Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00048
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1997
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: VID00048
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

Moraes et al.: Phytoseiids from northeastern Brazil


'Depto. Zoologia, ESALQ/USP, 13418-900 Piracicaba, SP, Brazil

CIAT, AA 6713, Cali, Colombia

3UFRPE, Recife-PE, Brazil

A new species of phytoseiid mite, Phyllodromus trisetatus n.sp., collected in north
eastern Brazil is described. Phyllodromus DeLeon, 1959 has been a monotypic genus
known only from Florida, U.S.A. Neoseiulus gracilis (Muma, 1962) is redescribed
based on the holotype and specimens from northeastern Brazil.

Key Words: predaceous mites, biological control, taxonomy, Gamasida


Es descrita una nueva especie de acaro fitoseido, Phyllodromus trisetatus n.sp., co
lectado en el noreste de Brasil. Phyllodromus DeLeon, 1959 fue un g6nero monotipico
conocido solamente de la Florida, U.S.A. Neoseiulus gracilis (Muma, 1962) es redescr
tito basado en el holotipo y en especimenes del noreste de Brasil.

Phytoseiid mites (Acari: Phytoseiidae) have received considerable attention world
wide because of their potential as natural enemies of phytophagous mites (McMurtry
1984). Few papers have reported on species of phytoseiid mites from Brazil (Moraes
et al. 1986), and only 4 papers on phytoseiid mites from northeastern Brazil (Farias
et al. 1981; Moraes & Oliveira 1982; Moraes & McMurtry 1983; Moraes et al. 1989).
The present paper provides a description of a new species of phytoseiid mite from
northeastern Brazil, and a redescription of Neoseiulus gracilis (Muma) based on the
holotype as well as specimens collected in northeastern Brazil.
All measurements are given in micrometers. Setal nomenclature is that of Rowell
et al. (1978) and Chant & Hansell (1971) for dorsal and ventral surfaces, respectively.
Dorsal and ventral idiosomal setal patterns are determined according to Chant &
Yoshida-Shaul (1989, 1991).


Phyllodromus DeLeon, 1959: 260; Muma, 1961: 290; Muma et al., 1970: 114

Phyllodromus trisetatus Moraes & Melo, n. sp.
(Figs. 1-5)

Diagnosis. This species is similar to the only other species in the genus, Phyllodro
mus leiodis DeLeon, 1959, but differs from it mainly by having JV1 on the ventrianal
plate and S2 and S4 setiform.

Florida Entomologist 80(3)

Female. (3 specimens measured). Dorsum -Dorsal plate faintly striate anterolat
erally, smooth or with faint circular pattern in the center, especially near J2; setal pat
tern 10A:9B; 388 (386-390) long, 223 (219-226) wide at s4 level,jl 18 (17-19),j3 37,j4
12 (11-14)j5 10 (1011),j6 18 (17-19), J2 12 (11-14), J5 7 (6-8), z2 38 (37-40), z4 44 (43
45), z5 11 (9-13), Z1 14, Z4 63 (62-65), Z5 61 (59-65), s4 56 (56-57), S2 52 (50-54), S4
13 (13-14), S5 11 (10-13), r3 38 (37-39), R1 16 (15-17). Setae j3, z2, z4, s4, S2, Z4, Z5
and r3 flattened and oblanceolate, with a small knob at the tips; other setae setiform.
Peritreme -Extending anteriorly to level slightly anterior to z2. Venter -Sternal and
genital plates smooth; ventrianal plate with a few transversal striae anterior to JV2
and reticulate posteriorly; metasternal plates smooth; metapodal plates punctuated.
All ventral setae setiform, except for JV5 which are flattened and oblanceolate. Dis
tances between setae ST1-ST3 44 (43-45), ST2-ST2 64 (63-66), ST5-ST5 62 (60-65).
Posterior margin of sternal plate expanded into a differentiated flap; sternal plate
with ST3 on hook-shaped posterior extensions. Ventrianal plate vase-shaped, 124
(122-128) long, 74 (73-76) wide at ZV2 level and 78 (76-80) wide at anus level; JV1 on
anterior margin of the plate; 2 small pores postero-laterad of JV2, in line with JV4.
Chelicera -Fixed digit 25 (22-28) long, with 7 teeth; movable digit 25 (24-26) long,
with 2 teeth. Spermatheca -Cervix deep bell-shaped, 18 (15-22) long; atrium en
crusted at the proximal portion of the cervix. Legs -Macrosetae absent on legs; chae
totaxy of GeII 2-2/1,2/1 1 and GeIII 1 2/1,2/0 1.
Male. Unknown.

4" K . t 5
._ rG

I_ 5~

Figs. 1-5. Phyllodromus trisetatus n. sp.: 1. female dorsal plate; 2. female ventral
surface; 3. female chelicera; 4. spermatheca; 5. female genu, tibia and tarsus of leg IV

September, 1997

Moraes et al.: Phytoseiids from northeastern Brazil

Locality and Type Material. Holotype female collected from Solanum erianthum,
Piritiba, State of Bahia, Brazil, on 25-VII-94, by A. R. de Luna; 2 paratype females col
elected from Waltheria indica, at Goiana, State of Pernambuco, Brazil, on 7-III-91, by
M. G. C. Gondim Jr. All types deposited at Depto. de Zoologia, ESALQ/USP.
Remarks. Phyllodromus trisetatus fits the description of the genus Phyllodromus,
except for having JV1 on the ventrianal shield; however, it seems that this should not
preclude the placement of the species in this genus, considering that it is not uncom
mon to observe variations even at the species level in relation to the location of prea
nal setae on or off the ventrianal plate. Phyllodromus leiodes is known only from
Florida, where it was collected from W indica, one of the plant substrates on which P
trisetatus was found in this study.
The trivial name of the new species refer to the presence of 3 preanal setae (JV1,
JV2 and ZV2) on the ventrianal plate.


Neoseiulus Hughes, 1948: 141; Muma, 1961: 295 (in part), DeLeon, 1965: 23; Muma
& Denmark, 1968: 235; Ragusa & Athias-Henriot, 1983: 660
Amblyseius (Neoseiulus), Karg, 1983: 313

Neoseiulus gracilis (Muma)
(Figs. 612)

Cydnodromus gracilis Muma, 1962: 9
Neoseiulus gracilis, Muma et al., 1970: 104
Amblyseius (Neoseiulus) atrii Karg, 1989: (new synonymy)
Material Examined: Sebring, Florida State, USA, from citrus litter, 11 IV 60, J. A.
Murrell holotypee female); Saint Lucia, Lesser Antilles, host?, 1980, S. Mahunka & L.
Mahunka-Papp holotypee female of Amblyseius (Neoseiulus) atrii Karg, 1989); Goi
ana, Pernambuco State, Brazil, from soil, 22-XI-90, M. G. C. Gondim Jr. (8 females, 9
males); Goiana, Pernambuco State, Brazil, from soil, 17-IV 91, M. G. C. Gondim Jr. (12
females, 6 males).
Female. (Figs. 6-10). Dorsum -Dorsal plate with a few striae anterolaterally; setal
pattern 10A:9B. The average measurements of 8 specimens collected in Brazil fol
lowed by the respective ranges and the measurements of the holotype are given sub
sequently: dorsal plate 329 (312-350) 342 long, 170 (155-179) 157 wide at s4 level,jl
12 (11-13) 15,j3 18 (17-19) 20,j4 15 (13-16) 18,j5 16 (16-17) 18,j6 16 (14-19) broken,
J2 19 (19-21) 20, J5 10 (9-11) 13, z2 16 (16-19) 18, z4 16 (16-17) 20, z5 17 (16-19) 18,
Z1 18 (17-19) broken, Z4 23 (22-25) 30, Z5 27 (24-30) 33, s4 17 (17-19) 23, S2 21 (19
22) 23, S4 20 (19-21) 23, S5 20 (17-22) 25, r3 12 (9-14) 17, R1 15 (14-16) 15. All setae
smooth. Peritreme extending anteriorly to level ofj 1. Venter -Sternal plate with a few
striae anteriorly and laterally; genital plate smooth; ventrianal plate with a few
transversal striae anterior to JV2 and reticulate posteriorly; metasternal plates
smooth; metapodal plates punctate. All ventral setae setiform. Distances between se
tae ST1-ST3 59 (54-62) 58, ST2-ST2 60 (54-65) 63, ST5-ST5 55 (51 59) 63. Ventrianal
plate shield-shaped, 114 (110-128) 121 long, 91 (88101) 101 wide at ZV2 level and 76
(73-81) 76 wide at anus level; 2 small pores posteromesad of JV2. Chelicera -Fixed
digit 31 (29-33) 28 long, with 4-5 teeth and a pilus dentilis; movable digit 31 (30-32)
30 long, with 1 tooth. Spermatheca -Cervix cup-shaped, 20 (19-22) 18 long; atrium
nodular. Legs -Macrosetae absent on legs of Brazilian specimens; macroseta found

Florida Entomologist 80(3)




6 7 10 12

Figs. 6 12. Neoseiulus gracilis (Muma) (drawings of specimens collected in Brazil)
6 female dorsal plate; 7. female ventral surface; 8. female chelicera; 9. spermatheca;
10. female genu, tibia and tarsus of leg IV; 11. spermatodactyl; 12. male ventrianal

only on basi-tarsus of leg IV of holotype, 30 long. Chaetotaxy of GeII 22/0,2/0-1; GeIII
Male. (Figs. 11-12) (5 Brazilian specimens measured). Dorsum -Dorsal plate stri
ate along the margins, 264 (254-276) long, 156 (151-166) wide at s4 level,jl 12 (10-12),
j3,j4,j6, J2, z2 and z4 16 (14-17),j5 and z5 15 (14-17), J5 10, Zl 18 (17-19), Z4 20 (19
22), Z5 19 (17-22), s4 and r3 14 (12-17), S2, S4 and S5 18 (17-19), R1 11 (10-12). All
setae smooth. Peritreme -Extending anteriorly almost to level ofj 1. Venter -Sterno
genital plate faintly striate. Ventrianal plate reticulate, sub-triangular, 109 (101-113)
long, 121 (120-125) wide at anterior corners, with 5 pairs of pores. Chelicera -Shaft
of spermatodactyl 13 (1214) long. Legs -Macrosetae absent on legs; chaetotaxy as in
Remarks. The measurements of the holotype female of A. (N.) atrii agrees well
with the measurements mentioned previously. Similarly to the holotype of N gracilis,
it also has a single macroseta, on basi-tarsus IV It is here considered a junior syn
onym of N gracilis. Hirschmann (1962) considered N gracilis junior synonym of Ne
oseiulus marinellus (Muma, 1962); Tuttle & Muma (1973) suspected that N gracilis
could be a senior synonym ofNeoseiulus mckenziei (Schuster & Pritchard, 1963). We
have not seen the types of N marinellus or N mckenziei. However, based on the orig
inal descriptions of the latter two, we consider them distinct from N gracilis and from
each other because of the different spermathecae. Apparently correctly, Ragusa &
Athias-Henriot (1983) synonymized N mckenziei under Neoseiulus barker Hughes,
1948, a species distinct from N gracilis.

