAWA. 1996. A consideration about the origin of
aphid intracellular symbiont in connection with gut bacterial flora. J. Gen.
Appl. Microbiol. 42: 17-26.
Infection of the European chafer, Amphimallon majalis by Bacillus popilliae:
Ultrastructure. J. Invertebr. Pathol. 31: 91-102.
MYERS, P., AND A. A. YOUSTEN. 1978. Toxic activity of Bacillus sphaericus SSII-1 for
mosquito larvae. Inf. Immun. 19: 1047-1053.
PURCELL, A. H., T. STEINER, F. MEGRAUD, AND J. M. BOVE. 1986. In vitro isolation of
a transovarially transmitted bacterium from the leafhopper Euscelidus varie-
gatus. J. Invertebr. Pathol. 48: 66-73.
PURCELL, A. H., K. G. SUSLOW, AND M. KLEIN. 1984. Transmission via plants of an in-
sect pathogenic bacterium that does not multiply or move in plants. Microb.
Ecol. 27: 19-26.
PURCELL, A. H., AND K. G. 1987. Pathogenicity and effects on transmission of a myco-
plasma-like organism of a transovarially infective bacterium on the leafhopper
Euscelidius variegatus. J. Invertebr. Pathol. 50: 285-290.
RICHARDS, F. F. 1993. An approach to reducing arthropod vector competence. ASM
News, 59: 509-514.
ROSELL, R. C., J. E. LICHTY, AND J. K. BROWN. 1995. Ultrastructure of the mouthparts
of adult sweetpotato whitefly, Bemisia tabaci. Int. J. Insect Morphol. & Em-
bryol. 24: 297-306.
SPLITTSTOESSER, C. M., C. Y. KAWANISHI, AND H. TASHIRO. 1978. Infection of the Eu-
ropean chafer, Amphimallon majalis by Bacillus popilliae: light and electron
microscope observations. J. Invertebr. Pathol. 31: 84-90.

Epler et al.: Redescription of Cricotopus lebetis 171

SRIVASTAVA, P. N., AND J. W. ROUATT. 1963. Bacteria from the alimentary canal of the
pea aphid, Acyrthosiphon pisum. J. Insect Physiol. 9: 435-438.
STUCKI, G., T. A. JACKSON, AND M. J. NOONAN. 1984. Isolation and characterization
of Serratia strains pathogenic for larvae of the New Zealand grass grub, Cost-
elytra zealandica. N.Z. J. Science 27: 255-260.
TANADA, Y., AND H. KAYA. 1993. Insect Pathology. pp. 147-151. Academic Press, NY.
WYSOKI, M., AND B. RACCAH. 1980. A synergistic effect of two pathogenic bacteria
from the Enterobacteriaceae on the geometrid Boarmia selenaria. J. Invertebr.
Pathol. 35: 209-210.
ZEIDAN, M., AND H. CZOSNEK. 1994. Acquisition and transmission of Agrobacterium
by the whitefly Bemisia tabaci. Molec. Plant-Microbe Interactions 7: 792-798.

Epler et al.: Redescription of Cricotopus lebetis


1461 Tiger Hammock Road, Crawfordville, FL 32327

2Entomology and Nematology Department, University of Florida
PO Box 110620, Gainesville, FL 32611-0620

'Fort Lauderdale Research and Education Center, University of Florida
3205 College Avenue, Fort Lauderdale, FL 33314-7799


The adult male and female of Cricotopus lebetis Sublette are redescribed and the
pupa and larva described for the first time. Larvae of C. lebetis mine in the stems of
the submersed aquatic weed hydrilla, Hydrilla uerticillata (L.f. Royle), causing suffi-
cient damage to the apical meristem to preclude further growth. The species is very
similar to C. tricinctus (Meigen) but can be distinguished from that species in the
adult male by the broader, more rounded inferior volsella; in the female by the lower
number of sensilla chaetica on the mid and hind basitarsi; in the pupa by the fusiform
thoracic horn; and in the larva by the simple S I and long setal tufts on abdominal seg-
ments I-VII.

Key Words: Chironomidae, Cricotopus, taxonomy, Hydrilla uerticillata, aquatic weed,


Se redescriben el adulto macho y hembra de Cricotopus lebetis Sublette y se des-
criben por primera vez la larva y pupa de esta especie. Las larvas de C. lebetis minan
los tallos de la hierba acuatica sumergida Hydrilla verticillata (L.f. Royle), causando
suficiente daho al meristemo apical como para impedir su crecimiento. C. lebetis es
muy similar a C. tricinctus (Meigen), pero se distingue de esta especie en que el macho
adulto posee una volsella inferior mas redondeada y ancha, mientras que la hembra

Florida Entomologist 83(2)

June, 2000

posee menos sensulos tipo chaetica en los basitarsi medio e inferior. Asmismo, la pupa
de C. lebetis present un cuerno toracico fusiforme y la larva tiene un S I simple y lar-
gos pinceles de cerdas en los segments abdominales I-VII.

The chironomid genus Cricotopus van der Wulp is common, widespread and speci-
ose. Hirvenoja (1973) revised the Palaearctic species but in the Nearctic the taxonomy
of the genus remains in less than satisfactory condition. Many undescribed species ex-
ist and the conspecificity of some Nearctic taxa with species originally described from
the Palaearctic is uncertain.
One such species is Cricotopus lebetis Sublette, a member of the sylvestris group of
the subgenus C. (Isocladius). Interest in the taxonomy of C. lebetis has been recently
stimulated by the discovery of the larvae of this species feeding within the stems of
hydrilla, Hydrilla uerticillata (L.f. Royle) (Hydrocharitaceae), a well known pest
aquatic plant that was introduced into Florida in the 1950's (Schmitz et al. 1991). Lar-
vae of C. lebetis mine in the stems of hydrilla, causing sufficient damage to the plant's
apical meristem to preclude further growth of the plant. This natural growth control
may prevent hydrilla from reaching the water's surface, eliminating the dense surface
mats which reduce biodiversity and interfere with navigation.
In this paper the adult male and female of C. lebetis are redescribed and the pupa
and larva are described for the first time. Information on the midge's life history and
potential use as a biocontrol agent for hydrilla is discussed in Cuda et al. (1999).


Morphological terminology and abbreviations follow Seather (1980), Oliver & Dil-
lon (1989), Epler (1988) and Sublette, et al. (1998). Measurements are in Pm, unless
otherwise stated, and consist of the range followed by the mean if three or more spec-
imens were measured.
For the descriptions below, the majority of the adult male material and all of the
adult female, pupal and larval material was from the F3 generation of laboratory
reared midges originally collected from the Plantation Inn Canal, Crystal River in
Citrus Co., Florida, on 23 September 1997 (see Cuda et al. 1999); data from two
paratype males are included in the adult male description.


Sublette (1964) described Cricotopus lebetis from adult male and female speci-
mens collected in Louisiana in 1957-1959. He noted that this new species would key
to C. tricinctus (Meigen) in Johannsen and Townes (1952) and that it was difficult to
separate C. lebetis from Palaearctic material of C. tricinctus (Meigen) on the basis of
color pattern. He also stated (1964: 118) that the "strong, almost right angled basal
lobe [= inferior volsella of Seather (1980) and Oliver & Dillon (1989)] on the basistyle
[= gonocoxite] as well as the shape of the dististyle" [= gonostylus] seemed to be dis-
tinctive for C. lebetis.
Beck & Beck (1966: 131) listed Cricotopus lebetus [sic] as a "recently substituted
American name" for C. tricinctus, but did not give any references or reason for this

Epler et al.: Redescription of Cricotopus lebetis

Hirvenoja (1973: 304), in the section Okologie und Verbreitung ("Ecology and Dis-
tribution") under C. tricinctus, mentioned Beck and Beck's (1966) listing of C. lebetis
as a synonym of C. tricinctus with some doubt as indicated by a "?" before his listing;
he did not list C. lebetis as a synonym of C. tricinctus.
Only Boesel (1983) formally listed C. lebetis as a new synonym of C. tricinctus, a
concept followed by Oliver et al. (1990).
Placement of C. lebetis in the papers above was based only on characters of the
adult stage. When characters from all life stages are considered, in particular the
pupa and larva, it is readily apparent that C. lebetis is a taxon distinct from C. tricinc-

Cricotopus (Isocladius) lebetis Sublette

Cricotopus lebetis Sublette, 1964: 118 (description of adult male and fe-
Cricotopus tricinctus (Meigen, 1818) partim: Boesel 1983:81(synonymy; in
key); Oliver et al. 1990: 24 (synonymy); and other North American au-
Male imago (n = 10, unless otherwise noted)

Color: In life, pale green with blackish-brown markings; these colors fade to pale
brown/stramineous with dark brown to brownish markings in alcohol preserved ma-
terial. In alcohol preserved material, brown to dark brown on antennae, head, tho-
racic vittae (vittae sometimes joined posteriorly by diffuse brown area), scutellum,
postnotum, median anepisternum, almost all to ventral 2/3 of preepisternum, ventral
half of anterior anepisternum II and approximate ventral half of epimeron. Wings
clear with light brown veins; halteres pale. Legs (Fig. 1) with fore and hind coxae
light, mid coxa brown; all trochanters light; fore femur light brown basally, much
darker in apical 1/3 to 12; mid and hind femora basally light with brown apical 1/3 to 1i;
tibiae with brown basal and apical bands, fore tibia slightly darker in middle than
mid and hind tibiae; fore tarsi brown, mid and hind tarsi light brown to stramineous.
Abdomen (Fig. 2) with T I and IV stramineous; T II with posterior 12 brown; T III with
posterior 4/5 brown; T V mostly brown, with paler anterior and posterior margins; T VI
with brown band across middle; T VII mostly stramineous, often with brown mark-
ings posterolaterally, sometimes almost completely dark; T VIII mostly brown, with
narrow posterior stramineous band; T IX mostly brown; gonocoxites and gonostyli
Length. Body (excluding head): 2.35-2.88, 2.55 mm (n = 5); thorax 0.70-0.85, 0.74
mm (n = 5); abdomen 1.65-2.05, 1.86 mm (n = 8).
Head. Temporal setae 6-8, 7; clypeal setae 4-8, 6; cibarial sensillae 2-9, 6. Lengths
of palpomeres 2-5 (n = 8): 30-45, 37; 50-63, 56; 53-73, 63; 88-107, 97.AR 0.77-0.93, 0.83.
Thorax. Setae: lateral antepronotal (n = 8) 0-2, 1; acrostichal (n = 6) 10-13, 12; dor-
socentral (n = 7) 8-14, 10; prealar (n = 9) 3-6, 4; scutellar (n = 9) 6-7, 6.
Wing. Length (n = 7) 1.08-1.30, 1.16 mm; width (n = 7) 310-380, 342. VR (n = 7)
1.14-1.23, 1.18. Costal extension (n = 6) 18-50, 31. Setae: brachiolum 1; squama (n =
9) 4-8, 6; R1 (n = 8) 2-4, 3.
Legs. Lengths of tibial spurs: fore 30-40, 34; mid 12-17, 14 (n = 9); 15-18, 17; hind
14-20, 18; 35-44, 38. Sensilla chaetica: mid 8-15, 12 (n = 9); hind 18-29, 23. Hind tibial
comb with 8-10, 9 setae (n = 8), setal length 27-43, 35. Pulvilli developed, about 12
length of claw. Lengths and proportions of legs:

Florida Entomologist 83(2)

440-600, 510
560-735, 627
260-375, 304
140-210, 164
110-150, 126
80-95, 83
65-80, 73
0.46-0.51, 0.48
3.09-3.42, 3.24
3.50-3.94, 3.75

475-605, 535
480-650, 551
190-275, 226
100-160, 124
85-115, 96
55-70, 63
60-70, 65
0.39-0.42, 0.41
3.53-3.96, 3.78
4.47-5.07, 4.82

June, 2000

430-640, 535
510-730, 604
260-375, 305 (n = 9)
125-185, 151 (n = 9)
120-160, 134 (n = 9)
65-85, 73 (n = 9)
60-80, 68 (n = 9)
0.48-0.55, 0.50
3.15-3.49, 3.34 (n = 9)
3.50-3.89, 3.74

Abdomen (n = 9). T III with 2-4, 3 median setae and 2-4, 3 lateral setae; T IV with
2-5, 3 median setae and 3-5, 4 lateral setae.
Hypopygium (Fig. 3). T IX with 5-10, 7 setae; laterosternite IX with 3-5, 4 setae.
Transverse sternapodeme width 83-155, 100 (n = 9); phallapodeme length 63-78, 70
(n = 9). Virga absent. Superior volsella well developed; inferior volsella (Figs. 4-6) lin-
guiform to triangular. Gonocoxite length 163-205, 179; gonostylus length 80-98, 86,
with well developed crista dorsalis (Fig. 7); GC/GS 1.99-2.13, 2.07. Megaseta length
14-16, 15 (n = 5).
Female imago (n = 5, unless otherwise noted)
Color. Mostly as in male, except thoracic vittae sometimes joined by diffuse brown
coloration, so that dorsum of thorax appears mostly brown (in fluid preserved speci-
mens). Legs light brown, with weaker vittate pattern than in male (pale areas not as
pale as in male). Abdominal tergites similar to male, except T IX and gonocoxite IX
Length. Body (excluding head) 2.23-2.65, 2.51 (n = 3); thorax 0.68-0.77, 0.74; abdo-
men 1.55-1.90, 1.78 (n = 3).
Head. Temporal setae 2-7, 5; clypeal setae 7-9, 8; cibarial sensillae 3-10, 6. Lengths
of palpomeres 2-5 (n = 4): 30-40, 34; 43-53, 49; 48-55, 52; 88-103, 95. Antenna with 5
flagellomeres; AR 0.40-0.55, 0.47.
Thorax. Setae: lateral antepronotal 1-2, 2; acrostichal 9-12, 11; humeral 2-3, 2;
dorsocentral 7-10, 9; prealar 3-4, 3; scutellar 5-8, 7.
Wing. Length (n = 4) 1.08-1.20, 1.00 mm; width (n = 4) 380-420, 395. VR (n = 4)
1.18-1.22, 1.20. Costal extension (n = 3) 40-60, 50. Setae: brachiolum 1; squama 4-11,
7; R (n = 4) 4-5, 4; R, (n = 4) 0-3, 2; R4,5 (n = 4) 1-3, 2.
Legs. Lengths of tibial spurs: fore 21-25, 23; mid 10-13, 12; 12-15, 14; hind 11-15,
13; 33-38, 36. Sensilla chaetica: mid 18-21, 20; hind 22-34, 27. Hind tibial comb with
9-12, 10 setae, setal length 35-37, 36 (n = 3). Pulvilli developed, about /2 length of claw.
Lengths and proportions of legs (n = 4):