September, 1997

Moraes et al.: Phytoseiids from northeastern Brazil


We are grateful to C. H. W Flechtmann (University of Sao Paulo, Brazil) and E.
Yoshida-Shaul (University of Toronto, Canada) for reviewing a previous version of
this paper, and to W C. Welbourn (Florida State Collection of Arthropods, U.S.A.) and
M. Moritz (University of Berlin, Germany) for loaning of holotypes.This work was par
tially sponsored by funds provided by IFAD (International Fund for Agricultural De
velopment), through an agreement with IITA (International Institute of Tropical


CHANT, D. A., AND R. I. C. HANSELL. 1971. The genus Amblyseius (Acarina: Phytosei
idae) in Canada and Alaska. Canadian J. Zool. 49: 703-758.
CHANT, D. A., AND E. YOSHIDA-SHAUL. 1989. Adult dorsal setal patterns in the family
Phytoseiidae (Acari: Gamasina). Internat. J. Acarol. 15: 219-233.
CHANT, D. A., AND E. YOSHIDA-SHAUL. 1991. Adult ventral setal patterns in the family
Phytoseiidae (Acari: Gamasina). Internat. J. Acarol. 17: 187-199.
DELEON, D. 1959. Two new genera of phytoseiid mites with a note on Proprioseius me
ridionalis Chant (Acarina: Phytoseiidae). Entomol. News 70: 257-262.
DELEON, D. 1965. A note on Neoseiulus Hughes 1948 and new synonymy (Acarina:
Phytoseiidae). Proc. Entomol. Soc. Washington 67: 23.
Predadores do acaro verde da mandioca no nordeste do Brasil. Pesquisa
Agropecuaria Brasileira 16: 313-317.
HUGHES, A. M. 1948. The mites associated with stored food products. Minist. Agr.
Fish., London, H. M. Stationary Office 168 p.
KARG, W. 1983. Systematische untersuchung der Gattungen und Untergattungen der
Raubmilben families Phytoseiidae Berlese, 1916, mit der Beschreibung von 8
neuen Arten. Mitt. Zool. Mus. Berlin 59: 293-328.
KARG, W. 1989. Zur Kenntnis der Raubmilbengattung Amblyseius Berlese, 1904 (Ac
arina, Parasitiformes, Phytoseiidae). Dtsch. ent. Z. N. F 36 13: 113-119.
MCMURTRY, J. A. 1984. A consideration of the role of predators in the control of acar
ine pests, pp. 109-121 in D. A. Griffiths, and C. E. Bowman (eds.). Acarology VI,
v. 1, Ellis Horwood Ltd., New York.
MORAES, G. J. DE, AND J. A. MCMURTRY. 1983. Phytoseiid mites (Acarina) of north
eastern Brazil with descriptions of four new species. Internat. J. Acarol., 9: 131
MORAES, G. J. DE, AND J. V. DE OLIVEIRA. 1982. Phytoseiid mites of coastal Pernam
buco, in northeastern Brazil. Acarologia, 23: 315-318.
1989. Explorations for natural enemies of the cassava green mite in Brazil, pp.
351 353 in 8th Symp. Int. Soc. Tropical Root Crops, Bangkok, Thailand, Oct. 30
Nov. 5, 1988.
MORAES, G. J. DE, J. A. MCMURTRY, AND H. A. DENMARK. 1986. A catalog of the mite
family Phytoseiidae: references to taxonomy, synonymy, distribution and habi
tat. EMBRAPA-DDT, Brasilia, 353 p.
MUMA, M. H. 1961. Subfamilies, genera, and species of Phytoseiidae (Acarina: Mesos
tigmata). Florida St. Mus. Bull. Biol. Sci., 5: 267-302.
MUMA, M. H., AND H. A. DENMARK. 1968. Some generic descriptions and name
changes in the family Phytoseiidae (Acarina: Mesostigmata). Florida Entomol.,
51: 229-240.
MUMA, M. H., H. A. DENMARK, AND D. DELEON. 1970. Phytoseiid of Florida. Arthro
pods of Florida and neighboring land areas, 6. Florida Dept. Agr. Cons. Serv,
Div. Plant Ind., Gainesville, 150 p.

324 Florida Entomologist80(3) September, 1997

RAGUSA, S., AND C. ATHIAS-HENRIOT. 1983. Observations on the genus Neoseiulus
Hughes (Parasitiformes, Phytoseiidae). Redefinition. Composition. Geography.
Rescription of two new species. Revue Suisse Zool., 90: 657-678.
ROWELL, H. J., D. A. CHANT, AND R. I. C. HANSELL. 1978. The determination of setal
homologies and setal patterns on the dorsal shield in the family Phytoseiidae
(Acarina: Mesostigmata). Canadian Entomol., 110: 859-876.
SCHUSTER, R. O., AND A. E. PRITCHARD. 1963. Phytoseiid mites of California. Hilgar
dia,34: 191-285.
TUTTLE, D. M., AND M. H. MUMA. 1973. Phytoseiidae (Acarina: Mesostigmata) inhab
iting agricultural and other plants in Arizona. Agric. Exp. St., Univ. Arizona,
Tucson, Tech. Bull., 208: 55 pp.


Florida Entomologist 80(3)


Subtropical Agricultural Research Laboratory, USDA-ARS
2301 S. International Blvd., Weslaco, TX 78596


Coatings applied to fruits have been shown to kill tephritid fruit fly immatures in
side of the fruits. The present research investigated the efficacy of coatings against
distinct life stages of Mexican fruit fly, Anastrepha ludens (Loew), and results showed
high levels of disinfestation of grapefruits of up to the early third instar (95%) for one
commonly used grapefruit coating, Citrus Lustr 402. Emergence was reduced signifi
cantly even for late third instars. Leaving one-third of each grapefruit uncoated re
duced efficacy considerably. Mixing Citrus Lustr 402 into the diet used to rear
Mexican fruit fly did not affect survival indicating that this coating is not toxic to lar
vae. This research supports the hypothesis that coatings act primarily to modify at
mospheres inside the fruits and kill larvae by restricting gaseous exchange. Fruit
coating could be incorporated as a component of an integrated systems approach to
quarantine security where a series of pest infestation-reducing steps decreases risk to
insignificant levels.

Key Words: Fruit wax, quarantine security, systems approach, Anastrepha ludens


Ha sido demostrado que ciertas cubiertas aplicadas a las frutas matan a los inma
duros de las moscas tefritidas en el interior de las mismas. En la present investiga
ci6n se estudi6 la eficacia de las cubiertas contra diferentes estadios de la mosca
mexicana de las frutas, Anastrepha ludens (Loew). Los resultados mostraron que la
desinfestaci6n de toronjas tratadas con la cubierta comunmente usada alcanza el 95%
en el tercer estadio temprano de la mosca. La eficacia se redujo considerablemente al
dejar un tercio de cada fruta sin cubrir. La mezcla de Citrus Lustr 402 con la dieta
usada para criar la mosca mexicana no afect6 la supervivencia, indicando que esa cu
bierta no es t6xica a las larvas. Esa investigaci6n sostiene la hip6tesis de que las cu

September, 1997

Hallman: Fruit fly mortality in coated fruit

biertas actuan primariamente modificando la atmdsfera dentro de las frutas y matan
las larvas mediante la restricci6n del intercambio de gases. La cubierta de las frutas
podria ser incorporada a los sistemas integrales de seguridad cuarentenaria donde
una series de pasos para la reducci6n de la infestaci6n de plagas disminuya el riesgo a
niveles insignificantes.

Tephritid fruit flies are major horticultural pests and probably the chief group of
quarantined pests worldwide. To prevent inadvertent introduction of fruit fly species
into areas of the world where they do not exist but could become established, fruits are
subjected to quarantine treatments, shipped from areas certified to be free of the
pests, or packed under systems which reduce the risk of infestation to negligible levels
(Sharp & Hallman 1994).
Fruit coatings have been shown to kill Caribbean fruit fly, Anastrepha suspense
(Loew), immatures in various fruits (Hallman 1996, Hallman & Foos 1996, Hallman
et al. 1994, 1995). One-hundred percent Caribbean fruit fly mortality was observed in
grapefruits coated with Sta-Fresh 600, a non-drying coating commercially applied to
melons in transit and washed off after arrival (Hallman et al. 1994). Apparently, coat
ings also killed Mediterranean fruit fly, Ceratitis capitata (Wiedemann), immatures in
fruits (Saul et al. 1985, 1987). Coatings probably kill fruit fly immatures inside of
fruits largely by creating a modified atmosphere (Hallman 1994).
None of the previous research addressed the effect of coating fruit on different fruit
fly stadia. The goals of the research reported herein were to determine if coating
grapefruit would kill Mexican fruit fly, Anastrepha ludens (Loew), and to determine
the mortality levels at different insect life stages. An experiment was also conducted
to determine if coatings would kill Mexican fruit fly larvae when mixed in their diet.


Mexican fruit flies were from a colony reared on a semi-artificial diet at the U.S.
Department of Agriculture, Subtropical Agricultural Research Laboratory in Weslaco,
Texas (Spishakoff & Hernandez-Davila 1968). Grapefruit cultivars 'Ruby Red' and
'Rio Red' (mean weight about 450 g) were placed 180-200 fruits at a time in an alumi
num screen cage (228 x 81 x 46 cm) with about 10,000 Mexican fruit fly adults for 24
hours. About half of the flies were females, and all were fed water, sugar, and yeast hy
drolysate. After exposure to oviposition, the grapefruits were cleaned with water and
light hand scrubbing and held at about 24'C until the Mexican fruit fly immatures
reached the desired stage: early egg (1 day), second instar (7-8 days), early third in
star (11 14 days), and late third instar (15 18 days). Several grapefruits were cut open
and the stage of fruit fly development verified before grapefruits were coated. The ex
perimental design was a randomized complete block with three replicates and 20
grapefruits per replicate-treatment combination, including uncoated controls.
The following coatings were used: Sta-Fresh 590 HS (FMC Corp., Lakeland, FL),
Citrus Lustr 402 (ELF Atochem North America, Inc., Monrovia, CA), and Nature Seal
2020 (EcoScience, Orlando, FL). Sta-Fresh 590 HS and Citrus Lustr 402 are commer
cially-used, high-gloss citrus coatings which contain alkali soluble resins, propylene
glycol, fatty vegetable acid soaps, and silicone antifoam. Nature Seal 2020 is a cellu
lose-based coating which is used on limes. Each grapefruit was hand coated with 0.9
1.0 ml, which is approximately the rate used commercially on grapefruits in southern
Texas, and allowed to dry in ambient air.

Florida Entomologist 80(3)

To determine if complete coating of fruit was necessary to achieve fruit fly mortal
ity, in one test one-third of the surface area of the grapefruits was covered with a sin
gle piece of 5.15-cm wide tape before coating and then removed about one-half hour
after coating, leaving about one-third of the fruit uncoated in a single patch. This test
was conducted on fruits 10-13 days after infestation when most larvae had developed
to early third instar. The experimental design was randomized complete block with
three replicates and 20 fruits per replicate including the uncoated controls.
All grapefruits were placed individually in 2-liter plastic containers containing
about 250 cm3 of sand which provided a burrowing and pupation site for emerging lar
vae. About two weeks after larvae began emerging, the fruits were dissected, and all
puparia and live and dead larvae were counted.
To test if the coatings were actually toxic to Mexican fruit fly larvae, 1 ml of coating
was mixed with 150 ml of diet, placed in 275-ml plastic containers, and infested with
100 early third instars. When the larvae completed feeding they were removed from
the diet and placed in 0.5-liter heavy paper containers with 100 cm3 of vermiculite and
held for pupation and adult emergence. This experiment was a randomized complete
block with four replicates, including the controls without coating.
All data were analyzed by the SAS ANOVA procedure after normality was tested
with the UNIVARIATE procedure (SAS Institute, Inc. 1988). Data which were not
normal were transformed by log(n + 1) and then tested again.


Analysis of variance [log(n + 1)] showed significant differences among numbers of
Mexican fruit fly larvae emerging from grapefruits coated at different life stages of
the insect (F = 5.14; df 3,6; P < 5%) and between the different coatings used (F = 20.3;
df = 3,6; P < 1%) (Table 1). The least number of larvae emerged from grapefruits
coated with Citrus Lustr 402. Standard errors of the mean overlapped only in the case
of third instars in grapefruits coated with Citrus Lustr 402 and Sta-Fresh 590, indi
casting no difference between these two coatings for third instars. Survival remained
low through the early third instar, but increased greatly among late third instars. The
life stage by coating interaction was not significant (F = 1.74, df = 9,18), indicating
that the relative effectiveness of the three coatings was similar among the various
Mexican fruit fly life stages.



Stage Citrus Lustr 402 Sta-Fresh 590 Nature Seal 2020

1 day old egg 0.03 +0.03 2.6 1.3 9.0 + 0.9
2nd instar 0.02+ 0.02 1.8 1.5 10.7 +4.2
Early 3rd instar 1.1 + 1.1 3.1+ 2.0 14.7 +3.6
Late 3rd instar 6.6+ 2.6 4.5+ 2.2 14.9 +3.7

'Mean number of larvae from uncoated control= 19.0 + 4.0.

September, 1997

Hallman: Fruit fly mortality in coated fruit

Mixing Citrus Lustr 402 into the diet did not affect survival of early third instar
Mexican fruit fly to the adult stage (mean of 81% survival for control versus 80% for
diet plus coating)demonstrating that the coating was not directly toxic to the insect.
Mortality of Mexican fruit fly in grapefruits that had about two-thirds of the sur
face area coated leaving a single, uncoated 5.15 cm wide strip was greatly reduced
compared with totally coated fruits. Mean number of insects per grapefruit ( SEM)
was 21.1 + 4.6, 7.8 1.5, and 8.5 + 0.9 for the control, Citrus Lustr 402, and Sta-Fresh
590, respectively. Grapefruits infested with early third instars and that were two
thirds coated yielded 37-40% of the total Mexican fruit fly larvae emerging from un
coated grapefruits compared with 5.8-16% for completely coated grapefruits. Never
theless, analysis of variance indicated significant differences between the control and
the two partially coated treatments (F = 7.09; df = 2,4; P < 5%).


Application of citrus coatings at commercial rates provided high levels of disinfes
station of Mexican fruit fly immatures from grapefruits. Although the level of reduc
tion was inadequate to provide quarantine security, which requires virtually 100%
mortality, the data suggest that coatings could be easily incorporated as a component
of a quarantine security system consisting of a series of pest mitigating steps to re
duce the risk of infestation to a negligible level (Hallman 1995). Partial coating re
duced the effect greatly, but still provided some abatement. The coating even provided
significant mortality of late third instars, many of which probably could have avoided
mortality simply by emerging from the fruit. In this study, Citrus Lustr 402 was
markedly better than the other two coatings used in reducing Mexican fruit fly sur
vival in grapefruits. It is arguable that the effect of coatings on fruit fly disinfestation
would be more pronounced than these studies indicated because most of the grape
fruits that were coated would have been culled due to their substandard condition
caused by remaining at room temperature for 7-18 days after infestation while wait
ing for the insects to reach the desired stage of development before coating. Commer
cially, fruits would be coated very soon after harvest.
Because the coating itself was not directly toxic to the larvae it seems likely that
the mode of action of coatings is simply a modified atmosphere where lowered oxygen
and raised carbon dioxide levels kill insects inside of fruits (Hallman et al. 1994).


I thank Mike Diaz and Rene Martinez, USDA-ARS, Weslaco, Texas, for technical
assistance and Julia Olivarez for rearing Mexican fruit fly Bill Weeks, Texas Produce
Association, supplied Sta-Fresh 590 and some of the grapefruits, and Craig Campbell,
then with EcoScience, provided Nature Seal 2020.


HALLMAN, G. J. 1994. Controlled atmospheres, pp. 121-136 in R. E. Paull and J. W
Armstrong [eds.], Insect pests and fresh horticultural products: treatments and
responses. CAB International, Wallingford, England.
HALLMAN, G. J. 1995. Incorporation of fruit coatings into quarantine security sys
tems. Proc. Ann. Internat. Res. Conf. on Methyl Bromide Alternatives and
Emissions Reductions, Nov. 6-8, San Diego, CA. pp. 89, 1-2.
HALLMAN, G. J. 1996. Mortality of Caribbean fruit fly in coated fruits, pp. 495-498 in
B. A. McPheron and G. J. Steck, eds, Fruit Fly Pests: A World Assessment of
their Biology and Management. St. Lucie Press, Delray Beach, FL.

Florida Entomologist 80(3)

September, 1997

HALLMAN, G. J., AND J. F. Foos. 1996. Coating combined with dimethoate as a quar
antine treatment against fruit flies (Diptera: Tephritidae). Postharvest Biol.
and Technol. 7: 177 181.
of feral Caribbean fruit fly (Diptera: Tephritidae) immatures in coated guavas.
J. Econ. Entomol. 88: 1353-1355.
1994. Mortality of Caribbean fruit fly (Diptera: Tephritidae) immatures in
coated fruits. J. Econ. Entomol. 87: 752-757.
SAS INSTITUTE. 1988. SAS/STAT User's Guide, release 6.03 edition. SAS Institute,
Cary, NC.
Laboratory trials of methoprene-impregnated waxes for preventing survival of
adult oriental fruit flies (Diptera: Tephritidae) from infested papayas. J. Econ.
Entomol. 80: 494-496.
SAUL, S. H., R. F. L. MAU, AND D. OI. 1985. Laboratory trials of methoprene-impreg
nated waxes for disinfesting papayas and peaches of the Mediterranean fruit
fly (Diptera: Tephritidae). J. Econ. Entomol. 78: 652-655.
SHARP, J. L., AND G. J. HALLMAN [eds.]. 1994. Quarantine Treatments for Pests of
Food Plants. Westview Press, Boulder, CO.
SPISHAKOFF, L. M., AND J. G. HERNANDEZ-DAVILA. 1968. Dried torula yeast as a sub
stitute for brewer's yeast in the larval rearing medium for the Mexican fruit fly.
J. Econ. Entomol. 61: 859-860.


Florida Entomologist 80(3)


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

2Centro di Studio CNR sulle Tecniche di Lotta Biologica (CETELOBI) 80055 Portici
Via Universita, 133, Italy


Coccobius donatellae Pedata and Evans, spec. nov. is described and illustrated
from specimens reared from Comstockiella sabalis (Comstock) on palmetto palm (Sa
balpalmetto) in Florida. Coccobius donatellae is the most common parasitoid that at
tacks this host in Florida and is believed to be the same species reported in the
literature as "Physcus sp." that was introduced into Bermuda from Florida in the
1920's. Evidence suggests that earlier reports of Encarsia portoricensis (Howard) as a
parasitoid of the palmetto scale are based on erroneous identifications of what were
probably Coccobius donatellae males. Recent collections in Florida confirm Aphytis di
aspidis (Howard), reported previously as Aphytis fuscipennis, and Encarsia citrina

September, 1997

Evans & Pedata: A New Coccobius Species

(Craw) as parasitoids of C. sabalis. Intraspecific variation occurring in C. donatellae
and in Coccobius testaceus (Masi), is discussed.

Key Words: Coccobius, Aphelinidae, Diaspididae, Comstockiella, armored scale, bio
logical control, parasitoid


Se describe y se ilustra Coccobius donatellae Pedata and Evans, spec. nov., criado
de Comstockiella sabalis (Comstock) sobre la palma palmetto (Sabalpalmetto) en Flo
rida. Coccobius donatellae es el parasite mas comun que ataca este hospedero en Flo
rida y se cree que es la misma especie reportada en la literature como "Physcus sp. que
fue introducida a Bermuda de Florida en los anos 1920. Se present evidencia que in
dica que los informes anteriores de Encarsia portoricensis (Howard) como parasite de
C. sabalis son basados sobre identificaciones err6neas de los machos de Coccobius do
natellae. Se confirm Aphytis diaspidis (Howard), reportado anteriormente como,
Aphytis fuscipennis, y Encarsia citrina (Craw) como parasitos de C. sabalis basado en
las recolecciones reci6n hechas en Florida. Se incluye informaci6n sobre la variaci6n
intraespecifica que ocurre en C. donatellae y en Coccobius testaceus (Masi).

Comstockiella sabalis (Comstock) is an armored scale insect (Homoptera: Diaspi
didae) known from the southern United States, Mexico, several of the Caribbean Is
lands, and from greenhouses in Germany (Nakahara, 1982). Although it is commonly
found on palm species, it rarely causes economic damage due to the severe attack of
parasitoids on this species throughout its geographic range. However, this has not al
ways been the case. C. sabalis invaded Bermuda in 1921 and quickly spread through
out the islands, severely damaging or killing Sabal bermudana Bailey trees (Russell
1934a). Parasitized C. sabalis specimens were collected in Florida and sent to Ber
muda in 1926 and 1929. No mention was made of the specific identity of parasitoids
introduced into Bermuda from Florida at that time; however, in a survey of the natu
ral enemies of the palmetto scale in Bermuda conducted in 1933, Physcus sp., Encar
sia portoricensis Howard, Aphytis fuscipennis Howard and two undetermined
Hymenoptera were reported as being reared from this host (Russell, 1934b). The par
asitoid referred to in the survey as "Physcus sp.", now placed in the genus Coccobius,
was particularly effective against the scale. Russell (1934b) reported that "the palmet
tos on which this species was placed that were once badly infested, were later free
from scale". Bennett and Hughes (1959) reared Physcus sp. and Aphytis fuscipennis
from C. sabalis collected in Bermuda in 1956 and stated that "it would seem that E.
portoricensis is no longer of importance as a control for this scale".
Recent collections of C. sabalis in Florida have helped to clarify our knowledge of
the natural enemies of C. sabalis in Florida and provided insight as to the identity of
the parasitoid species introduced into Bermuda from Florida in the 1920's. We suggest
that specimens identified in the Bermuda survey as "Physcus sp." and Encarsia por
toricensis, represent the female and male of Coccobius donatellae, respectively. This
species is the most common parasitoid reared from C. sabalis in Florida, and undoubt
edly plays a key role in its control. Evidence supporting our hypothesis that speci
mens reared from C. sabalis in the Bermuda survey that were identified as Encarsia
portoricensis were actually males of Coccobius donatellae consists of: Encarsia por
toricensis is a whitefly parasitoid that is not known to occur in Florida; males of Coc
cobius donatellae are similar to females of E. portoricensis in color, and in the number

Florida Entomologist 80(3)

of and relative lengths of antennal segments (6-segmented); and specimens deposited
in the Museum of Natural History, London from the 1933 Bermuda survey, identified
by Ferriere as Encarsia sp., were later identified as Coccobius males (A. Polaszek, per
sonal communication).
The third species mentioned in the survey, Aphytis fuscipennis Howard, was syn
onymized with Aphytis diaspidis (Howard) by Rosen and DeBach (1979), who did not
list C. sabalis as one of its hosts. We confirm the identity of A. diaspidis based on three
specimens of Aphytis diaspidis reared from C sabalis from the 1933 Bermuda survey
and deposited in the Florida State Collection of Arthopods, Gainesville, Florida. Our
collections in Florida confirm Aphytis diaspidis and Encarsia citrina (Craw) as para
sitoids of C. sabalis; it appears that both of these species play minor roles in control
ling populations of the scale.
The majority of the 79 described species of the genus Coccobius are parasitoids of
diaspine scales; 10 species have been reported as parasitoids of soft scales (Coccidae),
1 species from a mealybug (Pseudococcidae), and 1 species from a lac scale (Kerridae).
Hayat (1984) reviewed the 58 Coccobius species known worldwide at that time and
provided a taxonomic key to 48 of those species. Since then, twenty-one species have
been described; of these are, 7 from South Africa (Prinsloo, 1995), 10 from China
(Huang, 1990), 1 from Japan (Tachikawa, 1988), 2 from Turkmenia (Myartseva, 1995)
and 1 from Azerbaijan (Jasnosh and Mustafeva, 1992). Only 6 species are known to oc
cur in the continental United States; of these, 2 species (howardi, stanfordi) were de
scribed from California, 1 species (varicornis) from Washington, DC, and 3
flaviventriss, fulvus, testaceus) are introduced species. Coccobius donatellae is the
fourth species of this genus to be described from the continental United States.
Terminology follows that used by Hayat (1984). Figure 1 shows the mesosoma di
vided medially with the surface sculpturing on the left side and the station on the
right side. The metasoma is divided medially showing the dorsum on the left side and
the venter on the right side.

Coccobius donatellae Pedata and Evans, NEW SPECIES
(Figs. 1-6)

Female (Figs. 1-4)

Length: 0.70-0.90 mm, mean of 5 specimens = 0.82 mm. Coloration: Body (Fig. 1)
yellowish; basal half of head dark brown; pronotum, metanotum, metasomal tergites
I-VI, fuscous; legs white with central portion of femora and basal two-thirds of tibiae,
faintly fuscous; antennae yellowish, basal half of scape, dorsal margin of pedicel, Fl
and club, grayish; fore wing hyaline. Structure: Head slightly wider than mesosoma.
Antenna (Fig. 3) consists of radicle (R), scape (S), pedicel (P), 3 funicle segments (Fl
F3) and 2 club segments (F4-F5), length:width ratio for each segment as follows:
R:3.2, S:3.5, P:1.4, Fl:1.5, F2: 1.7, F3:1.6, F4:1.4, F5:2.5; relative length of each seg
ment to length of Fl segment: R: 1.1, S:2.8, P: 1.1, F1:1.0, F2:1.3, F3:1.3, F4:1.3, F5:2.2;
flagellar segments Fl F6 with 2,2,2,2 and 5-6 linear sensilla, respectively Mesosoma
with broad mesoscutum, 1.8x as wide as long with approximately 40 setae and small,
reticulate cells each with internal striations; scutellum with 3 pairs of setae, and
sculpturing similar to that of mesoscutum; mesophragma reaching base of metasomal
tergite II. Fore wing 2.6x as long as wide, discal station uniformly distributed with
narrow asetose area basally near posterior margin; marginal vein as long as costal
cell with 10-12 marginal setae; submarginal vein with 6-7 setae; longest marginal
cilia 0.2x as long as the maximum width of fore wing. Metasoma slender, 1.7x as long

September, 1997

Evans & Pedata: A New Coccobius Species

as mesosoma, tergites I-VI with reticulate lateral margins, tergites V VI with stipules,
centrally; tergites I-VII with 1,4,4,4,3,6,6 pairs of setae, respectively; ovipositor arises
at level of tergite II, slightly protruding from apex, 1.7-1.9x as long as tibia II (Fig. 2)
and 4.1x as long as valvular III.

Male (Figs. 5-6)

Coloration: Head with occiput yellow and basal half, dark brown; mesoscutum,
scutellum and axillae, light brown; pronotum, metanotum, metasoma and coxae, dark
brown; femora, except for pale apices, and proximal two thirds of tibiae, brownish;
tarsi yellow; antennae fuscous, fore wing hyaline. Differs structurally from the female
primarily by the 6-segmented flagellum (Fig. 5) and by the scape which has a ventral,
circular glandular area (Viggiani et. al., 1986) separated from the 6 medial pores.
Length:width ratios of antennal segments R-F6 as follows: R:2.1, S:2.8, P:1.3, Fl:1.4,
F2:1.5, F3:1.5, F4:1.5, F5:1.6, F6:1.8; relative length of each segment to length of Fl:
R:0.8, S:2.1, P:0.9, F1:1.0, F2:1.1, F3:1.1, F4:1.1, F5:1.2, F6:1.2.