370-430, 406
465-520, 500
210-245, 226
95-120, 106
60-90, 76
45-60, 50
50-60, 55
0.44-0.47, 0.45
3.82-4.18, 3.95
3.88-4.11, 4.01

420-490, 458
430-480, 468
175-200, 190
80-90, 86
60-75, 65
40-50, 44
0.40-0.42, 0.41
4.33-4.67, 4.46
4.85-4.89, 4.87

420-490, 461
490-530, 520
245-280, 268
100-125, 113
90-110, 101
50-60, 58
0.50-0.53, 0.51
3.75-4.02, 3.91
3.57-3.71, 3.67

Epler et al.: Redescription of Cricotopus lebetis

Abdomen (n = 3). T III with 2-3, 2 median setae and 2 lateral setae; T IV with 2 me-
dian setae and 2-3, 2 lateral setae.
Genitalia (Figs. 8-10). Notum 105-143, 124 long (measured to bifurcation); seminal
capsule diameter 50-60, 57 (n = 3), cercus length 85-103, 97 (n = 4) (measured from
ventral aspect). Spermathecal ducts with at least one loop. Coxosternapodeme as in
Fig. 10. T IX with 3-5, 4 setae; gonocoxite IX with 8-12, 11 setae.

Pupa (n = 10, unless otherwise noted)

Color. Exuviae pale yellow with narrow, light brown bands at posterior of T II (over
booklet row) and T III; T IV entirely pale; T V with posterior 2/3 light brown; T VI with
median light brown area; lateral margins of T VI-VIII and anal lobes light brown.
Length. Total 2.50-3.18, 2.79 mm (n = 8); cephalothorax 0.75-0.95, 0.81 mm (n = 7);
abdomen 1,70-2.30, 1.99 mm (n = 9).
Cephalothorax. Frontal setae 60-88, 73 (n = 6) long, 2 wide; dorsal median an-
tepronotal seta 58-73, 64 (n = 6) long; ventral median antepronotal seta 83-155, 96 (n
= 8) long; lateral antepronotal seta 25-40, 33 (n = 8) long. Median suture area smooth.
Thoracic horn (Fig. 11) fusiform with sparsely scattered minute spinules; 50-83, 67
long; 15-20, 17 (n = 9) maximum width. Precorneal setae lengths: PC, 88-105, 101 (n
= 9); PC2 65-93, 77 (n = 8); PC3 63-100, 82 (n = 7). Dorsocentral setae lengths: DC, 33-
48, 39; DC2 25-60, 36 (n = 9); DC3 30-40, 36 (n = 8); DC, 35-48, 39 (n = 9); DC, stouter
than DC2. Wing sheath without bacatiform papillae or nasiform tubercles.
Abdomen (Fig. 12). T II with 49-68, 56 booklets arranged in double row. Pedes spu-
rii B weakly developed on T II; pedes spurii A present on S IV-VI. Tergite I with one
anterolateral seta, T II-VIII with 3 lateral setae (2 dorsal and one ventral). Dorsal
shagreen on T I sparse, scattered minute spinules in weak, longitudinal lateral bands;
T II with scattered weak spinules over most of surface; T III with fine spinules over
most of surface; T IV-VI with larger spinules over most of surface, with coarser
spinules at center of tergites; T VII-VIII with anterior band of fine spinules; anal disc
with small anteromedian area of fine spinules. Conjunctiva III-IV, IV-V and V-VI with
spinules. Ventral shagreen consists of small posterolateral groups of minute spinules
on S I; S II-V with weak longitudinal bands of minute spinules; S VI-VII with small
anterolateral groups of minute spinules. Anal lobes with 3 macrosetae; anal lobe
length 200-218, 208 (n = 7). Lengths of anal lobe macrosetae (n = 9): seta 1 (anterior-
most seta) 78-88, 82; seta 2 (middle seta) 73-95, 85; seta 3 80-102, 91. The anal lobe
ratio (ALR) varies depending on which macroseta is measured and compared to the
anal lobe length, thus ALR1 (using anteriormost seta) 0.39-0.43, 0.40; ALR2 0.38-
0.45, 0.42; ALR3 0.40-0.47, 0.43 (all ALR n = 7).

Fourth instar larva (n = 11, unless otherwise noted)

Color. In life, the body is green with blue bands on the second and third thoracic
segment; the blue color is bleached on alcohol preserved specimens but the thorax re-
mains darker than the remainder of the body in such material. Head capsule pale yel-
low-brown, premandibles light brown, mentum and apical 1/3 to 12 of mandible dark
brown to black. Claws of parapods translucent to pale brown.
Head. Postmentum length 170-205, 186 (n = 9). Labrum (Fig. 13) with simple S I.
Total antennal length 61-73, 64 (Fig. 14). Length of antennal segments 1-5: 31-43, 37;
13-18, 16; 9-13, 11; 3-4, 4; 3-4, 4; 2. Ring organ 8-10, 9 (n = 5) from base of basal seg-
ment; sensory pits slightly above to around same level as ring organ. Lauterborn or-
gans extend to apex of antennal segment 3. AR 1.11-1.60, 1.34. Premandible (Fig. 15)
apically bifid; length 69-80, 74. Mandible (Fig. 16) length 120 = 137, 127; with 3 inner

Florida Entomologist 83(2)




Figs. 1-10. Cricotopus lebetis adult structures. 1. Male fore, mid and hind legs; 2.
Male abdomen; 3. Hypopygium; 4-5. Inferior volsella variation in Florida material; 6.
Inferior volsella, Louisiana specimen; 7. Variation of gonostylus due to angle of obser-
vation; 8. Female genitalia, ventral; 9. Female genitalia, lateral; 10. Female coxoster-

June, 2000





, /

Epler et al.: Redescription of Cricotopus lebetis

teeth; apical tooth length 13-16, 14 (n = 5); width of inner teeth 23-27, 25 (n = 5). Outer
margin of mandible mostly smooth; inner margin of mandible without spines; man-
dibular margin at base of seta subdentalis without minute teeth. Seta internal
present, usually with 6 branches. Mentum (Fig. 17) with 13 teeth; second lateral tooth
small and fused to first. Maxilla as in Fig. 18.
Body. Small claws of anterior parapods (Fig. 19) with apical tooth much larger
than inner teeth. Abdominal segments I-VII with long setal tufts (Fig. 20); setal tufts
with about 25-50 setae, longest setae about 385 long; tuft on VII with smaller and
fewer (about 11-20) setae. Anal tubules elongate-ovoid.


The color pattern of adults can be variable in many species of Cricotopus; C. lebetis
is no exception. In males, tergite VII is apparently most susceptible to color variation;
it may be almost totally unmarked with brown or, as in the majority of the Florida ma-
terial examined, marked with brown in the posterolateral corners. Sublette (pers.
comm.) has seen material of C. lebetis from Baton Rouge, Louisiana, in which T VII is
almost completely infuscate.
In general, the type series is darker than the Florida material examined. Sublette
(1964) stated that the female pronotum (= antepronotum) was infuscate; in the Flor-
ida material examined the antepronotum is stramineous, similar to other unmarked
body areas.
Adults of C. lebetis are most likely to be confused with C. tricinctus, as originally
noted by Sublette (1964). In males, the inferior volsella of C. tricinctus is generally
narrower and more triangular than that of C. lebetis. There is variation in the shape
of this lobe; figure 6 is from a Louisiana paratype and is similar to, but still shorter
and broader than, the volsella of some European C. tricinctus (see Hirvenoja 1973:
Fig. 189(3) and 1 ---". ".
Females of C. lebetis are also similar to C. tricinctus; lower counts of sensilla cha-
etica on the mid and hind basitarsi (means of 22 and 27) will separate the Florida C.
lebetis females examined from those of C. tricinctus (means of 48 for both legs), follow-
ing the data from Hirvenoja (1973). It must be noted that the descriptions above (for
all life stages of C. lebetis, with the exception of the two males from Louisiana) are for
individuals from one population in Florida. These individuals may be smaller than
populations of this species from other areas; setal counts and other measurements
may also show a wider range once more material is examined.
Pupae of the two species are also similar, but the thoracic horn of C. lebetis is fusi-
form compared to the elongate digitiform horn of C. tricinctus (see Hirvenoja 1973:
Fig. 190(2)).
Larvae of C. lebetis are similar to other members of the C. sylvestris group, but dif-
fer in bearing setal tufts on abdominal segments I-VII; in C. tricinctus larvae, only ab-
dominal segments I-VI possess setal tufts. All larvae of C. lebetis examined had simple
S I; the S I of related species are bifid (but note that the bifurcation of some of these
other species may be unequal, with one branch much larger than the other; see Hir-
venoja 1973: Fig. 191(3)).
An unresolved problem is that of the origin of North American Cricotopus lebetis.
Is this a species that was introduced with hydrilla, or is it native to North America
and has seized upon introduced hydrilla as a suitable host plant? Is C. lebetis a facul-
tative miner of hydrilla or does it attack other plants? Hirvenoja (1973) noted that Eu-
ropean C. tricinctus larvae mine in the leaves of Potamogeton, and that other aquatic
plant species may also be attacked.

Florida Entomologist 83(2)



June, 2000

14 15



Figs. 11-20. Cricotopus lebetis pupal structures; 13-20, larval structures. 11. Vari-
ation in thoracic horn; 12. Abdomen, dorsal; 13. Labrum; 14. Antenna; 15. Premandi-
ble; 16. Mandible; 17. Mentum; 18. Maxilla; 19. Small claws of anterior parapod; 20.
Lateral setae of abdominal segment II.

/ 19


Epler et al.: Redescription of Cricotopus lebetis

As seen above, C. tricinctus and C. lebetis have been confused in North America;
Sublette (pers. comm.) has seen true C. tricinctus specimens from the north central
U.S. and compared them with Palaearctic material from the British Museum (Natural
History) and from Hirvenoja's collection; thus both species occur in the Nearctic. The
Crystal River specimens constitute the first record of C. lebetis from Florida. The pos-
sibility exists that C. lebetis has been misidentified as C. tricinctus or other species in
other countries where hydrilla occurs (i.e., Japan). Cricotopus nitens (Kieffer, 1921)
and C. taiwanus Tokunaga, 1940, both described from Taiwan, have a similar color
pattern; if not distinct species, either may be a senior synonym. The best way to solve
this riddle would be to rear all life stages of hydrilla-associated Cricotopus throughout
the range of the plant.


We are extremely grateful to Dr. J. E. Sublette, Tucson, AZ, for his assistance in
identifying C. lebetis, reviewing an early draft of this paper and the loan of specimens.
We also thank B. A. Caldwell for providing a pre-submission review of this paper and
Dr. C. L. de la Rosa for providing the Spanish abstract. The senior author wishes to es-
pecially thank Dr. Barry Merrill and Judy Merrill (Merrill Consultants, Dallas, TX)
for providing laboratory and computer equipment. This is Florida Agricultural Exper-
iment Station Journal Series No. R-07159.


BECK, W. M. JR., AND E. C. BECK. 1966. The Chironomidae of Florida: A problem in
international taxonomy. Gewasser und Abwasser 41/42: 129-135.
BOESEL, M. W. 1983. A review of the genus Cricotopus in Ohio, with a key to adults of
species of the northeastern United States (Diptera, Chironomidae). Ohio J. Sci.
83: 74-90.
CUDA, J. P., B. R. COON, J. L. GILMORE, AND T. D. CENTER. 1999. Preliminary report
on the biology of a hydrilla stem tip mining midge (Diptera: Chironomidae).
Aquatics 21(4): 15-18.
EPLER, J. H. 1988. Biosystematics of the genus Dicrotendipes Kieffer, 1913 (Diptera:
Chironomidae: Chironominae) of the world. Mem. American Entomol. Soc. 36:
HIRVENOJA, M. 1973. Revision der Gattung Cricotopus van der Wulp und ihrer Ver-
wandten (Diptera, Chironomidae). Ann. Zool. Fennici 10: 1-363.
JOHANNSEN, 0. A., AND H. K. TOWNES. 1952. Tendipedidae (Chironomidae). Guide to
the Insects of Connecticut. Part VI. The Diptera or true flies of Connecticut.
Fifth fasicle: Midges and Gnats. St. Geol. Nat. Hist. Surv. Bull. 80: 3-26.
KIEFFER, J. J. 1921. Chironomides des Philippines et de Formose. Philippine J. Sci. 18:
OLIVER, D. R., AND M. E. DILLON. 1989. 2. The adult males of Chironomidae (Diptera)
of the Holarctic region-Key to subfamilies. Entomol. Scandinavica Suppl. 34:
OLIVER, D. R., M. E. DILLON, AND P. S. CRANSTON. 1990. A catalog of Nearctic Chi-
ronomidae. Research Branch Agriculture Canada Pub. 1857/B. 89 pp.
S/ETHER, 0. A. 1980. Glossary of chironomid morphology terminology (Diptera: Chi-
ronomidae). Entomol. Scandinavica Suppl. 14: 1-51.
SCHMITZ, D. C., B. V. NELSON, L. E. NALL, AND J. D. SCHARDT. 1991. Exotic plants in
Florida: a historical perspective and review of the present aquatic plant regu-
lation program. pp. 303-326 in Center, T. D., R. F. Doren, R. L. Hofstetter, R. L.
Meyers, and L. D. Whiteaker (eds.), Proceedings, Symposium on Exotic Plant

180 Florida Entomologist 83(2) June, 2000

Pests, 2-4 November 1988, Miami, Florida. U.S. Dept. of the Interior, National
Park Service, Washington, D.C.
SUBLETTE, J. E. 1964. Chironomidae (Diptera) of Louisiana. I. Systematics and imma-
ture stages of some lentic chironomids of west-central Louisiana. Tulane Stud.
Zool. 11: 109-150.
SUBLETTE, J. E., L. E. STEVENS, AND J. P. SHANNON. 1998. Chironomidae (Diptera) of
the Colorado River, Grand Canyon, Arizona, USA, I: Systematics and ecology.
Great Basin Naturalist 58: 97-146.
TOKUNAGA, M. 1940. Chironomoidea from Japan (Diptera), XII. New or little-known
Ceratopogonidae and Chironomidae. Philippine J. Sci. 72: 255-311 + 4 plates.

Florida Entomologist 83(2)

June, 2000


'Present address: USDA-APHIS, P.O. Box 1040, Waimanalo, HI 96795
and Hawaiian Evolutionary Biology Program, University of Hawaii
Honolulu, HI 96822

2Organization for Tropical Studies-Undergraduate Semester Abroad Program
Duke University, 410 Swift Avenue, Durham, NC 27705

3Students (in alphabetical order): Lisa Bell, Aisha Burden, Mark Fox,
Ilmi Granoff, Nihara Gunawardene, Melisa Holman, Allison Hornor,
Jane MacLeod, Julia Michalek, Casuarina McKinney-Richards,
Adam Ruff, Aaron Smith, Darcy Thomas, and Olivia Watson


Flower selection and pollen-collecting effort were monitored for 3 species of bees
that sonicate flowers of Solanum wendlandii Hook. for pollen in southern Costa Rica.
Between 0700-0900 hours, Bombus pullatus (Fkln.), Euglossa erythrochlora Moure,
and Pseudaugochloropsis graminea (Fabricius) foraged more frequently at new flow-
ers (that had opened the day of observation) than old ones (that had opened at least
1 day before observation). Between 0900-1100 hours, however, this preference was no
longer evident, and all 3 species visited new and old flowers with similar frequency. E.
erythrochlora and P. graminea spent more time harvesting pollen during 1) initial
(first or second) visits to new flowers than initial visits to old flowers and 2) initial vis-
its to new flowers than final (seventh or later) visits to new flowers. Similar, although
not statistically significant, trends were evident for B. pullatus as well. An experi-
ment using pollinator exclusion bags revealed that the reduced foraging effort at in-
dividual flowers was resource-dependent and was not simply a time-dependent

Key Words: Apidae, buzz pollination, Costa Rica, foraging behavior, Halictidae,

Shelly et al.: Buzzing Bees on Solanum 181


Se monitored la selecci6n de flores y el esfuerzo en recolectar polen de tres species
de abejas que frecuentan flores de Solanum wendlandii Hook. en al Sur de Costa Rica.
Entre las 0700 y 0900 h, Bombus pullatus (Fkln.), Euglossa erythrochlora Moure y
Pseudaugochloropsis graminea (Fabricius) forrajearon con mayor frecuencia flores
nuevas (flores que abrieron el mismo dia de hacerse la observaci6n) que flores viejas
(flores que abrieron por lo menos un dia antes de hacerse la observaci6n). Sin em-
bargo, esta preferencia no se observe entire las 0900 y 1100 h, ya que las tres species
visitaron flores nuevas y viejas con la misma frecuencia. E. erythrochlora y P. grami-
nea emplearon mas tiempo cosechando polen en visits iniciales (primera o segunda
visits) a flores nuevas que en visits iniciales a flores viejas. Ademas, emplearon mas
tiempo en visits iniciales a flores nuevas que en visits finales (s6ptima visit en ade-
lante) a flores nuevas. Se observe una tendencia similar (aunque no estadisticamente
significativa) en B. pullatus. Un experiment empleando bolsas para excluir poliniza-
dores demostr6 que la reducci6n en el esfuerzo por forrajear flores individuals estuvo
determinada por la disponibilidad de alimento y no por el horario.

The manner in which nectar rewards influence flower selection and floral handling
times has been well studied for a number of bee species (e.g., Cresswell 1990, Giurfa
and Nunez 1992, Dukas and Real 1993). Early studies of nectar-collecting by bees
(e.g., Waddington and Heinrich 1979, Pyke 1979) lent strong empirical support to the
development of optimal foraging theory (Krebs and McCleery 1984). In contrast,
fewer studies have investigated the foraging choices of bees harvesting pollen, and
these have yielded mixed results regarding the ability of bees to assess pollen rewards
from individual flowers. For example, Haynes and Mesler (1984) observed bumblebees
foraging on inflorescences of a lupine species and found bees did not discriminate be-
tween old, pollen-poor flowers and younger flowers that contained greater pollen re-
wards. Based on patterns of turning frequency and directionality, Hodges and Miller
(1981) similarly concluded that bumblebees were not adjusting their foraging move-
ments in response to pollen availability.
On the other hand, several studies have demonstrated that bees do modify their
foraging behavior in response to either anticipated or actual pollen rewards for in-
dividual floral visits. In the first demonstration of "distant assessment", Pellmyr
(1988) showed that bumblebees assessed pollen rewards prior to alighting, based on
age-dependent changes in floral shape, and rejected old flowers in favor of younger,
pollen-rich flowers. He also showed that bumblebees adjusted their handling time
with pollen availability and spent more time on younger flowers than older ones.
Likewise, Gori (1989) experimentally removed pollen and found that bumblebees re-
sponded by visiting fewer flowers per inflorescence. Buchmann and Cane (1989) and
Harder (1990) also reported a positive relation between pollen availability and han-
dling time for individual floral visits, indicating immediate assessment of pollen re-
The present study examined whether bees foraging on Solanum wendlandii Hook.
selectively visited younger (pollen-rich) flowers over older flowers and also spent more
time foraging on younger than older flowers. In addition, a pollinator exclusion exper-
iment was conducted to assess whether, among young flowers, floral handling time
varied between "virgin" vs. previously visited flowers.

Florida Entomologist 83(2)

June, 2000


Flowers of S. wendlandii have purple petals that fade with age (flowers probably
last no more than 3 days). Five, large, tubular anthers are present with distal sections
purple and basal parts yellow. Nectaries are absent, and flowers offer only pollen,
which is released through 2 minute apical pores per anther (Michener 1962). Bees are
able to gather the pollen efficiently only via sonication or buzzing of the anthers. Vis-
iting bees grasp the anther cone and rapidly contract their indirect flight muscles
(thus producing an audible sound or buzz), which transfers vibrations to the anthers
and expels pollen onto the bee. Bees then groom and transfer the pollen to special
structures (or corbiculae) on their hind legs for storage.
The study was conducted at the Las Cruces Biological Station of the Organization
for Tropical Studies in southwestern Costa Rica. The patch of S. wendlandii observed
was growing on a stone wall in the station clearing immediately adjacent to mixed pri-
mary and secondary pre-montane forest (elevation 1,100 m). Data were gathered from
18 September to 4 October, 1997, an interval falling toward the end of a 9-month rainy
season. Observations were restricted to sunny days with air temperatures ranging be-
tween 20-23C.
Visits by 3 species of buzz-pollinating bees were recorded continuously for individ-
ually tagged flowers ofS. wendlandii between 0700-1100 hours over 4 days. Based on
preliminary observations, the peak period of floral visitation occurred between 0800-
1000 hours. Buzz-pollinators were never seen at the flowers prior to 0715 hours, and
thus our data most likely describe complete sequences of bee visitation to the focal
flowers. The 3 principal buzz-pollinating species included (in order of increasing body
size) a halictid, Pseudaugochloropsis graminea (Fabricius), a euglossine, Euglossa
erythrochlora Moure, and a bumblebee, Bombus pullatus (Fkln.). Several other Eu-
glossa spp. were observed sonicating the flowers, but these were infrequent visitors.
On a given morning, 1-8 observers recorded the time of day, bee species, and dura-
tion of pollen collecting (to the nearest s) for individual foraging visits to 2-5 pairs of
flowers, each pair consisting of a "new" (i.e., newly opened the same day as the obser-
vation) and an "old" (i.e., open for at least 1 day prior to the observation) flower. Op-
erationally, the duration of pollen collection (here termed handling time) was equated
with audible buzzing of the anthers. Flower age was determined by tagging stems 1
day before making observations with a small piece of green tape. Tags were placed be-
low fully developed buds (set to open the following morning and be "new" flowers) and
the closest (already) open flower. Paired new and old flowers were less than 30 cm
apart in all cases.
We compared handling times of 1) "initial" visits (first or second visit observed over
all bee species) to new vs. old flowers, 2) initial vs. "final" visits (seventh or later visit
observed over all bee species) to new flowers, and 3) final visits to new flowers vs. ini-
tial visits to old flowers. Note that the terms "initial" and "final" refer to the sequential
order of visits compiled over all bee species for particular flowers and not to inter-flo-
ral visitation sequences for particular bees. Counting visits independently of species
identity provided only a rough index of pollen depletion, since possible interspecific
differences in pollen removal were not documented. Note also that initial visits to old
flowers refer, not to their first or second visits in absolute terms, but to the first or sec-
ond visits recorded during our observations. Being at least 1 day old, old flowers had
most likely been visited multiple times on the day(s) prior to our observations (a valid
assumption given that 49 of the 52 new flowers we observed received 2 or more visits
by sonicating bees).
As shown below, for 2 of the species there was a significant reduction in the time
spent gathering pollen at new flowers through the morning, i.e., between early (0700-

Shelly et al.: Buzzing Bees on Solanum 183

0900 hours) and late (0900-1100 hours) observation periods. To determine whether
this decrease was time-dependent (foraging rule: if late morning, spend less time per
new flower) vs. resource-dependent (foraging rule: if pollen supply depleted, spend
less time per new flower), we placed fine mesh bags on a total of 24 flower buds 1 day
before opening. Buds were enclosed completely with nylon mesh screening secured to
the stem with a wire clasp. Bags were sufficiently large that the exposed anthers of
newly opened flowers were well below the screening, out of reach of potential pollina-
tors. For each bagged flower, we tagged (but did not bag) a nearby (within 30 cm) fully
developed bud. The following morning we removed the bags at 0900 hours and contin-
uously recorded visits for the next 2 h; unbagged, new flowers were observed contin-
uously from 0700-1100 hours. On a given day, an observer monitored visits
continuously to 8 pairs of new flowers (i.e., 1 bagged, 1 unbagged). This experiment
was completed over 3 mornings.
Upon completing behavioral observations, we collected flowers to estimate pollen
supplies for (1) new, unvisited flowers (buds were bagged 1 day before opening and
bagged flowers were collected the following day, (2) new, visited flowers (buds were
tagged but not bagged 1 day before opening and flowers were collected the following
morning), and (3) old, visited flowers (already open flowers were tagged and collected
the following day. All flowers were collected at 1100 hours, and anthers were removed
with a forceps, placed in a drying oven (55C) for 24 h, and weighed to the nearest
0.001 g using a Mettler AE260 Analytical Balance.
Variation in floral handling times within and among bee species and anther
weights among flowers was first analyzed using one-way ANOVA to detect significant
inter-group variation overall and then the Tukey test to identify significant differ-
ences between specific groups (log10 transformed values were used in both tests to con-
trol for the association between mean and variance; Zar 1996). However, data on floral
visitation were not normally distributed (even after log,10 transformation), and conse-
quently for this parameter within-species variation was first analyzed using the
Kruskal-Wallis test (ANOVA by ranks) and then the non-parametric Dunn's test
(Daniels 1990). In all cases, there was direct correspondence between the 2 types of
tests: when ANOVA (or the Kruskal-Wallis test) detected (or, conversely failed to de-
tect) significant variation overall, the Tukey test (or Dunn's test) also identified (or,
conversely, failed to identify) specific, significant inter-group differences. Conse-
quently, only the results of the Tukey tests (or Dunn's tests) are presented. For pair-
wise comparisons in the flower bagging experiment, we used the Mann-Whitney test,
a nonparametric analogue of the Student's t test (Zar 1996).