Morphological variation

Individuals of Coccobius donatellae vary primarily in body size, number of mesos
cutal setae, number and size of reticulated cells of the mesoscutum, relative lengths
of the flagellar segments, and the relative length of the marginal fringe of the fore
wing to its maximum width. In general, smaller individuals have fewer mesoscutal
setae (30-36), larger and fewer reticulate cells on the mesoscutum, and longer mar
ginal fringes (0.22-0.26x maximum width of fore wing) than do larger individuals (40
46 mesoscutal setae, marginal fringe = 0.12-0.16x maximum width of fore wing). The
Fl antennal segment tends to be shorter (Figs. 4, 6) in smaller individuals, at times,
almost quadrate, 1.1-1.4x as long as wide, and 0.8x as long as the F2; whereas in
larger individuals, the Fl is usually more elongate, 1.5-1.7x as long as wide and ap
proximately as long as the F2.


The female of Coccobius donatellae can easily be distinguished from females of the
other 3 species described from the continental United States by the coloration of its
body which is almost entirely yellow; whereas the head, mesosoma and at least part
of the metasoma of the other species are dark brown. Coccobius donatellae is most
similar in coloration and structure to Coccobius testaceus (Masi), a European species
introduced into California for the control of Lepidosaphes ulmi L. and L. conchiformis
(Gmelin) (Flanders 1942). Females of C. testaceus can be distinguished from females
of C. donatellaeby the grayish F2 segment (Fig. 7), reported in the past as being pale,
and the pale apical half of F5 segment and relative length of the ovipositor:tibia II
(1.4-1.5:1). Males of C. testaceus differ from C. donatellae males by having the head
completely dark brown, the length of the pedicel only about one half as long as the Fl
segment, and by the larger, contiguous glandular area on the scape. Most C. testaceus
males have 2 rows of linear sensilla on the Fl (Fig. 8); however in smaller individuals
there may be a single row (Fig. 9).
Material examined

Female holotype (in Canada Balsam), 129, 66 paratypes (in Modified Hoyer's
Mounting Medium), 79, 5 paratypes (in Canada Balsam), 109,3 paratypes (card

Florida Entomologist 80(3)

Figures 1-9. (1 6) Coccobius donatellae 1) 2 habitus 2) 2 tibia II 3) 2 antenna (nor
mal) 4) 9 antenna (small individual) 5) 6 antenna (normal) 6) 6 antenna segments
R-F1 (small individual); (7-9) Coccobius testaceus 7) 9 antenna 8) 6 antennal seg
ments R-F2 (normal) 9) 6 antennal segments R-F2 (small individual).

mounted) as follows: United States, Florida, Levy County, Cedar Key, 18 VI 1988, by
F D. Bennett, reared from Comstockiella sabalis on Sabalpalmetto. Additional collec
tions: 4 Florida, Escambia Co., Pensacola, 10 III 1991, by F D. Bennett, reared from
Comstockiella sabalis on Sabal palmetto; 4 Florida, Osceola Co., Canoe Creek, 1 IV
1991, by F D. Bennett, reared from Comstockiella sabalis on Serenoa repens; 7 Flor
ida, Lee Co., Ft. Meyers, 23 XI 1991, by F D. Bennett, reared from Comstockiella sa
balis on Sabal palmetto; 2 9, 16, Florida, Alachua Co., Gainesville, 10 V 1996, by P.A.
Pedata, reared from Comstockiella sabalis on Sabal palmetto.

September, 1997

Evans & Pedata: A New Coccobius Species


Female holotype and 59, 56 paratypes are deposited in the United States Na
tional Museum of Natural History, Washington, D.C.; remaining paratype specimens
are deposited in Florida State Collection of Arthropods, Gainesville, Florida; the Nat
ural History Museum, London, England; and the Dipartimento di Entomologia e Zoo
logia Agraria, Universita di Napoli "Federico II", Portici, Italy.


Coccobius donatellae is named in memory of Donatella Pedata.


We thank Fred D. Bennett who collected the majority of the specimens used in this
study. Avas Hamon for identification of specimens of Comstockiella sabalis. Andrew
Polaszek for his assistance and Gennaro Viggiani for his advice and support. Funding
for the senior author provided, in part, by the National Biological Control Institute,
USDA/APHIS, Postdoctoral Fellowship in Systematics. Florida Agricultural Series
No. R-05675.


BENNETT, F. D., AND I. W. HUGHES. 1959. Biological control of insect pests in Ber
muda. Bulletin of Entomological Research 50: 423-436.
FLANDERS, S. E. 1942. The introduction of Physcus testaceus Masi into California.
Journal of Economic Entomology 35: 290-291.
HAYAT, M. 1984. Notes on some species of Coccobius and Prophyscus (Hymenoptera:
Aphelinidae), with special reference to Girault and Howard types. Oriental In
sects 18: 289-334.
HUANG, J. 1990. Systematic studies on Aphelinidae of China (Hymenoptera: Chalci
doidea). Contributions of the Biological Control Research Institute. Fujien Ag
ricultural University. Special Publication No. 5. 348 pp.
JASNOSH, V. A., AND G. A. MUSTAFEVA. 1992. A new parasite of the pomegranate
scale, Coccobius granati sp. n. (Hymenoptera: Aphelinidae). Zool. Zh. 71(2):
MYARTSEVA, S. N. 1995. New species of aphelinids (Hymenoptera, Aphelinidae) par
asites of scale insects on Tamarixin Turkenia. Entomol. Obozr. 74(2): 432-440.
NAKAHARA, S. 1982. Checklist of the armored scales (Homoptera: Diaspididae) of the
conterminous United States. USDA, APHIS-PPQ. 110 pp.
PRINSLOO, G. L. 1995. Revision of the southern African species of Coccobius Ratze
burg (Hymenoptera: Aphelinidae), parasitoids of armoured scale insects. Jour
nal of Natural History 29: 1517-1541.
ROSEN, D., AND P. DEBACH. 1979. Species of Aphytis of the World (Hymenoptera: Aph
elinidae) Series Entomologica vol. 17, Dr. W Junk BV Publishers, The Hague,
810 pp.
RUSSELL, T. A. 1934a. An account of the palmetto scale. Agricultural Bulletin, Ber
muda Department of Agriculture 13(3): 17-24.
RUSSELL, T. A. 1934b. The use of parasites against the palmetto scale. Agricultural
Bulletin, Bermuda Department of Agriculture 13(11): 81-86.
TACHIKAWA, T. 1988. A new and economically important species of Coccobius (Hy
menoptera: Aphelinidae) parasitic on Hemiberlesia pitysophila Takagi (Ho
moptera: Diaspididae) in Okinawa, Japan. Transactions of the Shikoku
Entomological Society 19: 6771.

334 Florida Entomologist80(3) September, 1997

VIGGIANI, G., BATTABLIA, D, AND R. JESU. 1986. Laccoppiamento di Physcus testaceus
Masi (Hym. Aphelinidae), con notizie preliminary sulle struttura dello scapo
antennale maschize. Bollettino del Laboratorio di Entomologia Agraria Filippo
Silvestri 43: 16.


Florida Entomologist 80(3)


'Agricultural Research Programs, Florida A&M University
Tallahassee, FL 32307-4100, rflowers@nsl.famu.edu

Department of Biology, University of Pennsylvania, Philadelphia, PA 19104


Host plant associations are given for 137 species representing 7 subfamilies and
92 genera of Costa Rican Chrysomelidae. A numeric score is introduced to objectively
describe confidence that a field observation of an interaction between a chrysomelid
and a plant represents true herbivory Literature host plant records, if they exist, are
given for included chrysomelid taxa.

Key Words: herbivory, Criocerinae, Chrysomelinae, Cryptocephalinae, Eumolpinae,
Galerucinae, Hispinae, Lamprosominae, host plants


Se presentan asociaciones de plants hospederas para 137 species de Chrysome
lidae de Costa Rica, representando 7 subfamilias y 92 g6neros de escarabajos. Se in
troduce una calificaci6n numerica para describir objetivamente la confianza en que
una observaci6n de campo de una interacci6n entire un escarabajo y una plant repre
senta un caso verdadero de herbivoria. Se presentan datos de plants hospederas de
la literature, si existen, para los taxa de escarabajos incluidos.

In recent years, there has been a surge of interest in relationships between tropi
cal plants and insects. The interest is driven by the related agendas of studying them
for their intrinsic scientific interest, and protecting tropical biodiversity through find
ing practical and non-destructive ways to use it. The latter agenda is exemplified by
the biochemical prospecting programs recently started in several areas of the world
(Reid et al. 1993).
Most plant-insect research begins with a basic event: an observation that a specific
plant is somehow important in the life cycle of a specific insect. Unfortunately, huge

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records

sections of the tropical insect fauna are still unusable as subjects of insect-plant re
search because that first step of linking plant and insect taxa has been largely ne
glected. In-depth studies of plant-insect interactions have focused on temperate zone
insects and on a few relatively well known tropical groups (e.g., Lepidoptera). Only a
small percentage of the fauna of tropical herbivores has been similarly studied.
The family Chrysomelidae (Coleoptera), or leaf beetles, is a natural subject for
studying plant-insect and inter-herbivore interactions (Strauss 1988). Of the esti
mated 37,000 species, world-wide, in this family, almost all, as far as we know, are
herbivores or seed predators. However, for about 70% of the described species, we do
not have records of host plants. Most of the known host plant records are Holarctic
(Jolivet 1988b). For Neotropical Chrysomelidae other than Bruchinae, the most spe
cific information treats economically important species (e.g., King & Saunders 1984,
Ostmark 1975, Jolivet 1979, Hilje et al. 1991). However, a review of known host plants
of the tortoise beetles (Cassidinae) of Panama was recently published by Windsor et
al. (1992); Moldenke (1971) listed host plants for some Mexican Chrysomelidae, and
Anaya (1989) reviewed the known host plants of North and Central American Chry
somelinae. Jolivet, in a series of papers (1977, 1978, 1982, 1987a, 1987b, 1988a, 1991;
Jolivet et al. 1986) and in a recent book (Jolivet & Hawkeswood 1995) summarized
current host plant data on a world level for the Chrysomelidae. However, in much of
this literature, beetle species are usually identified only to genus and their plant hosts
only to family. A few field studies have documented significant attacks by chry
somelids on plants in Central American ecosystems (e.g., Rockwood 1974, Memmott
et al. 1993), and some detailed field and laboratory studies have been undertaken for
several Neotropical species (Bach 1986, Begossi & Benson 1988, Buzzi & Winder
1986, Hsiao 1988, Strong 1977a,b). Apart from these ecological studies of specific
chrysomelids, many of the published host plant records are of dubious value, stating
merely that beetle X was taken on plant Y (or, all too often, "genus X feeds on plant
genus Y"). A further problem, also noted by Furth (1985), is that a large proportion of
such records are buried in taxonomic monographs and regional studies (e.g., Bechyne
& Bechyne 1975) and accessible only by reading these studies in their entirety. Much
more data on a much broader spectrum of chrysomelid taxa will have to be accumu
lated and made available before any credible generalizations about the nature of leaf
beetle-plant interactions can be made.
In this paper, we present feeding records of adults and larvae for 137 species of
Costa Rican Chrysomelidae, representing 7 subfamilies and 92 genera. The majority
of these observations were made by the senior author during a six-month sabbatical
at Costa Rica's Instituto Nacional de Biodiversidad (INBio) in 1991, and by the junior
author during the years 1978 to 1995 as a byproduct of an on-going intensive study of
the caterpillars of the dry forests of Sector Santa Rosa of the Guanacaste Conserva
tion Area (Janzen 1993, Janzen & Gauld 1996). Our records include results from di
rect observations of free-living feeding, feeding tests, and field associations. We have
omitted many records where a single beetle was seen or collected on a plant, except for
a few cases where the beetle was seen actively feeding.
Beetles were identified by the senior author (Criocerinae, Cryptocephalinae, Lam
prosominae, Eumolpinae) and the following specialists: Catherine N. Duckett (Uni
versity of Puerto Rico, Alticini), Vilma Savini P (Universidad Central de Venezuela,
Alticini), David G. Furth (U.S. Natural History Museum, Alticini), Shawn M. Clark
(West Virginia Department of Agriculture, Galerucini), Charles L. Staines (Maryland
Department of Plant Protection, Hispini), and Edward G. Riley (Texas A&M Univer
sity, Cassidini). Plants were identified by the authors and Quirico Jimenez (INBio),
Nelson Zamora (INBio), and Pablo Sanchez (Museo Nacional de Costa Rica).