The 3 species displayed the same temporal pattern of abundance (Fig. 1). Few flo-
ral visits occurred before 0730 hours. Activity peaked between 0730-0830 hours and
then declined steadily until the end of observations at 1100 hours. Despite frequent
checks, no bees were seen at the flower patch in the afternoon or early evening. For all
3 species, approximately 2/3 of all floral visits occurred prior to 0900 hours.
The species also exhibited the same pattern of preference for floral age (Table 1).
Individuals of all species preferred new over old flowers in the early morning but
showed no such preference later in the morning. The early morning preference for
new flowers was quite pronounced: new flowers were, on average, visited 3-4 times
more often than old flowers. Reflecting the decline in overall activity, visitation rates
to new flowers decreased significantly between early and late morning for all species.
Visitation rates declined with time for old flowers as well, though this difference was
not significant for any bee species.

184 Florida Entomologist 83(2) June, 2000

---- P. graminea
35 -A E. erythrochlora
30/ -o- B. pullatus

20 -


A- A
o ~~ ~ ~ ~ .. ---------------
7 8 9 10 11

Time of day (h)

Fig. 1. Total numbers of visits recorded over all focal flowers (N = 52 pairs of new
and old flowers) in relation to time of day.

During initial visits, P. graminea and E. erythrochlora spent significantly more
time collecting pollen from new flowers than old ones (Table 2). Also, when visiting
new flowers, these same species spent significantly more time collecting pollen during
initial visits than final ones. These same trends were apparent for B. pullatus as well,
although they were not statistically significant (Table 2). Handling times for initial
visits to new flowers were similar to initial visits to old flowers in all 3 species (Table 2).
Regarding interspecific comparisons, P. graminea spent significantly more time
collecting pollen during initial visits to new flowers than E. erythrochlora or B. pulla-
tus (Table 2). Handling times did not vary among species for final visits to new flowers
or initial visits to old flowers.
The incidence of initial and final visits to new flowers was not independent of the
time of day, and the majority (51/69 = 74% over all species) of final visits to new flow-
ers occurred after 0900 hours. Thus, as noted above, the decrease in buzzing durations
between initial and final visits to new flowers noted for P. graminea and E. erythro-
chlora may have reflected a time-dependent, rather than a resource-dependent, shift
in foraging behavior. However, in the flower bagging experiment mean handling times
were greater for the initial visits to "virgin" new flowers than for the concurrent, final
visits to unbagged, new flowers for all 3 species P. graminea: 26 s vs. 11 s, respec-
tively (n1 = 11, n2 = 8); E. erythrochlora: 15 s vs. 6 s, respectively (n1 = 6, n2= 6); B. pul-
latus: 18 s vs. 8 s, respectively (n, = 17, n2 = 17); P < 0.05 in all cases, Mann-Whitney
test). Pollen-collection times for initial visits to previously bagged, new flowers were
similar to those recorded for initial visits to unmanipulated, new flowers in the early
morning for all 3 species (P > 0.05 in all cases; Mann-Whitney test). Despite these

Shelly et al.: Buzzing Bees on Solanum 185


Early Late

New Old New Old

P. graminea 1.4a (1.4) 0.3b (0.6) 0.6b (1.0) 0.25b (0.6)
E. erythrochlora 0.8a (0.9) 0.3b (0.4) 0.4b (0.6) 0.2b (0.5)

B.pullatus 1.8a (1.5) 0.5b (0.5) 0.9b (1.0) 0.5b (0.5)

findings, bees visited virgin and previously available new flowers with similar fre-
quency: over all 3 species, the mean numbers of visits recorded between 0900-1100
hours were 1.85 (_1.2) and 2.0 (_1.3) for virgin and unbagged new flowers, respec-
tively (P > 0.05; Mann Whitney test).
Anther (mg dry) weights were significantly greater for new, unvisited flowers (x
+ 1 SD = 149 + 14 mg) than either new, visited (131 + 12 mg) or old, visited (125 + 9
mg) flowers (P < 0.05 in both cases; Tukey test using transformed [log,0 x] data).
Among visited flowers, anthers from new flowers weighed more than those from old
flowers, although this difference was not statistically significant (P > 0.05; Tukey test
using transformed [log x0]J data).


Foraging choices frequently involve "non-energetic" benefits (Rasheed and Harder
1997). That is, although their collection requires energy expenditure, the resources do
not provide energy directly to the forager but serve other functions. For bees, pollen
harvesting yields non-energetic benefits: pollen serves primarily as a protein source
for developing larvae, while nectar is the chief energy source for flight and other ac-


New Old

Initial Final Initial

P. graminea 37aA (24, 34) 15bA (10, 24) 10bA (6, 19)
E. erythrochlora 18aB (11, 19) 8bA (4, 14) 6bA (4, 14)

B.pullatus 14aB (13, 45) 9aA (6, 31) 8aA (6, 29)

Florida Entomologist 83(2)

June, 2000

tivities (Heinrich 1979). In addition to the fact that nectar is an easily measured and
manipulated resource, the research emphasis on nectar foraging by bees derives
largely from the working assumption that energy is an appropriate fitness "currency"
for foraging animals. Nonetheless, energy-mediated constraints on bee activity might
be expected to generate a common pattern of foraging behavior regardless of the type
of resource collected (i.e., energy- or non-energy-based). In fact, Rasheed and Harder
(1997) found that pollen-gathering bumblebees foraged in a manner qualitatively
similar to that reported for nectar-gatherers, i.e., in both cases, bees maximized for-
aging efficiency or benefit-to-metabolic cost ratio.
The present study provides additional evidence that pollen-foraging bees modify
their behavior in response to anticipated and actual pollen returns from individual
flowers. All 3 species studied preferred new over old flowers in the early morning. As
petal color changed greatly with age, it seems likely that bees used reflectance pat-
terns (in the visual or ultraviolet spectra) as long-distance cues of pollen supplies. By
late morning, however, the bees did not differentiate between new and old flowers,
and, as the measurements of anther weight suggest, this shift reflected an increased
similarity in the pollen abundance of new and old flowers. Thus, the bees presumably
could distinguish between new and old flowers late in the morning but, owing to re-
duced pollen levels in the new flowers, "ignored" this distinction. Additional data on
approach and rejection probabilities for flowers of different ages are required to con-
firm color-based discrimination in the early morning.
In contrast, bees were apparently unable to make long-range assessment of pollen
availability among new flowers. In the bagging experiment, bees visited virgin and
unbagged new flowers with equal frequency. This finding was not unexpected, since
the pollen is concealed within minute pores on the anther (Michener 1962). Potential
cues, such as pollen odor (Buchmann 1983) or "bruise marks" left on the anthers by
previous visitors (J. Cane, pers. comm.), were presumably either absent or weak. Con-
sistent with other studies, therefore, our data suggest that long-distance assessment
of nectar (Neff and Simpson 1990) or pollen (Pellmyr 1988) rewards may depend ex-
clusively on "gross" features of floral morphology, such as overall shape or color.
Upon alighting, bees clearly adjusted their harvesting effort to match pollen avail-
ability. In the bagging experiment, handling times for all 3 species were significantly
longer for virgin, new flowers than for unbagged, new flowers. Observations for P.
graminea and E. erythrochlora showed that initial visits to new flowers were signifi-
cantly longer than either initial visits to old flowers or final visits to new flowers.
Other studies (Pellmyr 1988, Buchmann and Cane 1989, Harder 1990) report this
same trend, supporting the general observation that, as bee visitation continues in a
flower patch, individual foragers encounter diminishing pollen returns per flower and
therefore spend less time at individual flowers.
In sum, our data indicate that pollen-collecting bees are sensitive to varying re-
source levels within individual flowers and respond by selecting and intensively han-
dling more rewarding flowers. Thus, while the nature of the rewards differ, pollen-and
nectar-foraging bees appear similar in attempting to maximize the rate of resource
collection. Instead of maximizing the rate of fuel (nectar) intake, however, pollen-col-
lecting bees forage in a manner that increases the rate of pollen delivery to developing


We thank Luis Diego Gomez for weather data and for identifying the plant, Raul
Rojas for much logistical support, Jim Ackerman for supplying a key for euglossines,

Shelly et al.: Buzzing Bees on Solanum 187

the staff of the Instituto Nacional de Biodiversidad for assistance in identifying the
bees, and Emma and Miranda Shelly for assistance with data collection. We also
thank Jim Cane for supplying references and encouragement and Jack Neff for help-
ful comments on an earlier draft.


BUCHMANN, S. L. 1983. Buzz pollination in angiosperms, pp. 73-113 in C. E. Jones and
R. J. Little (eds.). Handbook of experimental pollination biology. Van Nostrand
Reinhold, New York.
BUCHMANN, S. L., AND J. H. CANE. 1989. Bees assess pollen returns while sonicating
Solanum flowers. Oecologia (Berl) 81: 289-294.
CRESSWELL, J. E. 1990. How and why do nectar-foraging bumblebees initiate move-
ments between inflorescences of wild bergamot Monarda fistulosa (Lamiaceae).
Oecologia (Berl) 82: 450-460.
DANIELS, W. W. 1990. Applied nonparametric statistics. PWS-KENT Publishing, Boston.
DUKAS, R., AND L. A. REAL. 1993. Effects of recent experience on foraging decisions by
bumble bees. Oecologia (Berl) 94: 244-246.
GIURFA, M., AND J. NUNEZ. 1992. Foraging by honeybees on Carduus acanthoides:
pattern and efficiency. Ecol. Entomol. 17: 326-330.
GORI, D. F. 1989 Floral color change in Lupinus argenteus (Fabaceae): why should
plants advertise the location of unrewarding flowers to pollinators? Evolution
43: 870-881.
HARDER, L. D. 1990. Behavioral responses by bumble bees to variation in pollen avail-
ability. Oecologia (Berl) 85: 41-47.
HAYNES, J., AND M. MESLER. 1984. Pollen foraging by bumblebees: foraging patterns
and efficiency on Lupinus polyphyllus. Oecologia (Berl) 61: 249-253.
HEINRICH, B. 1979. Bumblebee economics. Harvard University Press, Cambridge, MA.
HODGES, C. M., AND R. B. MILLER. 1981. Pollinator flight directionality and the as-
sessment of pollen returns. Oecologia (Berl) 50: 376-379.
KREBS, J. R., AND R. H. MCCLEERY. 1984. Optimization in behavioral ecology, pp. 91-
121 in J. R. Krebs and N. B. Davies (eds.). Behavioural ecology: an evolutionary
approach. Blackwell Scientific Publications, London.
MICHENER, C. D. 1962. An interesting method of pollen collecting by bees from flowers
with tubular anthers. Rev. Biol. Trop. 10: 167-175.
NEFF, J. L., AND B. B. SIMPSON. 1990. The roles of phenology and reward structure in
the pollination biology of wild sunflower (Helianthus annuus L., Asteraceae).
Israel. J. Bot. 39: 197-216.
PELLMYR, 0. 1988. Bumble bees (Hymenoptera: Apidae) assess pollen availability in
Anemonopsis macrophylla (Ranunculaceae) through floral shape. Ann. Ento-
mol. Soc. America 81: 792-797.
PYKE, G. H. 1979. Optimal foraging in bumblebees: rules of movement between flow-
ers within inflorescences. Anim. Behav. 27: 1167-1181.
RASHEED, S. A., AND L. D. HARDER. 1997. Foraging currencies for non-energetic re-
sources: pollen collection by bumblebees. Anim. Behav. 54: 911-926.
WADDINGTON, K. D., AND B. HEINRICH. 1979. The foraging movements of bumblebees
on vertical "inflorescences": an experimental analysis. J. Comp. Physiol. 134:
ZAP, J. H. 1996. Biostatistical analysis. Prentice-Hall, Upper Saddle River, New Jersey.

Florida Entomologist 83(2)

June, 2000


Caroni Research Station
Waterloo Road, Carapichaima, Trinidad & Tobago, West Indies

Rice Thrips, Stenchaetothrips biformis Bagnall (Thysanoptera: Thripidae) was
first observed in Trinidad at the rice department of Caroni (1975) Ltd.', on July 29
The specimens were identified by A. K. Walker of the British Museum of Natural
History. The Identification was facilitated by CARINET, the Caribbean loop of Bionet
The thrips were observed on seedling rice (Oryza sativa L grown under irrigated
conditions. At the time of discovery, the rice plants were showing silver streaks typical
of the damage caused by Stenchaetothrips biformis.
Prior to this observation, monitoring of other seedling pests Hydrellia sp. (Ephy-
dridae: Diptera) and Sogatodes orizicola (Muir) (Delphacidae: Homoptera) had been
in progress so it is probable that the thrips were discovered soon after their arrival in
the area.
The field in which the thrips were collected was subsequently sprayed with insec-
ticide for control of Hydrellia and the fate of the thrips is unknown. Since the discov-
ery, two rice crops have been planted. No more thrips have been found during
monitoring exercises.
The pest complex at the Caroni Rice Department at present includes Hydrellia sp.,
Tagosodes spp., Diatraea saccharalis (F.) (Pyralidae: Lepidoptera), Rupela albinella
(Cram.) (Pyralidae: Lepidoptera), and Oebalus poecilus (Dall.) (Pentatomidae: Het-
Stenchaetothrips biformis may be controlled by flooding. Should the species be-
come economically important their presence may compromise the current practice of
draining field for control of Hydrellia sp.
This record follows the recent arrival of Stenchaetothrips biformis in Guyana in
July-August 1994 (Munroe 1995), and Venezuela, near Calabozo, January-February
1995 (Cermeli et al. 1995). Calabozo is roughly 700 km west-southwest of Trinidad.


Specimens of Rice Thrips, Stenchaetothrips biformis Bagnall (Thysanoptera:
Thripidae) was collected on seedling rice in Trinidad, West Indies. This constitutes the
first record of rice thrips in Trinidad.


CERMELI, M., E. GARCIA, AND M. GAMBOA. 1995: Stenchaetothrips biformis (Bagnall)
(Thysanoptera: Thripidae) nueva plaga del arroz (Oryza sativa L.) en Venezu-
ela. Boletin de Entomologia Venezolana N.S. 10(2):209-210.
MUNROE, L. 1995. A new pest in Guyana's rice fields. Caraphin News No. 11 pp. 1-2.

Caroni (1975) Ltd. Is a state owned agroprocessing company which is the largest producer of
rice in Trinidad.

Scientific Notes


1University of Arkansas, Dept. of Entomology
321 Agri Building, Fayetteville AR 72701

2University of Arkansas, Rice Research and Extension Center
PO Box 351, Stuttgart AR 72160

The rice water weevil, Lissorhoptrus oryzophilus Kuschel (Coleoptera: Curculion-
idae), is a pest of rice in the southeastern U.S. and California. Although adult weevils
feed on rice leaves causing longitudinal scars, larvae feed on the roots causing eco-
nomic injury by reducing yield. Carbofuran (Furadan, FMC Corp., Philadelphia, PA)
has been used to manage rice water weevil larvae since the late 1960's. However, car-
bofuran is no longer legal for use in rice. The new alternative insecticides, lambda-cy-
halothrin (Karate, Zeneca Inc., Wilmington, DE) and diflubenzuron (Dimilin,
Uniroyal Chemical Co., Middlebury, CT) require application within 10 d after perma-
nent flood (Bernhardt 1997). Lambda-cyhalothrin is targeted against adult weevils,
and diflubenzuron is targeted against eggs. The two scouting methods used to deter-
mine the need for carbofuran application against larvae were the leaf scar method
and larval core sample (Tugwell & Stephen 1981, Morgan et al. 1989). Inspection for
leaf scarring from adult feeding is inadequate, and the inspection of rice roots for lar-
vae is too late in determining the need of an insecticide application with the newer in-
secticides. Therefore, a new scouting method is urgently needed.
A double-ended barrier trap was developed based on weevil swimming behavior
and tested on F, adult weevils. In a preliminary test on 22 July 1998, 16 double-ended
barrier traps were placed next to rice plants in a small late rice plantation (Hix et al.
1999). Barrier trap catch means of adult weevils were 73.9 (_9.4 SE) per trap on 23
July 1998 and 54.4 (_6.4 SE) per trap on 24 July 1998. The rice root core sample mean
for this bay was 72.9 (_7.0 SE) larvae per core on 1 August 1998. The Arkansas eco-
nomic threshold is 10 larvae per core.
In this note, we describe how to assemble this trap from the materials depicted in
Fig. la. When appropriate, English measurements are listed in parentheses.

Aluminum screen or aluminum flashing-35 cm x 11 cm
2 Boll weevil traps with collecting cups-Technical Precision Plastics,
Mebane, NC
4 Fishing bobbers-5 cm (2 in)
2 PVC pipes-10.5 cm x 2.1 cm (0.5 in schedule 40)
8 Plastic cable ties-10.2cm
20 gauge stainless steel wire
Dowel rod if aluminum screen is used 35 cm x 0.5 cm (3/16 in)

Trap construction begins by marking the aluminum screen or flashing 6.0 cm from
the corners on the long sides and 3.0 cm from the corners on the short sides. The cor-
ners are then trimmed off by cutting between these marks providing a barrier with
correct taper for insertion into the boll weevil traps. If screen is used for the barrier,
a dowel should be wired to the screen at 3.0 cm from the upper edge of the barrier with
stainless steel wire (Fig. la). This imparts rigidity to the trap allowing it to float prop-

Florida Entomologist 83(2)

June, 2000

Fig. 1. (a) Boll weevil tops with collecting cups, 5 cm (2 in) fishing bobbers, PVC
pipe, 10.2 cm plastic cable ties, screen (cut to specifications), and dowel rods with ar-
row indicating points to which they are attached to screen; (b) tabs (denoted by ar-
rows) in boll weevil traps may be used for proper screen alignment and trap strength.

early. If aluminum flashing is used for the barrier, 2 small holes should be drilled on
each of the 4 tapered edges. The barrier is inserted into each boll weevil trap using
plastic tabs (marked by arrows in Fig. ib) on the boll weevil traps for proper barrier
orientation and trap strength. The barrier is then wired to the boll weevil traps with
stainless steel wire. The float system for the trap is made by epoxying the bottom of
the fishing bobbers into each end of 2 PVC pipes. The top holes of the bobbers are
sealed with epoxy to make them water tight. Hot glue could be used for the float as-

Scientific Notes

Fig. 2. (a) Cable ties are looped around the PVC (denoted by double arrows) pipe
of the float assembly and cable ties previously attached to the trap braces on the up-
per side of the trap assembly, the single arrow indicates space through which a flag
maybe placed to hold the trap in position; (b) placement and appearance of properly
placed trap in the field.

sembly but is inferior to epoxy for durability in the field. A plastic cable tie is attached
to each of the 4 braces on the upper side of the trap assembly. Plastic cable ties are
then looped around PVC pipe of the floats and through each cable tie previously at-
tached to the braces of the boll weevil traps (Fig. 2a). Traps are positioned in the field
via flags placed through the space in one of the trap ends (Fig. 2). The trap should be

Florida Entomologist 83(2)

June, 2000

suspended beneath the float assemblies allowing about 1.5 cm of the barrier to emerge
from the water.
Traps should be checked every 24 h or 36 h for weevils. Weevils can survive being
submerged for about 96 h, but frequent inspection of the traps will minimize a buildup
of debris and decomposition of other insects caught in these traps. This trap design
might be useful in collecting other species of aquatic weevils and small to medium
sized aquatic insects.


The construction of rice water weevil barrier traps is described. These traps are
constructed by attaching a barrier between boll weevil traps and suspending them in
the water by flotation devices. Floating barrier traps were designed to monitor adult
rice water weevils from the first day of permanent flood until ten days post flood.


This project was funded in part by the University of Arkansas Experiment Station
and the Arkansas Rice Promotion Board.


BERNHARDT, J. L. 1997. Control of the rice water weevil with Karate. Arthropod Man-
agement Tests. 22: 289.
Trapping adult rice water weevils with floating cone and barrier traps, in The
B. R. Wells Rice Research Studies 1998. Arkansas Agri. Exp. Stat. Res. Series
468: 135-141.
MORGAN, D. R., N. P. TUGWELL, AND J. L. BERNHARDT. 1989. Early rice field drainage
for control of rice water weevil (Coleoptera: Curculionidae) and evaluation of an
action threshold based upon leaf-feeding scars of adults. J. Econ. Entomol. 82:
TUGWELL, N. P., AND F. M. STEPHEN. 1981. Rice water weevil Lissorhoptrus oryzophi-
lus seasonal abundance, economic levels, and sequential sampling plans. Ar-
kansas Agri. Exp. Stat. Bul. 849.

Scientific Notes


USDA, ARS, 5230 Konnowac Pass Rd., Wapato, WA 98951, USA

1Current address: Department of Biology, Oral Roberts University, Tulsa, OK 74171

Alarm pheromones have been demonstrated for a number of species of social Vesp-
idae including several hornets and yellowjackets (Vespines) (Landolt et al. 1997).
Maschwitz (1964a, b) first demonstrated alarm pheromone responses in the yellow-
jackets Vespula vulgaris L. and V germanica (Fab.) in response to crushed wasps and
body parts. Pheromone-mediated alarm has since been observed in other vespines:
Dolichovespula saxonica (Fab.) (Maschwitz 1984), the southern yellowjacket V squa-
mosa (Drury) (Landolt & Heath 1987, Landolt et al. 1999), the eastern yellowjacket
V maculifrons (Buysson) (Landolt et al. 1995), Provespa anomala Saussure
(Maschwitz & Hanel 1988), and Vespa crabro L. (Veith et al. 1984). 2-Methyl-3-
butene-2-ol was identified as a component of the alarm pheromone of V crabro (Veith
et al. 1984), and N-3- methylbutylacetamide was isolated and identified as an alarm
pheromone of the southern and eastern yellowjackets (Heath & Landolt 1988,
Landolt et al. 1995).
The source of alarm pheromones in social wasps generally is the venom, although
the head is implicated as an additional source of alarm pheromone for V vulgaris (Al-
diss 1983) and V squamosa (Landolt et al. 1999). Alarm behavior in V germanica and
V vulgaris occurred in response to the squashed sting apparatus, sting sac, and sol-
vent extract of the sting sac (Maschwitz 1964b) and in D. saxonica as a response to
crushed venom glands (Maschwitz 1984). Veith et al. (1984) stimulated alarm in V
crabro with squashed venom sacs or venom. Landolt & Heath (1987) isolated an
alarm pheromone of V. squamosa in solvent extracts of the venom sac and glands. Al-
diss (1983) observed alarm in V vulgaris in response to crushed conspecific heads, and
Landolt et al. (1999) stimulated alarm and attack in the southern yellowjacket with
a solvent extract of conspecific heads. Alarm pheromones known in several species of
Polistes also originate in the venom (reviewed by Landolt et al. 1997).
Despite repeated demonstrations of alarm responses of social wasps to conspecific
body parts and extracts of body parts, the alarm signalling process itself remains un-
known. We do not know how wasps release alarm pheromone. It is hypothesized that
alarm pheromone in venom is released when wasps spray venom or is deposited when
wasps sting (Aldiss 1983, Maschwitz 1964b, Greene et al. 1976). An alarm pheromone
originating in the head of workers may be released at the mouthparts and applied or
evaporated from the mandibles (Landolt et al. 1999).
We report here experimental evidence that an alarm pheromone is deposited on a
substrate or target when attacked by southern yellowjackets. We also demonstrated
persistence of that alarm pheromone activity that is uncharacteristic of alarm phero-
mones generally. Alarm pheromones in social insects typically are quite volatile and
short-lasting, an advantage in permitting normal colony activities to resume once a
threat has passed (Matthews & Matthews 1978).
Preliminary observations that led to this study indicated possible contamination
of protective clothing and equipment following attacks by southern yellowjackets.
This included seemingly unprovoked responses by yellowjackets to investigators one
or more days following other experiments and a residual odor on material and objects

Florida Entomologist 83(2)

June, 2000

that had been attacked by workers. We conducted experiments to determine if attack-
ing wasps leave a material that elicits alarm and attack in other workers.
Observations and experiments with southern yellowjackets were made in Alachua
County, Florida. All testing was done with vigorous underground colonies. The bioassay
for these tests involved a cork (3.7 cm x 3.7 cm) connected to a wooden dowel (3 m long
by 1.2 cm diam.) with 2 interlocking eye hooks screwed into the dowel and cork. The
eye hooks permitted movement of the cork on the dowel that made it easier to detect
wasp contact with the target. Three colored push pins (red, blue, and green) were
stuck into the bottom of the cork. This target (cork with pins) was waved from side to
side about 0.3 m in front of a colony entrance to induce attack from guard workers
present in the nest entrance. During these presentations, workers generally attacked
the cork as well as the hooks and push pins. During attacks workers made sting
thrusts and also appeared to bite the target, with their mandibles open and contact-
ing the cork, hooks, or pins.
An experiment was conducted to determine if freshly attacked targets elicit alarm,
as evidence of the deposition of alarm pheromone by wasps onto targets during earlier
attacks. Corks were first presented at nest entrances until hit by wasps, with 6-10
wasps contacting the cork. This treated cork was placed in a glass jar in an ice chest
and transported to the laboratory and placed in a freezer. This procedure was re-
peated to accumulate 5 treated corks. A new dowel and cork was used for each repli-
cate of this procedure. A treated cork was subsequently exposed to ambient field
conditions for 3 min and was then presented to a second test colony. The cork was
moved slowly to 1/3 m upwind of the colony where it remained for the assay duration.
Alarm behavior and hits to the cork were noted for 2 min, with the use of a tape re-
corder. As a control, an unexposed cork and dowel were presented in the same manner
before each assay of a treated cork. Five treated corks were each tested 4 5 h apart
over several days. Numbers of hits and landings on the five treated corks ranged from
1 to 76 (mean + SE = 33.4 + 30.4), significantly greater than the no hits or landings
that occurred on the five control corks (p = 0.036 by Student's t test).
A second experiment was conducted to determine if the alarm activity of material
applied by yellowjackets to the cork remained active after exposure at ambient tem-
peratures. The bioassay protocol was the same as above except that each treated cork
was aired in the laboratory for 15 h before field tests. The five treated corks were
tested over 4 different days, with 2 tested 2.5 h apart on one day. Again, the five con-
trol corks elicited no response, with no evidence of alarm and no contacts of yellow-
jackets with the corks. The 15 h old corks however, elicited alarm and attack to the
corks. Numbers of hits and landings ranged from 4 to 83 per assay (mean + SE = 35
+ 33.9), significantly greater than the response to the control (t = 2.33, p = 0.04).
These two experiments demonstrate that alarmed V squamosa apply chemicals to
targets that they attack and that such attacked targets may subsequently remain ac-
tive in eliciting alarm and attack responses from southern yellowjackets. At this time,
the source of the pheromone applied is not known. During stinging attacks on the
corks, hooks, and pins, venom could have been applied to those surfaces. Also, wasps
attacking the targets were seen to bite on the cork, hooks, and pins, with the possibil-
ity that other alarm pheromones from gnathal or cephalic glands which open to the
mouthparts (Landolt & Akre 1979) could be applied to mark the targets. The southern
yellowjacket is known to possess alarm pheromone both in the venom (Landolt &
Heath 1987, Heath & Landolt 1988) and in the head (Landolt et al. 1999).
The possible adaptive significance of such a long lasting alarm signal is apparent
when considering the functions of sting autotomy in social insects, the lack of sting
autotomy in yellowjackets, and the nature of predators of social wasps and bees. Sting