Florida Entomologist 80(3)

Our data are organized into a table with three supplementary appendices. Table 1
lists observations by chrysomelid taxon, gives field data in summary form, and lists
voucher specimens. Appendix 1 is a key to plant family name abbreviations. Appendix
2 gives the full localities for locality codes used in Table 1. Appendix 3 gives miscella
neous field observations, as well as relevant literature citations for many of the chry
somelid taxa. In Table 1 we have followed the higher classification of Reid (1995)
which reduces several well-known subfamilies to tribal status and confirms earlier
opinions (eg. Crowson 1955, Lawrence 1982) that Bruchidae, or seed weevils, are a
subfamily of Chrysomelidae. Bruchinae are not included in this report; for informa
tion on their host associations, see Janzen (1980a), Johnson (1990), and literature ci
stations therein. While not all workers fully agree with all aspects of Reid's
classification, it represents the latest and most comprehensive phylogenetic arrange
ment of the Chrysomelidae. For differing views, see Kingsolver (1995), Verma & Sax
ena (1996), and Reid (1996).

Explanation of Table 1
Leaf Beetle

Scientific names follow Wilcox (1983) and Flowers (1996). In a few cases, approxi
mate species identifications are indicated by "nr." before the species name: e.g., Pla
giodera nr. uniforms. In some cases only generic identifications were possible, and
distinct morphospecies are numbered as such.


Names follow current usage in the Costa Rica National Herbarium and in the bot
any department at INBio. In cases where species identification is approximate, the
term "cf." is used (e.g., Solanum cf. torvum).

Plant Family

Classification follows the listings of the Flora of Costa Rica by the Missouri Botan
ical Garden and INBio, viewable on the World Wide Web at http://cissus.mobot.org/
manual.plantas/lista.html. Families are coded by initial letters of their family
names. See Appendix 1 for full listing.


A, adult; L, larva; P, pupa


See Appendix 2 for full locality data.


Date of initial collection is given in cases where beetles were reared from larvae or
held for testing.

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records


DHJ&WH: Daniel H. Janzen & Winnie Hallwachs
RWF: R. W. Flowers
Names of other collectors are given as they appear on voucher data labels.


This is an attempt to objectively communicate our level of confidence that an ob
served association involved actual feeding by the chrysomelid.
6 Chrysomelids were observed in the field actually eating plant material.
5 Chrysomelids fed on plant when confined.
4 10 or more chrysomelids were collected from a plant and feeding damage that
could reasonably be attributed to the beetles was present.
3 Five to nine chrysomelids were collected from a plant and feeding damage that
could reasonably be attributed to the beetles was present, or 10 or more chry
somelids were collected from a plant but obvious feeding damage attributable to
the beetles was not present.
2 Two to four chrysomelids were collected from a plant and feeding damage that
could reasonably be attributed to the beetles was present, or five to nine chry
somelids were collected from a plant but obvious feeding damage attributable to
the beetles was not present.
1 Two to four chrysomelids were collected from a plant but no noticeable feeding
damage was observed.

Number (No.)

Number of vouchered specimens. In general, one feeding record equals one
voucher; the few exceptions are mentioned in the Note column.


Specimens collected by the senior author have voucher codes in the form "(Collec
tion No.) RWF(Year)" and are deposited in INBio. Those collected by the junior author
have codes in the form "(Year)-SRNP-(Number)" and are nominally specimens of IN
Bio but are on temporary loan to the University of Pennsylvania.


These are numbered consecutively and appear in Appendix 3.

Appendix 2. Localities

Localities cited in Table 1 are listed on an approximate north-south gradient. The
first letter of each locality code corresponds to the first letter of its province. Localities
in the Area de Conservaci6n Guanacaste also include Lambert Coordinates in paren
theses. Lambert Coordinates are used in Costa Rica in preference to latitude-longi
tude because the 1:50,000 topo sheets are gridded with Lambert Coordinates and,
being metric, Lambert positions are easier to use.

Florida Entomologist 80(3)


The data presented in these tables represent only the beginnings of the task of
working out host plant relationships for the Central American Chrysomelidae. Our
data cover less than 7% of the estimated 2000 chrysomelid species present in Costa
Rica alone (Flowers, unpublished data). In some cases, our data confirmed previously
published relationships between chrysomelid genera and host plant families (summa
rized in Jolivet & Hawkeswood 1995); 30 of our records represent host plant family
range extensions, and 19 records are for chrysomelid genera in which, apparently, no
host plants have been recorded previously.
Most previously published host plant studies for the Neotropical Chrysomelidae
(aside from focused studies on specific taxonomic groups, (e.g., Bach 1986; Begossi &
Benson 1988; Windsor 1986) make no distinctions between accidental or casual asso
citations of plant and beetle and true host relationships. The dangers in not making
these distinctions have been demonstrated to us on several occasions when we found
chrysomelid species that move off their food plants for resting or defecating. An exam
ple is Omophoeta simulans (Alticini, see Table 1), a group of which was first observed
sitting on leaves of a Luehea sapling (Tiliaceae). Although large numbers of beetles
were on the Luehea, and their frass was also evident on these leaves, closer inspection
revealed that no feeding was taking place on the Luehea and that the true food plant
(Evolvulus nummularis; Convolvulaceae) was growing beneath the shrub. Similar
warnings about possible confusion of Alticini food plants due to the beetle's mobility
have been given by Hawkeswood and Furth (1994). Nevertheless, collection records
can still provide useful information-for many taxa opportunistic collecting has pro
vided the only information we have on possible host plants-if their limitations are
clearly acknowledged. For our data we have included a "reliability scale" to roughly
measure our confidence that a given association represents a true chrysomelid-host
plant relationship. While ecological studies of narrow groups of chrysomelids or
plants will always provide the most unambiguous data on feeding requirements, re
cent emphasis on and support for inventory collecting can rapidly increase knowledge
of the feeding habits of a broad range of chrysomelids, if observations are qualified in
some manner.
We intend to continue expanding on the present work, and we encourage other col
lectors of Chrysomelidae to record, categorize and publish the plant associations they
observe. Rapidly expanding our knowledge of chrysomelid-plant interactions is im-
portant for two reasons. On the practical side, knowing host plants for more chry
somelid species will facilitate programs in chemical prospecting which are currently
focused on plants. When a family of plants is being surveyed for active chemicals, the
insects feeding on those plants represent another level of chemical derivatives avail
able for screening. The phytophagous insect may produce novel chemical varieties
which cannot be synthesized directly from the host plant.
A second area where more host plant data are needed is in the testing of hypothe
ses of the evolution of host plant selection. At present there are two competing theo
ries of what chiefly influences this evolution: phylogenetic and ecological mediation.
Phylogenetic mediation (cospeciation) postulates that most cases of herbivory arise
from cospeciation or parallel descent. This theory has become a popular explanation
of host plant selection, under the name "coevolution" (though we caution the reader
that this is not the original meaning of the word, see Janzen 1980b). Phylogenetic me
diation has been demonstrated in the Chrysomelidae for Phyllobrotica species (Gal
erucinae) and their hosts in the Lamiales (Farrell and Mitter 1990). However, their
study represents one of the few documented examples of coevolution (Anderson 1993).

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records


i- t- Cn m LfM

N-c aNa -aN

c c vcc
0 - i CD LOf CD CD

7c C-

. N

Lf LfL
r C COC Lfl C<

^ U 0

< aac .: L

2 2 .

C- H

mWL ~ o o:~5
H O '- "C on

Florida Entomologist 80(3)

September, 1997

o 0

- N

- c00


c' c'

~ ~ :
LLI [Li L0











t- 00


Nt- t
at o

- C

0 0


R 3
z c

t in

. iS


I ;

-> -
(^ (

;3 5 3 t

- CC


0 0 00 0

< < < < <

-f -nOn on^
[i O- On < <


; -,o

t o

B 6~



c -


Flowers & Janzen: Chrysomelid Feeding Records



c' -~ i

- LO

CD CD 1'^ 1'^ CD0 ^ CD CD Lf C'j CD









Lflb~ b' ~ C '
-O c'J
0 o i

H H H HH 0



'e" IE Vn
S's-o c u\ ifU if I sie "it,^5 ,-

E~r 3 G^Gszs~a ^^S u u
'2 a 2a "

"s -

^~ 1 I
E .

3y~ 3r 3ri 3ro G
I+ 1 ^^ ll
0,S J i]




- LO Lf

at t
- a



H L-

N- -

< <
0 0

bbbbbbbb bbb

Florida Entomologist 80(3)

- C

at at

) c)
00 00


cI ii c C f LO CC Lto



3: 3:

0 -


~' Lf Lfl b'


L N r 4 C

0~nr ~: Q

00 00 ;'

C~C -
cc ccc
at ata

c'Jc'J c

- C
at a

C -

C' C1 -1 -1 C

0 0 0 0 0<

< < < 0 >

U O].




z 0

~ A ) . _>

3 -E 3g s? .- S s

Ij Ib l Is l
c; C- E g g R 'i

o o, v, V) ~3. ^ a o ) g N ^ s ^ fi g E
^ s ^. ^- '- '- C ^30^ *

-0 Z

z F ai

a' ^ ?^ *J

23: 4 t
-Q G ^ t ^

September, 1997



Flowers & Janzen: Chrysomelid Feeding Records


- Lf Co Cf


.IOn ~
C4 *03


2~~ ~
C4L2j e

cc 2
3~0 ~00

< < < < <


t t


3 -c
-(o ma , _
vl dicy.~v,$ 8i ~>

d- C-
-l c'
E- d-


-0 -0b h



C\3 -


|: I



K, K,

I I:



Ro R
0, 0,
0, 0.~e

344 Florida Entomologist 80(3) September, 1997


cj M c.0 cj M c0 I M
x o 00Lfl en L C C
Ci tn CD nt CM CD b' LO en LO LO

S oo oo oo oo oo en oo oo

< uC C CC C C C
S CC C C t- C< < Lf o
M 2


7 D
Cfl 3 C i t S C
to cc

-i 1t 1 CC CC

3 v 3

on o 0 0


< on

< S ^ M M

7 -- S- j

|<^ 0, 1

Flowers & Janzen: Chrysomelid Feeding Records

^- CMJ



-- o, c co~

< b <
d~< bb

g -t




o at C'OtCD

cf cD co~


[L ~ [IL


-1 C C1 OtC

Qr o < C

on U

^ -c
s as ^ s

s- 0,, S

*c d, FT s s-

| I o E s

i I
1^1t11 11

Florida Entomologist 80(3)

September, 1997

t-- '00
c' c'

- C4


a0 O


00 00 00 0

CD b~b CD CD CD b' CD










Lfl D

CM CM "" M 1

|^1^^~~ C

c' c'J lv

< << < < < <

O ^ z
20 02
M c2 U Q4

-c c
C '-'
Ion ~ 3 s
-> ~ K>a
~ 3 I~ n

M i


vt 0,'*-
S-|3 ;t

at a




<1 <

a c.


I d 0


Flowers & Janzen: Chrysomelid Feeding Records


00 00
00 CD 00
C' )' 0' CD CC C






< <
cD a

- a

at -


N' at
at at
at -

C'J C,

- N- at 2 -
u u 0 0 0 0

< < < < < < <

on On O2^^ O

S. C L. CO L L

**- i- -0
-N N

? ^--,- -,:i
( b bfl.
5rrz~o (D -
*^ -"




rf : '
0 ^ C

mn Lf , LO tI-





2 ~,
t~L2~ ~

-- -

< < <

^0 0 M~r 2 r C/ C2

~2n n

9 |g g
^ u ,C. 0 ,-a
a :

B &- .
0, Q E >
5 ?
ZS toni

c'J ?

(j - ^
Q -~,-4; 4; 3

-t (^ ,3 .
cu a -

Florida Entomologist 80(3)

t- 00
m co c

00L ^



LO 00 C i N CD
\ c.0

00 N-





at cj c' cJ '


2; m~
aml2~ e

0 i 0 < o

~ <<< < <


t- CD

-i ^f CM Cn

a at at

c' ~ cJ
CD~ -

< U b-








i -s -i-r a

-5 c'J5

+( 4 u rum

a ~ ~ Z E ~ Z4c;




1 i I *
0, $s,

(c ~ c
-, B
:3 ; i

5c o- >'j

September, 1997



<< H )HO 2
M n W o o no w 2


c' '

Flowers & Janzen: Chrysomelid Feeding Records

0 CM


CD, -C' Lfl -

cn c'j '

at at
- tjl
- at



5 2-5 2- cl"
e: C.Th C.> e
3< 3<

at a
at a

< 250 0 0

< << < <

O o< < <
c O 0 0



0 0

<< < < < < <

cOn o
m< < 2

< 0

ZI Cll N ^ .
2 C

4 t
s *i | sc a w 1 s- g |
s s- *

-1 I& J ilt U. 1 IE il l f
Q 2 aa C
I C-K; i 1 i

^" m ff iaia ^ ^
Q-, -c'

cc $4 t~, t ,-,
^- -"s -
I~~~ ~~ C"t %*

0, 1^ I g^
3+ ^- 'fl d^ ^G'

Florida Entomologist 80(3)

September, 1997

m Lfl

t- LO b'
- CM

SM Lo C c i MO L CJ N


.5 y


c'J~c C D C
~~-C 0C- -1 -1 -1 -

0o Q5 ~ 3 0

< << << << < < d< <<<












^ -a 2 o
-). c'
s I ^ ^s `s



v3 v3 v3 a S
*- (
z ^0v
, ^ ^ '

- 0 -



^ -t


g- .c'J
n i^

"^ ^^^S

22 com f

Flowers & Janzen: Chrysomelid Feeding Records

CD tN-

a a

m t
ati Lfl
Cfl Lfl

a a




CD N' CJ Cl J CD Lfl b CD




at -n

I cI

c' -

Lnat -
0)at at
at- at



at at

c'J -

c' c' c' N < 0 0

^~~ d d M a a"


12, 12
on < < << 2 2< <

-c |*
i C.on Kg "zi i 1
^ C. c C^ 1 1 1

0, -3 F 1 a
*-, ^ ^ 3 a

C ; ;- s
i" 1 ^ ^ ".