Scientific Notes

autotomy is the loss of the sting and venom sac after a stinging episode, such as occurs
in the honey bee,Apis mellifera L. (Free 1987) and in the social wasp Polybioides raph-
igastra (Saussure) (Sledge et al. 1999). In both of these species, sting autotomy prob-
ably permits the prolonged release of alarm pheromone from the sting apparatus
following stinging, marking the intruder and focusing subsequent attacks (Free 1987,
Sledge et al. 1999). A similar strategy has been suggested for species ofApis (Pickett
et al. 1982, Schmidt et al. 1997), includingA. mellifera and Apis cerana (Fab.), based
on the large quantities of eicosenol in the sting apparatus. It is hypothesized that this
compound serves as a carrier to prolong the release of more volatile pheromone com-
pounds and to mark an intruder to focus the defending bees. Although the southern
yellowjacket does not exhibit sting autotomy, the deposition of a long lasting alarm
pheromone during stinging attacks on vertebrate predators may similarly serve to
chemically mark the animal. This would focus attacks on the intruder and also alert
the colony if and when this predator approached the nest again.
Several vertebrate predators, such as skunks and raccoons, commonly prey on yel-
lowjacket colonies (Akre & Reed 1984). A possible strategy of a vertebrate predator is
exemplified by the honey buzzard, a successful nest predator of European yellowjack-
ets (Cobb 1979, cited in Akre & Reed 1984). This bird was persistent in its attacks on
excavated subterranean Vespula nests over a period of days. The buzzard was driven
away from nests repeatedly by wasps, but later reapproached the nest to continue the
attack. Such an "attack, retreat, reattack" scenario is a likely strategy for other ver-
tebrate nest predators. Thus, a long lasting alarm pheromone that marks a predator
may be highly advantageous to yellowjackets. However, caution must be exercised in
interpreting how these results of alarm pheromone persistence from a wooden cork
may relate to alarm pheromone on vertebrate skin, fur, or feathers. Additional exper-
imentation with leather or feather targets could help address this question, as would
chemical analysis of odors emitted by attacked objects.


Alarmed southern yellowjacket workers attacking corks placed near colony en-
trances applied an alarm pheromone that stimulated alarm and attack behavior in
another colony 3 min or 15 h after pheromone deposition. Observations of wasps at-
tacking corks indicated deposition or application of alarm pheromone could be made
both from the sting and from the mandibles. This long lasting material may serve to
mark an attacking vertebrate predator so that it is quickly detected and attacked
again upon its return to a wasp colony.


AKRE, R. D. AND H. C. REED. 1984. Vespine defense, 3, pp. 55-94, in H. R. Hermann,
(ed.), Defensive mechanisms in social insects. Praeger, New York, 259 pp.
ALDISS, J. B. J. F. 1983. Chemical communication in British social wasps (Hy-
menoptera: Vespidae). Ph.D. Dissertation, Univ. Southampton, U.K. 252 pp.
COBB, F. K. 1979. Honey buzzard at wasps' nest. Brit. Birds 72: 59-64.
FREE, J. B. 1987. Pheromones of social bees. Cornell Univ. Press, Ithaca, New York.
218 pp.
GREENE, A., R. D. AKRE, AND P. J. LANDOLT. 1976. The aerial yellowjacket, Dolicho-
vespula arenaria (Fab.): Nesting biology, reproductive production, and behavior
(Hymenoptera: Vespidae). Melanderia 26: 1-34.
HEATH, R. R., AND P. J. LANDOLT. 1988. The isolation, identification, and synthesis of
the alarm pheromone of Vespula squamosa (Drury) (Hymenoptera: Vespidae)
and associated behavior. Experientia 44: 82-83.

Florida Entomologist 83(2)

June, 2000

LANDOLT, P. J., AND R. D. AKRE. 1979. Occurrence and location of exocrine glands in
some social Vespidae (Hymenoptera). Ann. Entomol. Soc. America 72: 141-148.
LANDOLT, P. J., AND R. R. HEATH. 1987. Alarm pheromone behavior of Vespula squa-
mosa (Hymenoptera: Vespidae). Florida Entomol. 70: 222-225.
LANDOLT, P. J., R. R. HEATH, H. C. REED, AND K. MANNING. 1995. Pheromonal medi-
ation of alarm in the eastern yellowjacket (Hymenoptera: Vespidae). Florida
Entomol. 78: 101-108.
LANDOLT, P. J., R. L. JEANNE, AND H. C. REED. 1997. Chemical communication in so-
cial wasps, pp. 216-235, in R. K. Vandermeer, M. Breed, M. Winston, and K. Es-
pelie (eds.), Pheromone communication in social insects: ants, wasps, bees, and
termites. Westview Press, Boulder, CO.
LANDOLT, P. J., H. C. REED, AND R. R. HEATH. 1999. An alarm pheromone from the
heads of worker southern yellowjackets, Vespula squamosa (Drury) (Hy-
menoptera: Vespidae). Florida Entomol. 82: 356-359.
MASCHWITZ, U. W. 1964a. Alam substance and alarm behavior in social Hymenoptera.
Nature 204: 324-327.
MASCHWITZ, U. W. 1964b. Gefahrenalarmstoffe und Gefahrenalarmierung bei so-
zialen Hymenopteren. Z. Verg. Physiol. 47: 596-655.
MASCHWITZ, U. W. 1984. Alarm pheromone in the long cheeked wasp Dolichovespula
saxonica (Hymenoptera: Vespidae). Deutsch Entomol. 31: 33-34.
MASCHWITZ, U.W., AND H. HANEL. 1988. Biology of the southeast Asian nocturnal wasp,
Provespa anomala (Hymenoptera: Vespidae). Entomol. Generalis 14: 47-52.
MATTHEWS, R. W., AND J. R. MATTHEWS. 1978. Insect behavior. John Wiley and Sons,
NY, 507 pp.
PICKETT, J. A., I. H. WILLIAMS, AND A. P. MARTIN. 1982. (Z)-11-Eicosen-l-ol, an im-
portant new pheromonal component from the sting of the honeybee Apis mel-
lifera L. (Hymenoptera: Apidae). J. Chem. Ecol. 8: 163-175.
1997. (Z)-11-Eicosen-1-ol, a major component ofApis cerana venom. J. Chem.
Ecol. 23: 1929-1939.
duces alarm behavior in the social wasp Polybioides raphigastra
(Hymenoptera: Vespidae): an investigation of alarm behavior, venom volatiles
and sting autotomy. Physiol. Entomol. 24: 234-239.
VEITH, J., N. KOENIGER, AND U. W. MASCHWITZ. 1984. 2-methyl-3-butene-2-ol, a ma-
jor component of the alarm pheromone of the hornet, Vespa crabro. Naturwis-
senschaften 71: 328-329.

Scientific Notes


Department of Biology, McMurry University, Abilene, Texas 79697

Tardigrades are animals currently classified in the phylum Tardigrada and are
phylogenetically most closely related to the arthropods (Aguinaldo et al. 1997). They
live in a variety of microhabitats, being active when a film of moisture is present and
becoming dormant when dry. Their small size (mean length 300-350 m) allows many
of the species to live relatively undisturbed on the surface of a variety of plants and
in the soil. The anterior end of the complete digestive system is equipped with sharp
stylets and a muscular pharynx which allows a piercing-sucking method of feeding.
Some species have a large diameter buccal tube which permits ingestion of solid par-
ticles up to the size of fungal spores. Bacteria have often been observed on the surface
of tardigrades and are ingested by a number of tardigrade genera (Kinchin 1994). It
has been assumed that these bacteria are food, although symbiosis has been sug-
gested (Kinchin 1989, 1993).
During a previous study of lichen-dwelling species of tardigrades, we discovered
that up to half of the individual tardigrades examined harbored members of the phy-
topathogenic bacteria Xanthomonas and Pseudomonas (Krantz et al. 1999). Evidence
developed in that study also suggested that these bacteria dwell inside the tardigrade,
perhaps having originated from detritus or food. Furthermore, the bacteria appear to
be shed as the animal moves about. Tardigrades harboring phytopathogenic bacteria
might transport those bacteria to suitable plant hosts where they could then cause
disease. Experimental evidence reported here strengthens this interpretation by
showing that in the laboratory tardigrades can acquire and transfer Xanthomonas
campestris pv. raphani Lincoln. Infections can be produced in healthy radish plants
from these transferred bacteria.
Xanthomonas campestris pv. raphani ATCC 49079 was obtained from The Ameri-
can Type Culture Collection, Rockville, MD. Individual tardigardes, Macrobiotus
hufelandi Schultze (Parachela: Macrobiotidae), were collected from samples of lichen
growing on trees along the shore of a fresh water lake. The samples were soaked in
water for 3.5 h during which the dormant forms of the tardigrades became active. The
tardigrades were concentrated by filtering the water containing them through 90 or
120 pm mesh screens. They were immediately collected from the screens, rinsed 3
times in tap water that had been passed through a sterilizing nylon membrane filter,
and then used in transmission experiments.
Xanthomonas campestris pv. raphani was grown in TGCP broth (per liter: yeast
extract, 2.5 gm; glucose, 20 gm; peptone, 2.5 gm; NaCl, 1 gm; K2HPO4, 1 gm; MgSO,@7
H20, 0.5 gm; CaCO,, 10 gm) or on nutrient agar (Difco, Sparks, MD) at 30C. Radish,
Raphanus sativus L. ('Cherry Blossom' seed BWI Bulk Seeds, Inc., U.S.A.), were
washed with 70% (v:v) isopropanol before planting in sterile 30 x 290 mm tubes filled
to 1/3 of their volume with moist, sterile potting soil and fitted with sterile sponge
plugs. The tubes were incubated at 30C in growth chambers on a continuous 24 h
light cycle.
Tardigrades were inoculated with X. campestris pv. raphani by adding bacteria
from a 24 h-old culture to a concentration of approximately 1 x 106 cells / ml of water
in which some lichen samples were soaked for 3.5 hours. Confirmation of inoculation
was obtained by previously described methods (Krantz et al. 1999) as follows. Tardi-

Florida Entomologist 83(2)

June, 2000

grades were rinsed 3 times in 50 pl droplets of sterile tap water by using a small wire
(Irwin) loop to capture the tardigrades and transfer them from droplet to droplet.
They then were placed on the surface of a nutrient agar plate, in the center of a circle
(2 cm in diameter) which was marked on the undersurface of the petri dish. The plates
were incubated for 5 days at 30C during which the tardigrades freely moved across
the surface of the agar. Microorganisms shed by the tardigrades developed into mac-
roscopic colonies during this period. Smooth, round, lemon-yellow colonies typical of
X. campestris pv. raphani which arose completely outside of the circle were used to in-
oculate leaves of radish seedlings by first dispersing a single colony in 5 ml of sterile
water and subsequently placing 50 pL of this suspension on the leaf of a radish plant.
Isolates which produced leaf spot within 7 days were scored as beingX. campestris pv.
raphani originating from inoculation of the tardigrade during soaking of the lichens.
Uninoculated tardigrades were tested this way also to determine ifX. campestris pv.
raphani was associated naturally with the tardigrade population.
To determine if M. hufelandi could be inoculated with X. campestris pv. raphani
from the lesions on infected radish leaves, single infected leaves were removed asep-
tically from separate plants and placed topside-up on separate nutrient agar plates.
A single rinsed tardigrade suspended in 50 pl of sterile water was deposited with a mi-
cropipet onto the leaf spot lesion of each leaf. The plates were incubated for 48 h at
30C and scored for the presence of X. campestris pv. raphani colonies on the agar
away from the leaf. The presence of these colonies would indicate that the bacteria
had been transferred off the leaf by the tardigrade. As controls, infected leaves with-
out tardigrades and uninfected leaves with tardigrades also were tested.
To test whether tardigrades harboring X. campestris pv. raphani could facilitate
an infection on healthy radish plants, 25 individual radish seedlings were inoculated
either with a single inoculated or uninoculated tardigrade prepared as described
above. The plants were incubated as described above for up to two weeks and were
scored for the appearance of bacterial leaf spot on the leaf where the tardigrade had
been placed.
Every attempt to inoculate M. hufelandi with X. campestris pv. raphani was suc-
cessful. In the 5 replicate experiments in which tardigrades were suspended with
these bacteria in water, and subsequently rinsed and placed on nutrient agar plates,
17 to 23 suspected colonies of X. campestris pv. raphani arose per plate. All of these
colonies were used successfully to produce leaf spot disease in healthy radish plants,
confirming that they represented the pathovar used to inoculate the tardigrades and
not naturally-occurring microbiota. Tardigrades which were not inoculated with X.
campestris pv. raphani did not produce colonies capable of causing disease in radish
plants. Plates inoculated with the last drops of sterile water used to rinse the tardi-
grades produced no colonies.
In all 5 replicates in which uninoculated tardigrades were introduced onto dis-
eased radish leaves lying on the surface of nutrient agar plates, and subsequently
were allowed to move freely across the surfaces of the plates, 15 to 24 colonies of the
pathogen arose per plate, indicating that tardigrades can acquire and move X.
campestris pv. raphani from diseased leaves. Plates in similar experiments using un-
infected, non-diseased leaves produced no colonies of X. campestris pv. raphani nor
did plates containing only diseased leaves without tardigrades.
In all 10 replicates in which inoculated tardigrades were introduced onto the leaves
of healthy radish plants, the radish plants displayed symptoms of leaf spot disease
within 2 weeks. The symptoms were identical to those observed when a washed pure sus-
pension ofX. campestris pv. raphani only was used to inoculate radish leaves. Inoculated
tardigrades therefore were able to shed viable X. campestris pv. raphani which then