--| 10,2 i -
4c .| .2^ 1^^ 3:B3
il 11
~z~E"'"~ ~ ~ c n



Florida Entomologist 80(3)


September, 1997


CD t- 00 C' ^f CD -




-~t C -
cn~N c' z
0 o0 0

<~~~3 << < <











Hw- <2 0
ono 0 O

3 2

g .^ .^ a

;"R ^ -9 8R 3 .

o cj c--s 3 1/

| .( ^" ^ .

; I

+c~ d



- c'



Flowers & Janzen: Chrysomelid Feeding Records

Lf Lfl Lf



b' b' CD Lfl C CD


at O
atat at
atat -

o Lf C C'N
N cn cj '
o o < < 0 Oo o1

< < ~< < <<

Z <
N a

Ha <

3 2

g -- b ^ (

0, v
0, w ,
,-, G -(: U 2

^ 1 I 01,Q1

1^1^^ iilU

0 0

aci at
0) 0)


2 m


o o

yi -

h o

Florida Entomologist 80(3)

September, 1997

CD t-
Lo Lo

CM M ^ CIA - CD -i M Cf O0 CIA CD LO Ot


SO 0 O LO i 0i o 000
35 N N 5

< <



0 000D
4 rf r f rf

om 0

co2 2S -. a
S S .-.K.

(i-i ^~ ,z ^ i
~c J
SS C -a e a

a ~ a i g- s"

*| i; 3 ^ ,5
i; It li 1^ 1^:o

*^ 5S- US U (ff Uf







Flowers & Janzen: Chrysomelid Feeding Records



9~ E

o o o 65o5

^ S ^ ^ ^ 3 i i
0, I -^ -
s & i a g a 1 o $ 5

C ;-3 ^C ; 2 ^ 2^ ^
`5 ca c i ^ ^ ^ a ca
ro d oao 3 ooo

I 3.

PiC ? mm
^ a

; ^S-
;^ls 1 ^

n o G

0 o

4 4


. I
s *
.3 | c .
^* R.
-S $ ah

Florida Entomologist 80(3)

tN- -



t- LO


a h

o- a


< ono



C- r
oi0 P

'- S n -

0, ^
cm c -- E

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records

The alternative hypothesis is that ecological mediation (colonization and host
transfer) is the primary explanation for current host associations. In cases of ecologi
cal mediation, phylogenies of herbivores and host plants are not congruent, and host
shifts are not necessarily between sister taxa of plants Anderson (1993). In a survey
of the Curculioninae (Curculionidae), Anderson (1993) found that in taxa where sys
tematics and plant associations were reasonably well known, evidence for cospecia
tion of plant and insect taxa is lacking, and ecological mediation appeared to be the
rule. However, like the Chrysomelidae, the majority of curculionine taxa lack any host
plant data. Until host plants are known for a much larger proportion of phytophagous
insect taxa, speculations on the evolution of host plant selection by insects will con
tinue to be based on small subsets of the phytophagous insect universe.


We sincerely thank the staffs of the Area de Conservaci6n Guanacaste (ACG), and
the Instituto Nacional de Biodiversidad (INBio) for their assistance and many kind
nesses during the course of this study. This research was funded in part by a grant
(FLAX 91005) from the CSRS, USDA, to Florida A&M University, a National Science
Foundation Mid-Career Fellowship (BSR-9003898) to the senior author, and NSF
DEB-9400829 to the junior author.


ANAYA R., S. 1987. Chrisomelinos (Coleoptera: Chrysomelidae) del Valle de Mexico.
Centro de Entomologia y Acarologia, Colegio de Postgraduados, Chapingo,
Mexico. 84 p.
ANDERSON, R. S. 1993. Weevils and plants-phylogenetic versus ecological mediation
of evolution of host plant associations in Curculioninae (Coleoptera, Curculion
idae). Mem. Entomol. Soc. Canada, No. 165: 197-232.
BACH, C. E. 1986. A comparison of the responses of two tropical specialist herbivores
to host plant patch size. Oecologia 68: 580-584.
BECHYNE, J. 1955. Troisi6me note sur les Chrysomeloidea N6otropicaux des collec
tions de 1'Institut Royal des Sciences Naturelles de Belgique (Col. Phytophaga),
deuxi6me parties (1). Bull. Inst. r. Sci. nat. Belgique 19: 128.
BECHYNE, J., AND SPRINGLOVA DE BECHYNE. 1975. Notas sobre la series fil6tica de Mo-
nomacra y sus formas convergentes (Col. Phytophaga, Alticidae). Rev. Fac.
Agron. (Maracay), 8: 25-140.
BEGOSSI, A., AND W. W. BENSON. 1988. Host plants and defense mechanisms in Oedi
onychina (Alticinae), pp. 57-71 in Biology of Chrysomelidae, P. Jolivet, E. Petit
pierre and T H. Hsiao, eds. Kluwer, the Netherlands. 640 pp.
BuzzI, Z. J., AND J. A. WINDER 1986. Stages and life cycle of Drepanocassis profana
(Boh., 1855) (Coleoptera, Chrysomelidae, Cassidinae) on Hyptis suaveolens (L.)
Poit. (Labiatae) in Brazil. Revta. brasilense Entomol. 30: 31-41.
CROWSON, R. A. 1955. The natural classification of the families of Coleoptera. London.
187 pp.
FARRELL, B. D., AND C. MITTER 1990. Phylogenesis of insect/plant interactions: have
Phyllobrotica leaf beetles (Chrysomelidae) and the Lamiales diversified in par
allel? Evolution 44: 1389-1403.
FLOWERS, R. W. 1991. Aggregations of Cassidinae (Chrysomelidae) in Santa Rosa and
Guanacaste National Parks, Costa Rica. Biotropica, 23: 308-310.
FLOWERS, R. W. 1996. La subfamilia Eumolpinae (Coleoptera: Chrysomelidae) en
America Central. Rev. Biol. Trop., Pub. esp. No. 2: 1 60.
FLOWERS, R. W., D. G. FURTH, AND M. C. THOMAS. 1994. Notes on the distribution and
biology of some Florida leaf beetles (Coleoptera: Chrysomelidae). Coleopts Bull.

Florida Entomologist 80(3)

FURTH, D. G. 1985. Some flea beetles and their foodplants from Kenya (Chrysomel
idae: Alticinae). Coleopts Bull. 39: 259-263.
HAWKESWOOD, T. J., AND D. G. FURTH. 1994. New host plant records for some Austra
lian Alticinae. Spixiana 17: 4349.
HILJE Q., L., C. ARAYA F., AND R. SCORZA R. 1991. Plagas y enfermedades forestales
en America Central; guia de campo. Serie T6cnica, Manual T6cnico 4, Centro
Agronomico Tropical de Investigacion y Ensenanza (CATIE), Turrialba,
262+xii pp.
HSIAO, T. H. 1988. Host specificity, seasonality and bionomics of Leptinotarsa beetles,
p. 581-599 in Biology of Chrysomelidae, P. Jolivet, E. Petitpierre and T H.
Hsiao, eds. Kluwer, the Netherlands. 640 pp.
JANZEN, D. H. 1980a. Specificity of seed-attacking beetles in a Costa Rican deciduous
forest. Journal of Ecology 68: 929-952.
JANZEN, D. H. 1980b. When is it coevolution? Evolution 34: 611-612.
JANZEN, D. H. 1993. Caterpillar seasonality in a Costa Rican dry forest, pp. 448-477
in Caterpillars: Ecological and evolutionary constraints on foraging, N. E.
Stamp and T M. Casey, eds. Chapman and Hall, New York,
JANZEN, D. H., AND I. D. GAULD (in press). Patterns of use of large moth caterpillars
(Lepidoptera: Saturniidae and Sphingidae) by ichneumonid parasitoids (Hy
menoptera) in Costa Rican dry forest, in Forests and Insects, A. D. Watt, N. E.
Stork and M. D. Hunter, eds., Chapman and Hall, London
JOHNSON, C. D. 1990. Systematics of the seed beetle genus Acanthoscelides
(Bruchidae) of northern South America. Trans. American Entomol. Soc. 116:
JOLIVET, P. 1977. Selection trophique chez les Eupoda (Coleoptera Chrysomelidae).
Bull. Soc. Linn. Lyon 46: 321-336.
JOLIVET, P. 1978. Selection trophique chez les Clytrinae, Cryptocephalinae et Chlam
isinae (Camptosoma) et les Lamprosomatinae (Cyclica)(Coleoptera Chrysomel
idae). Acta Zool. Path. Antverpiensia 70: 167-200.
JOLIVET, P. 1979. Les Chrysomelidae (Coleoptera) des Citrus et apparentes (Ruta
ceae) en zone temper6e et tropical. Bull. Mens. Soc. Linn. Lyon 48: 197-256.
JOLIVET, P. 1982. Les Eumolpinae (Col. Chrysomelidae) des Apocynaceae et des Ascl
epiadaceae (Gentianales). Bull. Mens. Soc. Linn. Lyon 51: 214-222.
JOLIVET, P., E. PETTITPIERRE, AND M. DACCORDI. 1986. Les plantes-h6tes des Chry
somelidae. Quelques nouvelles precisions et additions (Coleoptera). Nouv. Re
vue Entomol. 3: 341 357.
JOLIVET, P. 1987a. Apercu de la selection trophique chez les Galerucinae. Etude par
genre (Coleoptera Chrysomelidae). Bull. Ann. Soc. r. belge Entomol. 123: 283
JOLIVET, P. 1987b. Selection tropique chez les Megascelinae et les Eumolpinae (Cy
clica)(Coleoptera Chrysomelidae). Bull. Mens. Soc. Linn. Lyon 56: 199-208,
JOLIVET, P. 1988a. Selection trophique chez les Cassidinae (Coleoptera Chrysomel
idae). Bull. Mens. Soc. Linn. Lyon 57: 301-320.
JOLIVET, P. 1988b. Food habits and food selection of Chrysomelidae. Bionomic and
evolutionary perspectives, pp. 1-23, in Biology of Chrysomelidae. P Jolivet, E.
Petitpierre and T H. Hsiao, eds. Kluwer, the Netherlands. 640 pp.
JOLIVET, P. 1989. Selection trophique chez les Hispinae (Coleoptera Chrysomelidae
Cryptosoma) Bull. mens. Soc. linn. Lyon, 58: 297-317.
JOLIVET, P. 1991. Selection trophique chez les Alticinae (Coleoptera Chrysomelidae).
Bull. mens. Soc. linn. Lyon, 60: 2640; 53-72.
JOLIVET, P., AND T. J. HAWKESWOOD. 1995. Host-plants of Chrysomelidae of the
world. Backhuys, Leiden. 281 pp.
KING, A. B. S., AND J. L. SAUNDERS. 1984. The invertebrate pests of annual food crops
in Central America. Overseas Development Administration. London. 166 pp.
MEMMOTT, J., H. C. GODFRAY, AND B. BOLTON. 1993. Predation and parasitism in a
tropical herbivore community. Ecological Entomology 18: 348-352.

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records

MOLDENKE, A. R. 1970. A revision of the Clytrinae of North America north of the Isth
mus of Panama (Coleoptera: Chrysomelidae). Stanford University.
MOLDENKE, A. R. 1971. Host-plant relations of phytophagous beetles in Mexico. Pan
Pacific Entomol. 47: 105-116.
MONROS, F. A. 1949. Descripci6n de las metamorfosis de Lamprosoma chorisiae Mon
r6s y consideraciones taxon6micas sobre Lamprosominae (Col. Chrysomel
idae). Acta Zoo. Lill. 7: 449-466.
OSTMARK, H. E. 1975. Banana pests in the genus Colaspis including description of a
new species (Coleoptera: Chrysomelidae). Florida Entomol. 58: 18.
REID, C. A. M. 1995. A cladistic analysis of subfamilial relationships in the Chry
somelidae sensu lato (Chrysomeloidea), pp. 557-631 in Biology, Phylogeny, and
Classification of Coleoptera: Papers Celebrating the 80th Birthday of Roy A.
Crowson, J. Pakaluk and S. A. Slipinski, eds. Muzeum I Instytut Zoologii PAN,
REID, C. A. M. 1996. More on the family Bruchidae. Chrysomela Newsletter 31: 3.
C. JUMA. 1993. Biodiversity prospecting: using genetic resources for sustain
able development. World Resources Institute. Baltimore, MD. 341+ix pp.
ROCKWOOD, L. L. 1974. Seasonal changes in the susceptibility of Crescentia alata
leaves to the flea beetle Oedionychus sp. Ecology 55: 142-148.
SCHRODER, W. W., B. PUTTLER, S. S. IZHEVSY, AND D. GANDOLFO. 1994. Viviparity and
larval development of Platyphora quadrisignata (Germar) in Brazil. Coleopts
Bull. 48: 237-243.
STRAUSS, S. Y. 1988. The Chrysomelidae: a useful group for investigation of herbi
vore-herbivore interactions, pp. 91 106 in Biology of Chrysomelidae, P Jolivet,
E. Petitpierre and T H. Hsiao, eds. Kluwer, the Netherlands. 640 pp.
STRONG, D. R. 1977a. Rolled-leaf hispine beetles (Chrysomelidae) and their Zingib
erales host plants in Middle America. Biotropica 9: 156-169.
STRONG, D. R. 1977b. Insect species richness: hispine beetles of Heliconia latispatha.
Ecology 58: 573-582.
VERMA, K. K., AND R. SAXENA. 1996. The status of Bruchidae a a family Chrysomela
Newsletter 32: 3.
WILCOX, J. A. 1983. Checklist of the beetles of North and Central America and the
West Indies. Vol. 8. The Leaf Beetles and the bean weevils. Family 129. Chry
somelidae. E. J. Brill. New York, 166 pp.
WINDSOR, D. M. 1986. Natural history of a subsocial tortoise beetle, Acromis sparsa
Boheman (Chrysomelidae, Cassidinae) in Panama. Psyche 94: 127-150.
WINDSOR, D. M., E. G. RILEY, AND H. P. STOCKWELL. 1992. An introduction to the bi
ology and systematics of Panamanian tortoise beetles (Coleoptera: Chrysomel
idae: Cassidinae), pp. 372-391 in D. Quintero and A. Aiello, eds. Insects of
Panama and Mesoamerica. Oxford Univ. Press, Oxford, 692+xxii pp.

Florida Entomologist 80(3)


ACA Acanthaceae CNV Convolvulaceae MYR Myrsinaceae
AMA Amarantaceae DIO Dioscoreaceae ONA Onagraceae
APO Apocynaceae ERY Erythroxylaceae PAS Passifloraceae
ASC Asclepiadaceae ERI Ericaceae PIP Piperaceae
AST Asteraceae EUP Euphorbiaceae POA Poaceae
BIG Bignoniaceae FAB Fabaceae: POL Polygonaceae
BOR Boraginacaea Papilionoidea RUB Rubiaceae
BUR Burseraceae FLA Flacourtiaceae RUT Rutaceae
CAE Fabaceae: HIP Hippocrateaceae SPI Sapindaceae
Caesalpinoidea LAU Lauraceae SPO Sapotaceae
CAP Capparidaceae LOG Loganiaceae SIM Simarubaceae
CLU Clusiaceae MLP Malpighiaceae SOL Solanaceae
CEC Cecropiaceae MLV Malvaceae STE Sterculiaceae
COC Cochlospermaceae MAR Marantaceae URT Urticaceae
COM Combretacaea MEL Melostomataceae VER Verbenaceae
CON Connaraceae MIM Fabaceae: VIT Vitaceae
Mimosoidea ZIN Zingiberaceae

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records


G1 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Pitilla,
Estacion Pitilla, 8 km S Santa Cecilia, 700 m (N330000, E380400)
G2 Guanacaste Prov., Area de Conservacion Guanacaste, Sector El Hacha,
Cerro el Hacha, 10 km SE La Cruz, 300 m (N331700, E365400)
G3 Guanacaste Prov., Area de Conservacion Guanacaste, Area Recreativa
Junquillal, 3 km N Cuajiniquil, 0 m (N328000, E351700)
G4 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Orosi, Esta
cion Maritza, 20 km SE La Cruz (N326500, E372200)
G5 Guanacaste Prov., Area de Conservacion Guanacaste, Estacion Pocosol, 20
km S La Cruz, 250 m (N319000, E361100)
G6 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Orosi, Esta
cion Maritza, sendero Casa Fran, 21 km SE La Cruz, 600 m (N326000,
G7 Guanacaste Prov., Area de Conservacion Guanacaste, Estacion Santa
Rosa, 28 km NNW Liberia, 250 m (N313700, E359000)
G8 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Bosque Humedo, 30 km NNW Liberia, 300 m (N314800, E360500)
G9 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Cafetal, 31 km NNW Liberia, 300 m (N315500, E360200)
G10 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Bosque San Emilio, 29 km NNW Liberia, 300 m (N313800, E359800)
G11 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Sendero Natural, 28 km NNW Liberia, 250 m (N313100, E359900)
G12 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Finca Rosa Maria, 26 km NNW Liberia, 250 m (N311000, E359500)
G13 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Casona, 28 km NNW Liberia, 250 m (N313000, E359900)
G14 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Area Administrativa, 29 km NNW Liberia, 250 m (N313500, E358900)
G15 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Cliff Top Light, 31 km NNW Liberia, 300 m (N315200, E360200)
G16 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Casetilla Entrada, 33 km NNW Liberia, 300 m (N317800, E362600)
G17 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Laguna Escondida, 30 km NNW Liberia, 250 m (N314500, E357900)
G18 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Llano Guacimal, 32 km NNW Liberia, 300 m (N317000, E361600)
G19 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Canyon del Tigre, 18 km NW Irigaray, 200 m (N310000, E356800)
G20 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Naranjo,
Playa Naranjo, 0 m (N307000, E354500)
G21 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Naranjo, Sen
dero Real, 10 m (N309000, E354000)
G22 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Cruz de Piedra, 33 km NNW Liberia, 300 m (N317200, E360900)
G23 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Cacao,
EstacionCacao, 9 km N Quebrada Grande, 1000 m (N323100, E375500)

Florida Entomologist 80(3)


G24 Guanacaste Prov., Area de Conservacion Guanacaste, Sector Santa Rosa,
Finca Jenny, 30 km NNW Liberia, 200 m (N316200, E364200)
G25 Guanacaste Prov., Area de Conservation Guanacaste, Sector Santa
Rosa,Vado Rio Poza Salada, 17 km NW Irigaray, 10 m (N308900, E355700)
G26 Guanacaste Prov, Area de Conservacion Guanacaste, Sector Santa Rosa,
Quebrada Guapote, 27 km NNW Liberia, 200 m (N312700, E361700)
G27 Guanacaste Prov, Area de Conservacion Guanacaste, Sector Santa Rosa,
Quebrada Costa Rica, 250 m (N312200, E357500)
G28 Guanacaste Prov, Potrerillos, Rio Tempisque, 23 km NNW Liberia, 100 m
(N310900, E367400)
G29 Guanacaste Prov, Finca La Pacifica, 5 km NW Canas.
G30 Guanacaste Prov, 12 km NW of Bebedero, Hacienda Horizontes.
G31 Guanacaste Prov, Bebedero, Ingenio Taboga.
Al Alajuela Prov, Finca San Gabriel, 2 km SW Dos Rios, 600 m
A2 Alajuela Prov, Reserva Forestal San Ramon, 900 m
A3 Alajuela Prov, Bijagua, 20 km S Upala, 500 m
A4 Alajuela Prov, Canton La Guacima, Rio Segundo, 780 m
A5 Alajuela Prov, Canton Ciruelas, Rio Ciruelas, 800 m
H1 Heredia Prov, Estac. Biol. La Selva, 50 m
S1 San Jose Prov, San Pedro, Univ. Costa Rica
S2 San Jose Prov, El Rodeo, 1.5 km S Ciudad Colon
Cl Cartago Prov, Pavones, nr. Turrialba
C2 Cartago Prov, Madreselva, nr. Empalme
C3 Cartago Prov, Carretera Interamericana, 7 km S. Cartago
C4 Cartago Prov, Cerro Asuncion, paramo vegetation, 3396 m
C5 Cartago Prov, Tapanti, Refugio Vida Silvestre
P1 Puntarenas Prov, Reserva Forestal Monteverde
P2 Puntarenas Prov, Peninsula de Osa, Est. Boscosa, Reserva Forestal Golfo
P3 Puntarenas Prov, Peninsula de Osa, Cerro de Oro

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records

Appendix 3-Notes to Table 1.

1. These beetles were sitting on heavily eaten leaves, 1.5 m above ground.
2. Additional specimens were observed at time of collection, and C. Chavez reported
seeing this species frequently on the same host plant.
3. This species was tested on the host plant. Jolivet (1978) gave Asteraceae, Mimo
saceae, Ericaceae and Fagaceae as other host plant families of this genus.
4. This species was found feeding at shoot tips of its host plant.
5. The host plant is an abundant roadside weed on the entrance road in Sector
Santa Rosa and elsewhere in this sector. Beetles have been collected both in the
rainy and dry seasons. The larvae make cone-shaped cases, apparently utilizing
hairs of the host's leaves. Moldenke (1971) listed both Malvaceae and Convolvu
laceae as host plant families for this species.
6. The vouchers were collected from a swarm of this species feeding on the low bush
in dense dry forest. The intense feeding and mating activity was similar to that
observed in other Clytrinae (Flowers et al. 1994, Moldenke 1971). Jolivet (1978)
lists Mimosaceae as the predominant host for this genus.
7. The beetle was seen eating bark of new stems. Monr6s (1949) and Jolivet (1978)
described bark feeding by other members of this genus.
8. This species was very abundant on the leaves of its host at several regenerating
pasture sites in 1991. This cosmopolitan genus has been recorded from Arali
aceae from the Palearctic and from Myrtaceae from Puerto Rico (Jolivet 1978).
9. In 1991 this species was very abundant in the pastures and open areas after the
onset of the summer rains. Individuals were also collected on other pasture
shrubs. The collection of Jan Bechyne in Maracay Venezuela contains several
specimens of this species collected in El Salvador and bearing the (apparently)
manuscript name "saltator". The only other host record for this genus is Theo
broma cacao L. (Sterculeaceae) for an unidentified species (Jolivet 1987b).
10. Jolivet (1987b) stated that all host observations of the genus Chalcophana have
been Asteraceae.
11. Although only one voucher was preserved, numerous adults were observed, and
several were tested on the leaves of the plant host. Jolivet (1987b) noted that this
genus is both cosmopolitan and polyphagous.
12. Adults were feeding at night on very new expanding leaves of a 1.5 m shoot at
base of tree. Jolivet (1987b) stated that the only reliable feeding records for this
genus are from Fabaceae.
13. The only host records in the literature for Percolaspis are from Poaceae and Theo
broma cacao (Jolivet 1987b).
14. This species has been found feeding on several species of Rubiaceae. Adults are
agile leapers when disturbed. Jolivet (1987b) gave a single record for this genus:
Persea (Lauraceae) for a Cuban Phanaeta.
15. Adults of this genus were found on new foliage and in some years defoliated their
16. This species was very common feeding on various species of Melastomataceae.
Jolivet (1987b) described Typophorus as polyphagous but does not list any Melas
tomataceae among its host plants.
17. The voucher is one of many collected, seen and reared at Estaci6n Pitilla and San
Gabriel on various species of Solanum.
18. Larvae skeletonize host plant leaves. This species extensively defoliates its host
during some years. Literature records for New World Plagiodera are limited to
Salix, Populus (Salicaceae), Croton (Euphorbiaceae), and Lueha (Tiliaceae); how

Florida Entomologist 80(3)

ever, species in the Philippines and India have been reported on Xylosoma and
Flacourtia (Flacourtiaceae) (Jolivet & Hawkeswood 1995).
19. This is the most commonly collected of the Costa Rican species of Platyphora.
During one feeding test, two very small larvae were observed in the plastic bag
which up till then held a single female, suggesting that Platyphora bicolor is vi
viparous. Schroder et al. (1994) described the biology of the viviparous Platy
phora quadrisignata (Germar) from southern Brazil.
20. Adults and larvae were frequently found feeding on host plant throughout the
1991 rainy season. Apparently, our observations represent the only known host
plant data for Stilodes.
21. A group was followed from egg to adult. Larvae feed and rest on underside of
leaves. Pupation takes place in leaf litter.
22. Larvae are sooty black, covered with branched hair-like projections, and with red
heads. Pupae are yellow. This chrysomelid was parasitized by Myopharous (Ta
chinidae: Diptera).
23. In 1991 this beetle caused a major defoliation of its host plant, a pioneer species
in cleared pastures.
24. The host plant of this galerucine was found growing along the edge of a small
patch of forest.
25. The host plant was a low understory tree in tropical dry forest.
26. This galerucine was seen on several occasions feeding on young leaves of its host
plant. This genus has been recorded from Acacia (Fabaceae) in the USA (Jolivet
27. A large group of these Masurius (which may represent more than one species)
was found feeding on the two host plants growing within a few yards or each
other along a trail in montane forest.
28. This and the following species were reared to adult.
29. In addition to the voucher specimens, other specimens were collected two years
earlier on the same host plant.
30. RWF has observed adults of this species every year since 1989 defoliating basal
shoots of a tree growing in front of the main administration building at the Uni
versity of Costa Rica. Jolivet (1987a) listed Cordia and Lantana (Verbenaceae) as
hosts of this genus.
31. RWF observed on individual at night eating a hole in the middle of a leaf of the
Ipomoea host plant.
32. In both cases, beetles were observed feeding on the host plant. Jolivet (1991)
listed Labiaceae and Verbenaceae as probable hosts for this genus and noted
other citations of Lauraceae, Buddlejaceae, Asteraceae, Umbelliferae, Sterculi
aceae, and Fabaceae.
33. In addition to the vouchered specimen from Byrsonima crassifolia, this species
was abundant on this host plant at Estacion Maritza (G4) in 1991.
34. Unlike many other chrysomelids which were found associated only with young
foliage, A. salvadorense was found actively feeding late in the rainy season on
older leaves.
35. A large group of these beetles was found on a broken stalk of the host plant, feed
ing on sap and milky latex. The host plant was growing in the shaded understory
of montane forest.
36. Adults were reared from larvae feeding on the host plant.
37. Jolivet (1991) listed Samanea (Fabaceae/pap.) as a host of this genus.
38. Jolivet (1991) cited Theobroma and Tecoma (Bignoniaceae) as other known host
plants of this alticine genus.