Scientific Notes

caused an infection on the leaves. Leaves treated with uninoculated tardigrades or cell-
free spent TCGP broth fromX. campestris pv. raphani cultures did not become diseased.
Tardigrades were easily inoculated with X. campestris pv. raphani. No failures to
inoculate were observed in the experiments. Although it is uncertain where the inoc-
ulated bacterial cells resided, either in or on the animal, they appeared to associate
quickly and tightly with tardigrades. The xanthan exopolysaccharide produced byX.
campestris pv. raphani may have mediated attachment of the bacteria to the tardi-
grade exoskeleton. Or, the tardigrades may have ingested the bacteria while changing
from the dormant to active form during the soaking period. Because the tardigrades
were not washed free of the attachedXcampestris pv. raphani it is possible that a spe-
cialized relationship exists between them. Even when tardigrades were placed on the
lesions of infected leaves, the pathogen associated with the tardigrades at least well
enough to be transported off of the leaves and across the surface of nutrient agar
plates. Perhaps in nature tardigrades can transport Xanthomonas spp. through the
environment, and perhaps even from plant-to-plant, since tardigrades are mobile and
their small size also would allow them to become airborne. Xanthomonas spp. have
been reported to travel through the environment and possibly from plant-to-plant by
a wide variety of means including insects and small airborne particles (Bashan 1985).
The experiments in which inoculated tardigrades were placed on uninfected
plants showed that tardigrades can deposit Xanthomonas on a susceptible host, which
then can become infected. It is unknown whether the pathogen was injected into the
leaves during feeding by the tardigrades or merely deposited on the surface during
other activities, or some combination of the two. However, every plant that received a
single, infected tardigrade developed bacterial leaf spot, indicating that the bacteria
remained viable and infectious during their association with the tardigrades and that
the animals effectively transported the pathogens. The results as a whole suggest that
some cases of X. campestris infection in nature may be produced by bacterial cells
shed from tardigrades.


Tardigrades were inoculated with X. campestris pv. raphani by soaking in a sus-
pension of the bacteria or by contact with leaf spot lesions. Bacteria shed from these
tardigrades caused leaf spot disease in radish plants.


dence for a clade of nematodes, arthropods and other moulting animals. Nature
387: 489-493.
BASHAN, Y. 1985. Field dispersal of Pseudomonas syringae pv. tomato, Xanthomonas
campestris pv. vesicatoria, and Alternaria macrospora by animals, people,
birds, insects, mites, agricultural tools, aircraft, soil particles, and water
sources. Canadian J. Bot. 64: 276-281.
KINCHIN, I. M. 1989. Hypsibius anomalus Ramazzotti (Tardigrada) from gutter sedi-
ment. Microscopy 36: 240-244.
KINCHIN, I. M. 1993. An observation on the body cavity cells of Ramazzottius (Hyps-
ibiidae, Eutardigrada). Quekett J. Microscopy 37: 52-55.
KINCHIN, I. M. 1994. The biology of tardigrades. Portland Press, London. 186 pp.
KRANTZ, S. L., T. G. BENOIT, AND C. W. BEASLEY. 1999. Phytopathogenic bacteria as-
sociated with Tardigrada. Zool. Anz. 238: pp. 259-260.

Florida Entomologist 83(2)

June, 2000


1Department of Entomology, University of California
Riverside, California 92521, USA

2University of California Cooperative Extension, Ventura County
Ventura, California 93003, USA

The glassy-winged sharpshooter, Homalodisca coagulata (Say), is primarily a south-
eastern US species, but due to an apparent accidental introduction it has recently be-
come a problem pest of crops and ornamentals in southern California (Blua et al. 1999);
its spread into central and northern California is quite likely. Triapitsyn et al. (1998) re-
ported results of the 1997 survey of H. coagulata egg parasitoids in Florida and Louisi-
ana, where it is native, and also in southern California. In Monticello, Florida, its main
egg parasitoids were the mymarid wasp Gonatocerus ashmeadi Girault throughout the
season, and the trichogrammatid wasp Zagella sp. during July and August. This tri-
chogrammatid was later found (S. V. Triapitsyn, unpublished data, and J. D. Pinto, per-
sonal communication) to be different from the closely related species Zagella floridae
Viggiani, described from a female and a male collected fromjasmine in Fort Lauderdale,
Florida (Viggiani 1985), thus representing a new, undescribed species.
Despite the fact that parasitization of H. coagulata eggs by G. ashmeadi may at
times reach 80% in Florida and Louisiana, and up to 100% in California (Phillips
1998, Triapitsyn et al. 1998), it has become clear that the large numbers of H. coagu-
lata in southern California are due in part to poor natural control of the first, early
spring, generation of the sharpshooter.
In an effort to find a more effective natural enemy of the spring brood, we visited
Texas (Weslaco), and the state of Tamaulipas, Mexico, in late April 1999. Both states
have never been surveyed for H. coagulata natural enemies, yet this leafhopper is
known to occur there (Young 1958, Turner & Pollard 1959). Young (1958, 1968) re-
ported this species from Mexico, without providing further details about its distribu-
tion. We examined a specimen of this species from Llera de Canales, which is in the
tropical part of Tamaulipas. It is the most southern record of H. coagulata known in
North America, based on the published locality data and according to our assessment
of H. coagulata distribution (Fig. 1). This leafhopper is abundant from eastern Texas
to southern Georgia and northern Florida, but its populations in central Texas (west
of Brown and Kerr Counties) and in central Florida decrease considerably (Turner &
Pollard 1959). It appears to be rare also in northeastern Mexico, according to our own
observations and the study of the leafhopper collection of the Universidad Aut6noma
de Tamaulipas in Ciudad Victoria.
Homalodisca insolita (Walker) and several Oncometopia species are the prevalent
sharpshooters in southern Florida, Mexico, and central America (Young 1968), although
H. insolita not long ago invaded the traditional areas of H. coagulata distribution
(Turner & Pollard 1959). The sharpshooters Oncometopia clarior (Walker) and 0. sp.
near nigricans (Walker) occur on citrus in both Nuevo Le6n and Tamaulipas, Mexico (see
"Material Examined"). In Weslaco, Texas, we could not find any H. coagulata on citrus,
but we saw last year's egg masses, characteristic of this sharpshooter, with parasitoid
exit holes on leaves of a mescal bean tree, Sophora secundiflora (Ortega) de Candolle.

Scientific Notes

*Boundary Population

Fig. 1. Homalodisca coagulata distribution map in North America.

In parks and citrus groves of central Tamaulipas, we were not able to find any
adult or nymph stages of H. coagulata despite the fact that we saw many last year's
egg masses characteristic of the glassy-winged sharpshooter, all of them with evi-
dence of parasitization. Only on one occasion did we collect H. coagulata adults and
nymphs and also found its parasitized eggs both on citrus and peach, grown in the
shade of larger trees in a private garden near Valle Hermoso, in Rio Bravo del Norte
Valley (northern Tamaulipas).
In the laboratory of the Universidad Aut6noma de Tamaulipas, the mymarid
wasps Gonatocerus triguttatus Girault taxonomicc determination by S. V. Triapitsyn)
emerged from H. coagulata egg masses collected in Valle Hermoso, thus providing us
with the first known host record for G. triguttatus. Interestingly, the type series of this
species was reared from an egg-mass of a leafhopper on orange in Trinidad (Girault
1916). Gonatocerus triguttatus was redescribed and illustrated by Huber (1988), who
also indicated its distribution in Texas (Cameron, Hidalgo, and Val Verde Counties)
and its presence in several states in Mexico, without, however, giving details of the
Mexican material; these are provided below under "Material Examined". This species
is probably widely distributed throughout Central America and parts of South Amer-
ica (first author has seen unidentified specimens from Brazil, Costa Rica, Guatemala,
Guyana, and Mexico that may be conspecific with G. triguttatus).
We would expect that G. triguttatus would also attack other sharpshooter species
in the genera Homalodisca and Oncometopia. Our observations in the citrus groves in
Tamaulipas indicate that the early spring populations of the sharpshooters H. coagu-
lata, 0. clarior, and 0. sp. near nigricans are under good natural control by egg para-
We believe that, subject to obtaining proper permits, prompt introduction of G.
triguttatus into southern California from southeastern Texas or northeastern Mexico
is warranted. Finding live G. triguttatus beyond H. coagulata's range would be very
difficult because its host associations there are unknown. Other obvious candidates

Florida Entomologist 83(2)

June, 2000

for importation into California are Gonatocerus fasciatus Girault from Louisiana and
Zagella sp. from Florida (Triapitsyn et al. 1998). If established, these species may en-
hance the overall natural control of H. coagulata in southern California.
Material Examined. Gonatocerus triguttatus: MEXICO: Baja California Sur: Las
Barracas (ca. 30 km E. of Santiago), 18-IV-1984, P. DeBach, 11 females; 10 km N. of
La Paz, 28-X-1983, J. D. Pinto, 2 females. Nuevo Le6n: El Carmen, 10-VII-1983, A.
Gonzalez H., 2 females; Hacienda El Canada, 12-VII-1983, A. Gonzalez H., G. Gordh,
M. A. Rodriguez P., 2 females; Linares, 23-X-1961, H. Suarez, 1 female (on citrus); San
Juan, 14-VII-1983, A. Gonzalez H., F. Reyes V., 2 females [CNCI, Canadian National
Collection of Insects, and UCRC, University of California at Riverside, det. J. Huber].
Tamaulipas, nr. Valle Hermoso, 22-IV-1999, S. Triapitsyn & P. Phillips: 6 females (ex.
H. coagulata eggs on peach); 18 females, 3 males (ex. H. coagulata eggs on citrus).
Homalodisca coagulata: MEXICO, Tamaulipas: Llera de Canales, 20-VIII-1994, D.
Covarrubias, 1 female (on lemon); nr. Valle Hermoso, 22-IV-1999, S. Triapitsyn & P.
Phillips, 1 female, 2 males. Oncometopia clarior: MEXICO, Tamaulipas: Ciudad Vic-
toria, 28-VI-1997, Hernandez & Villegas, 1 female; Llera de Canales, 21-VI-1997,
Mtz., Monrreal & Teran, 1 male (on orange). Oncometopia sp. nr. nigricans: MEXICO:
Nuevo Le6n, Montemorelos, 24-IV-1999, S. Triapitsyn & P. Phillips, 1 male (on citrus).
Tamaulipas, nr. Santander Jim6nez, 22-IV-1999, S. Triapitsyn & P. Phillips, 1 male
(on Hibiscus sp.) [unless stated otherwise, all above specimens, including vouchers,
deposited in the collection of Universidad Aut6noma de Tamaulipas, Ciudad Victoria,
We thank Svetlana N. Myartseva for her help with parasitoid rearing, Enrique
Ruiz Cancino and Vladimir A. Trjapitzin for letting us use the facilities, arranging the
loan of specimens, and providing various assistance (all Universidad Aut6noma de
Tamaulipas, Ciudad Victoria, Mexico). Gordon Gordh (USDA-ARS, Weslaco, Texas) is
acknowledged for facilitating our survey efforts in Texas. We are indebted to Raymond
J. Gill (California Department of Food and Agriculture, Sacramento) for the sharp-
shooter identifications. Robert W. Brooks (Snow Collections, Natural History Mu-
seum, University of Kansas, Lawrence) provided data about the geographical
distribution of H. coagulata in Texas and John T. Huber (CNCI) pointed at additional
records of G. triguttatus in Mexico.


A survey of egg parasitoids of the glassy-winged sharpshooter, Homalodisca coag-
ulata (Say), was conducted in Tamaulipas, Mexico and Texas, USA in April 1999. The
mymarid Gonatocerus triguttatus Girault was reared from egg masses of this leafhop-
per on citrus and peach in northern Tamaulipas. This discovery is the first known host
record of G. triguttatus; its other, probable, host associations are indicated. This wasp,
whose geographical distribution also includes the eastern Rio Grande basin in Texas
and the states of Baja California Sur and Nuevo Le6n, Mexico, is identified as a po-
tential biological control agent for introduction into California against H. coagulata.


BLUA, M. J., P. A. PHILLIPS, AND R. A. REDAK. 1999. A new sharpshooter threatens
both crops and ornamentals. California Agric. 53(2): 22-25.
GIRAULT, A. A. 1916. New miscellaneous chalcidoid Hymenoptera with notes on de-
scribed species. Ann. Entomol. Soc. America 9: 291-308.
HUBER, J. T. 1988. The species groups of Gonatocerus Nees in North America with a
revision of the sulphuripes and ater groups (Hymenoptera: Mymaridae). Mem.
Entomol. Soc. Canada 141: 1-109.

Scientific Notes 203

PHILLIPS, P. A. 1998. The glassy-winged sharpshooter: a potential threat to California
citrus. Citrograph 83(12): 10-12.
asitoids of Homalodisca coagulata (Homoptera: Cicadellidae). Florida Ento-
mol. 8 (2): 241-243.
TURNER, W. F., AND H. N. POLLARD. 1959. Life histories and behavior of five insect
vectors of phony peach disease. Tech. Bull. United States Dep. Agric. 1188, 28
VIGGIANI, G. 1985. A new species of Zagella (Hym. Trichogrammatidae) from Florida.
Boll. Lab. Entomol. Agr. "Filippo Silvestri", Portici 42: 15-17.
YOUNG, D. A. 1958. A synopsis of the species of Homalodisca in the United States.
Bull. Brooklyn Entomol. Soc. 53(1): 7-13.
YOUNG, D. A. 1968. Taxonomic study of the Cicadellinae (Homoptera, Cicadellidae).
Part 1. Proconiini. United States Nat. Mus. Tech. Bull. 261, 287 pp.

Scientific Notes


'IRD (ex ORSTOM), Entomologie Agricole, B.P. 5045, 34032 Montpellier, France

2Embrapa-Meio Ambiente, C.P. 69, 13820-000 Jaguariina/SP, Brazil

Tetranychus ogmophallos Ferreira & Flechtmann is a new tetranychid found on
pinto peanut, Arachis pintoi (Krap. & Greg.) (Fabaceae), in an experimental field of
the Centro Nacional de Pesquisa Agropecudria do Cerrado (EMBRAPA) in Planal-
tina-DF, Brazil (Ferreira & Flechtmann 1997). Pinto peanut is a Brazilian legume
mainly used as a nitrogen fixator in pasture fields or as green manure in tropical crops
such as coffee, banana, and oil-palm (Grof 1984, De la Cruz et al. 1994, Suarez-
Vasquez et al. 1992). Because of its agronomic characteristics pinto peanut is cur-
rently exported from Brazil to other countries of Latin America and to Australia
(Asakawa & Ramirez 1989, Vilarreal & Chavez 1991, Cook et al. 1990). Thus, through
exports of pinto peanut T ogmophallos could be accidentally introduced into new ar-
eas, and is a quarantine issue.
The objective of this work was to compare the suitability of three legumes of eco-
nomic importance, common bean, Phaseolus vulgaris L., peanut, A. hypogeae L., and
soybean, Glycine max Merrill, as host plants of T ogmophallos. Studies were con-
ducted under laboratory conditions at 26 + 0.5C, 75 + 10% RH and a photoperiod of
13:11 (L:D). Experiments were carried out using the progeny of about 50 females from
the quarantine laboratory of CENARGEN, EMBRAPA, Brasilia. Mites were reared on
bean leaves in the same environmental conditions as above. To study survival rate,
developmental time, and sex-ratio, 4-5 females were placed on the lower surface of a
leaf disk (4 cm2) maintained on water-soaked cotton. After 1 h, females and excess
eggs were eliminated to obtain one egg per disk; 70 disks were kept for each treat-
ment. The disks were monitored three times a day (at 7 a.m., 1 p.m., and 7 p.m.) until
adult emergence. For oviposition study, one female teliochrysalis (last pre-imago in-
star) and two males were placed on a leaf disk and the males were removed 48 h after

Florida Entomologist 83(2)

June, 2000

the female had emerged. The number of eggs laid per female was monitored daily. The
disks were changed every 4 days. Thirty females were followed per legume. Demo-
graphic parameters (net reproductive rate, R0, generation time, G, and intrinsic rate
of natural increase, rm) were determined using a program developed by Hulting et al.
(1990). Biological parameters (survivorship, developmental time, fecundity, and lon-
gevity) were compared between treatments using a one-way ANOVA (LEAS 1989). If
ANOVA revealed significant differences, means were compared using the Scheffe
method. R0 and rm were compared using the SNK test (Hulting et al. 1990).
Development from egg to adult occurred on all three plant species tested. Develop-
mental time (Table 1) recorded on peanut was significantly different than those re-
corded on common bean and on soybean (F = 108.16; df = 2-144; P < 0.0001). On each
plant species, 85 + 5% of eggs reached adult stage, with no significant differences be-
tween treatments. The sex-ratio was 80% female for all 3 legumes.
Longevity in days for adult females differed significantly among host plant
(Table 1) (F = 21.7; df = 2-79; p < 0.05). The highest total fecundity was obtained on
common bean (F = 24.9; df = 2-79; p < 0.05), whereas no significant difference was
found between peanut and soybean.
The R0 calculated on common bean was significantly higher than R0 obtained on
soybean (q = 3.6, df = 3, 19; p < 0.05) and on peanut (q = 4.1, df = 3, 19; p < 0.05). No
significant difference was found between the latter two legumes (q = 0.5, df = 3, 19; p
> 0.05) (Table 1).
The rm values differed significantly between host plants. The highest rm was ob-
tained on common bean (q = 8 with soybean and 7.3 with peanut, df = 3, 19; p < 0.05)
and the lowest on peanut (q = 4.1 with soybean, df = 3, 19; p < 0.05) (Table 1).
Results of the present study indicate that T ogmophallos is not only able to develop
on common bean, soybean, and peanut, but displayed high rates of increase when
reared on these three plants. Values of the biological and demographic parameters
were in the range observed for other Tetranychus spp. under similar environmental
conditions (Gutierrez 1976, Carey & Bradley 1982, Tsai et al. 1989, Rai et al. 1995).
Because tetranychids are polyphagous and T ogmophallos showed high rates of in-
crease when reared on these three different plants, it seems reasonable to speculate
that T ogmophallos also could develop on a wide range of hosts. Thus, further inves-
tigations are needed in order to precisely characterize this range.


Arachis Phaseolus
Parameters hypogeae Glycine max vulgaris

Developmental time' 14.2 + 0.2 11.9 + 0.8 11.7 + 0.4
(egg to adult)
Total fecundity 60.0 2.9 63.9 5.7 104.3 7.8
Longevity of females' 16.5 0.8 15.4 1.2 25.3 1.1
Mean generation time'(G) 20.1 18.9 20.7
Net reproductive rate (Ro) 41.05 + 2.0 45.7 + 4.1 75.9 + 4.9
Intrinsic rate of increase (rm) 0.190 + 0.003 0.215 + 0.001 0.232 + 0.004

Means are followed by +SE.

Scientific Notes


The suitability of common bean, Phaseolus vulgaris, soybean, Glycine max, and
peanut, Arachis hypogeae, as food substrates for the mite Tetranychus ogmophallos
was evaluated. The mite performed better on common bean (rm = 0.232) although it de-
veloped and reproduced well on soybean (rm = 0.215) and peanut (rm = 0.190).


ASAKAWA, N. M., AND R. RAMIREZ. 1989. Inoculation and planting methodology for
Arachis pintoi. Pasturas Tropicales 11: 24-26.
CAREY, J. R., AND J. W. BRADLEY. 1982. Developmental rates, vital schedules, sex-ra-
tios and life tables for Tetranychus urticae, T. turkestani and T. pacificus (Ac-
arina: Tetranychidae) on cotton. Acarologia 23: 333-345.
COOK, B. G., R. J. WILLIAMS, AND G. P. M. WILSON. 1990.Arachispintoi Krap. et Greg.
nom. nud. (pinto peanut) cv. Amarillo. Australian J. Exp. Agric. 30: 445-446.
DE LA CRUZ, R., S. SUAREZ, AND J. E. FERGUSON. 1994. The contribution of Arachis
pintoi as a ground cover in some farming systems of Tropical America, pp. 102-
108. In P. C. Kerridge, and B. Hardy [eds.] Biology and agronomy of forage Ara-
chis. Centro Intern. Agric. Trop. Pub. 240, Cali, Colombia.
FERREIRA, D. N. M., AND C. H. W. FLECHTMANN. 1997. Two new phytophagous mites
(Acari: Tetranychidae, Eriophyidae) from Arachis pintoi from Brazil. Syst.
Appl. Acarol. 2: 181-188.
GROF, B. 1984. Forage attributes of the perennial groundnut Arachis pintoi in a trop-
ical savanna environment in Colombia. In Proc. International Grassland Con-
gress, 15., Kyoto, 1985. Centro Intern. Agric. Trop. Cali, Colombia, p. 168-170.
GUTIERREZ, J. 1976. Etude biologique et 6cologique de Tetranychus neocaledonicus
Andr6 (Acarien: Tetranychidae). Cahiers Orstom. Paris.
HULTING, F. L., D. B. ORR, AND J. J. OBRYCKI. 1990. A computer program for calcula-
tion and statistical comparison of intrinsic rates of increase and associated life
table parameters. Florida Entomol. 73: 601-612.
LEAS. Logiciels d'Enseignement et d'Analyses Statistiques. 1989. Chaire de Math6-
matiques. Unit6 de Biom6trie de l'Ecole Nationale Sup6rieure Agronomique de
Montpellier France.
RAI, A. B., A. S. SEJALIA, C. B. PATEL, AND S. KUMAR. 1995. The rate of natural in-
crease of red spider mite when reared on okra. Gujarat Agricul. Univ. J. 21:
SUAREZ-VASQUEZ, S., M. WOOD, AND S. NORTCLIFF. 1992. Crecimiento y fijacion de ni-
tr6geno por Arachis pintoi establecido con Brachiara decumbens. Cenicafe 43:
TSAI, S. M., K. S. KUNG, AND C. I. SHIH. 1989. The effect of temperature on life history
and population parameters of Kanzawa spider mite, Tetranychus kanzawai
Kishida (Acarina: Tetranychidae) on tea. Plant Protect. Bull. Taiwan 31: 119-
VILARREAL, M., AND 0. CHAVEZ. 1991. Adaptaci6n y producci6n de gramineas y legu-
minosas forrajeras en San Carlos, Costa Rica. Pasturas Tropicales 13: 31-38.

Florida Entomologist 83(2)

June, 2000


MATTHEWS, M. 1999. Heliothine Moths of Australia: A Guide to Pest Bollworms
and Related Noctuid Groups. CSIRO Publishing, Collingwood, Vict., Australia. x +
320 p. (23 color pl.) (17 x 25 cm), plus CD-ROM. ISBN 0-643-06305-6. Hardback. $90.

Although written from a project to identify the bollworms and related pest species
of this important economic group of moths, Matthews has provided a careful taxo-
nomic revision of the group that all students of the Noctuidae will find useful. The
book treats 38 species from Australia, with 18 new synonymies and 8 new species, in
5 genera. The book has excellent illustrations, including 23 color plates of all adults
and known larvae, plus 460 other figures halftonee photographs or SEM micrographs)
illustrating genitalia and other morphological characters of adults and immatures.
The CD-ROM included with the book includes a complete listing of all label data from
the over 14,800 specimens examined for the study and taxonomic lists. The illustra-
tions are all clear and finely printed. The color plates are very sharp and true to the
color tone of the moths and larvae. The book is finely printed and well bound.
Following an introduction on the methodology used, there are chapters on the eco-
nomic importance of Heliothinae in Australia, heliothine biology, systematics, mor-
phology, phylogeny, and identification keys to genera and species. Thereafter, the
author presents the species in the format of a traditional taxonomic revision, with di-
agnoses and descriptions of all taxa. Each species is discussed in detail, with new de-
scriptions, whether of known or new species, followed by notes on bionomics. Each
species has a range map included for Australia and Tasmania. After the taxonomic
section, there follows a checklist of all species for Australia, and then a chapter giving
detailed nomenclatural notes for all scientific names, including genera and all syn-
onyms. Prior to the monochrome figures of genitalia, there is a short section with
notes on the slides and specimens used for the figures.
The color plates include 2 plates showing some Australian habitats of heliothines,
followed by 2 useful color plates showing adults greatly enlarged and overprinted
with feature names for maculation and wing venation terms, plus details of the legs.
There are 8 color plates showing the Australian species from museum specimens
(about life size). There then is one color plate with enlarged views of the wing features
that allow identification of two closely related species: Helicoverpa armigera and H.
punctigera. Finally, there are 10 color plates illustrating adults in nature, larvae and
pupae of many of the species.
The author gives a detailed treatment for each species for Australia, but rather
less detailed discussion of the genera. This is partly due to his earlier work on the
world genera of Heliothinae (1991. Classification of the Heliothinae), where he al-
ready went into detail as to the generic limits for the subfamily. He gives further evi-
dence, particularly getting into molecular data, of the complexity of the heliothine
genera of Australia. Since Hardwick split Helicoverpa from the well-known genus He-
liothis in his 1965 monograph on North American heliothines, there has been contin-
ued argument from specialists as to whether Helicoverpa should be a subgenus or a
full genus. Part of these varying opinions were based no doubt on the similarities of
the species to be found in North America. Matthews, in covering all the many Austra-
lian species, clearly shows in this new study that the variation of the group is much
more complex outside of North America, thus further supporting Hardwick's more
preliminary work in splitting Heliothis. Thus, the Australian fauna in particular dem-
onstrates that the old concept of Heliothis is too broad to include so many different
species groups. Although many of the Australian species have characters that are

Book Review 207

very similar, Helicoverpa in particular is distinct enough to be a genus on its own, and
the other Australian groups vary so much from Heliothis that one concludes with Mat-
thews that they are best treated as separate genera: thus, Adisura, Heliothis, Helio-
cheilus, Australothis, and Helicoverpa.
Economic entomologists using this book will be able to accurately identify all Aus-
tralian heliothine adults and larvae: most species are distinct enough that the color
plates of the adults will suffice for identification, and only a few may require genitalic
dissection for species confirmation. The widespread Old World pest, Helicoverpa ar-
migera, occurs in Australia, thus the book is of use in other regions as well, particularly
since it is so carefully prepared and presented. The book is particularly important for
those involved in checking ports of entry for exotic pests, since anything from Australia
can be checked using this book as an identification guide for this group of moths.
Being virtually the finest and best illustrated revision of this group of moths for
any region of the world, all researchers on Noctuidae will need this book on their ref-
erence shelf, and likewise for economic entomologists.
J. B. Heppner
Florida State Collection
of Arthropods, FDACS, DPI
P. O. Box 147100
Gainesville, Florida 32614

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