September, 1997

Flowers & Janzen: Chrysomelid Feeding Records

39. Both this and the following host plant were growing close together in a mixed
stand next to a road.
40. The host plant, growing in a wet depression in a cleared area, sustained heavy
feeding damage from this alticine in 1991. Jolivet (1991) listed Cleome, Solanum,
Beta (Chenopodiaceae), Labiaceae, Cordia, and Adiantum (Adiantaceae) as host
plants of Leptophysa.
41. These beetles were swept from a tree that showed heavy feeding damage to the
leaves. No active feeding was observed, but this collection was made during an
abnormal dry spell during what was supposed to be the wet season.
42. This Longitarsus is a flightless species.
43. Field observations by DHJ indicate that the adult appears on the host plant to
oviposit; larvae are free living and cut islands out of leaf margin.
44. The host plant is a small prostrate weed. The beetles were first observed resting
and defecating on a shrub of Luehea (Tiliaceae) which grew over the Evolvulus.
When no feeding damage on the Luehea was seen, despite the beetle activity, a
wider search revealed the true host plant.
45. These small pinkish-orange flea beetles were observed feeding on newly expand
ing leaves (which are also reddish to pinkish orange) of their ericaceous hosts.
46. This alticine was collected abundantly from a very dense stand of its host plant.
In 1994 it was found equally abundantly in the same stand of plants.
47. RWF has observed this species over several years, actively feeding on Euphorbi
aceae even during the dry season in quite arid habitats.
48. These represent five different morphospecies of Syphrea collected on various
49. This species feeds by scraping pits in the expanding leaves of this host plant. The
following plant record may be an alternate dry season food source.
50. Huge numbers of this species were found defoliating the host plant during the
voucher year. In 1991, on the other hand, no specimens were found and no dam
age to the host was apparent. This is the species called Oedionychis sp. in Rock
wood (1974). Bechyn6 (1955) restricted the definition of true Oedionychis to a
small group of flightless Mediterranean flea beetles. New World species formerly
in Oedionychis are now placed in Walterianella, Alagoasa and other genera.
51. Jolivet (1991) listed Venezuelan records of Gardinia (Rubiaceae) and Tabebuia
(Bignoniaceae) for this genus.
52. The genus Cephaloleia is well known from various species of Heliconia and other
Zingiberales (Strong 1977a,b). This species was regularly encountered in rolled
up terminal leaves of Costus at this and other localities.
53. This hispine was very abundant in a dense stand of grass growing on a river sand bar.
54. A large number of these hispines were feeding on and heavily damaging leaves
of a shrub of its host growing along the bank of a river in deep shade.
55. These cassids have black larvae with long black caudal brushes; the pupae have
a creamy white thorax. Adults were reared.
56. Windsor et al. (1992) gave Ipomoea lindenii Mart. & Gal. as host plant for true
C. egregia.
57. This species periodically defoliates its host.
58. Feeding on young leaves of Alibertia was seen; some feeding damage was also
seen on the two bignoniaceous plants as well.
59. The cassid caused a major defoliation in 1979, but has been rare since. The 1991
record was from a single tree growing by the seashore and heavily damaged by a
group of the cassids. Jolivet (1988a) also listed Tabebuia and other Bignoniaceae
as hosts for this genus.

366 Florida Entomologist80(3) September, 1997

60. These records are of beetles aestivating in the dry season; see Flowers (1991) for
more details on this behavior. Windsor et al. (1992) listed several species of Cor
dia as the true host plants of this species.
61. The host plant was an understory plant in a pine plantation. Jolivet (1988a) also
listed Hyptis (Labiaceae) as a host plant for this genus.
62. The record from Bursera simaruba is for beetles hiding under bark plates during
the dry season. Jolivet (1988a) listed Phaseolus (Fabaceae) and Passiflora (Pas
sifloraceae) for this genus.
63. In 1991 this species was common during the rainy season. A colony at the Ad
ministration Area in Sector Santa Rosa (G14) was followed for several months,
during which time predatory pentatomids were observed resting on foliage above
the cassids, and occasionally descending to feed on them.
64. Windsor et al. (1992) listed Cordia spinescens for a P nr. alutacea from Panama.
65. Windsor et al. (1992) also list Solanum seaforthianum Andr. and Physalis cor
data Mill (Solanaceae).


Florida Entomologist 80(3)


'USDA-ARS South American Biological Control Laboratory
Bolivar 1559 (1686) Hurlingham, Buenos Aires Province, Argentina

USDA-ARS Center for Medical, Agricultural, and Veterinary Entomology
PO. Box 14565, Gainesville, Florida, 32604, USA


The longevity of colonies of the black imported fire ant, Solenopsis richteri Forel,
and the survival of starved workers and sexual females was compared between
healthy colonies and colonies infected with the microsporidium Thelohania solenop
sae Knell, Allen, & Hazard. The colonies were collected in the field and reared for ap
proximately four mo. Individual workers and sexual were held without food until
death. The body weight of infected and healthy workers was compared. After 3 mo of
laboratory rearing, longevity of infected colonies was significantly shorter than that of
healthy ones; mortality of infected colonies was 92% and mortality of healthy colonies
was 49%. At 27C, mortality rate of workers from infected colonies was higher than in
healthy workers. Workers from infected colonies lived between 8.8 and 29.2% less
than healthy workers. At 22 C, no statistical significance was observed. At 21 C, only
the initial mortality of sexual females was higher in infected than in healthy individ
uals. The weight of infected workers was very similar to that of healthy workers. T so
lenopsae should be considered for the biological control of the imported fire ants in the
United States.

Key Words: Solenopsis invicta, imported fire ants, microsporidium, ant longevity

September, 1997

Briano & Williams: Effect ofT. solenopsae


La longevidad de colonies de la "hormiga colorada" (u "hormiga brava") Solenopsis
richteri Forel y la supervivencia de obreras y hembras sexuadas en inhanici6n fueron
comparadas entire colonies sanas y colonies infectadas con el microsporidio Theloha
nia solenopsae Knell, Allen y Hazard. Las colonies fueron colectadas en el campo y
criadas durante aproximadamente cuatro meses. Las obreras y sexuados fueron man
tenidos sin alimento hasta su muerte. Se compare el peso corporal de obreras enfer
mas y sanas. Despu6s de 3 meses de cria en laboratorio, la longevidad de las colonies
infectadas fue significativamente menor que la de las colonies sanas; la mortalidad de
las colonies infectadas fue del 92% y la mortalidad de las sanas fue 49%. A 27'C, la
tasa de mortalidad de obreras de colonies enfermas fue mayor que la de obreras sanas.
Obreras de colonies enfermas sobrevivieron entire 8.8 y 29.2% menos que las obreras
sanas. A 22C, no se observe significancia estadistica. A 21 C, solo la mortalidad ini
cial de las hembras sexuadas fue mayor en los individuos enfermos que en los sanos.
El peso de las obreras enfermas fue muy similar al de las obreras sanas. T solenopsae
deberia ser considerado para el control biol6gico de la "hormiga colorada" en los Esta
dos Unidos.

The presence of a microsporidian pathogen in the red imported fire ant, Solenopsis
invicta Buren, was first reported by Allen & Buren (1974) from Brazil, and was later
described as Thelohania solenopsae Knell, Allen, & Hazard (1977) (Microsporida: The
lohaniidae). A similar microsporidium was discovered in the black imported fire ant,
Solenopsis richteri Forel, and other Solenopsis species, in Argentina and Uruguay
(Allen & Silveira Guido 1974). The presence of microsporidia was later reported in
surveys of fire ant natural enemies conducted in South America (Jouvenaz 1983,
1986; Jouvenaz et al. 1980, 1981; Wojcik et al. 1987; Briano et al. 1995). A comparative
study conducted by Moser (1995) confirmed that these microsporidia were conspecific.
Thelohania solenopsae is the most common microorganism of fire ants in Buenos
Aires Province, Argentina (Briano et al. 1995). Recently, it was discovered infecting
colonies of S. invicta in the United States (Williams et al. 1997). Briano et al. (1995a,
1995b, 1996) reported for Argentina a high intracolonial prevalence of the infection
and a detrimental effect on native fire ant field colonies and populations. They sug
gested that T solenopsae may be a suitable candidate for the biological control of the
red and black imported fire ant in the United States.
Although Knell et al. (1977) reported that field-collected colonies of S. invicta in
fected with this microsporidium cannot be maintained under laboratory conditions as
long as healthy colonies, this detrimental effect was never quantified. We speculated
that a similar effect of T solenopsae could be expected in S. richteri. Our primary ob
jective was to compare the longevity of field-collected healthy fire ant colonies, and the
survival of individual workers and female sexual, with those infected with T solenop
sae. This work reports the results of laboratory tests conducted since 1992.


Longevity of Colonies

In January 1992, 38 colonies of S. richteri were collected along the roadsides of Rt.
12, km 104, Isla Talavera, Buenos Aires Province, Argentina. This sampling site was
selected based on previous surveys that revealed high prevalence of T solenopsae
(Briano et al. 1995).

Florida Entomologist 80(3)

This microsporidium was detected in 16 colonies, being the other 22 colonies
healthy. The colonies were separated from the soil by flotation according to the tech
niques described by Banks et al. (1981). Colonies with no queen were removed from
the study, consequently, only 14 infected colonies were considered. Fifteen of the 22
healthy colonies collected at the same site were used as controls.
Because all colonies were polygyne, each one was fragmented into separate, equal
subcolonies comprised of one queen selected at random, 50 small and 50 large work
ers. The fragmented colonies were kept in plastic rearing trays (40 x 30 x 15 cm)
dusted with talc to prevent escape. The test was conducted in a walk-in rearing cham-
ber (28.6 + 1.3C and 60-90% RH). The colonies were fed twice a week with approxi
mately 100 adult house flies and V2 egg yolk; a water source and honey-agar cubes
were always present in the rearing trays.
The egg laying of the queens was checked daily only to confirm their fertility Mor
tality of the colonies was recorded and compared between infected and healthy colo
nies. A colony was considered dead when its queen was found dead. Growth of colonies
and mortality rate of individual workers was not quantified.

Survival of Workers. Test I

In December 1995, 4 colonies of S. richteri (2 Thelohania-infected and 2 healthy)
were excavated from Rt. 205, km 180, Saladillo, Buenos Aires Province. They were
separated from the soil by flotation (Banks et al. 1981). Although the exact percentage
of infected workers present in the infected colonies was not determined, based on pre
vious work (Briano et al. 1996), we estimated that it was, on average, 88%. A total of
160 workers was selected from the colonies. Twenty small (head width: 0.67 + 0.06
mm) and 20 large workers (head width: 1.08 + 0.17 mm) were separated at random
from each of the 4 colonies. The selected workers were put in groups of 10 in individual
cells (4 x 4 x 2 cm) of plastic rearing trays (40 x 20 x 2 cm). A plastic lid covered each
cell preventing escape. The workers were held without food until death. A small piece
of moistened cotton was present in the cells as a source of moisture. The test was con
ducted in a walk-in rearing chamber (27.3 1.6'C; 70-90% RH). Mortality was re
corded daily and survival of small and large workers was compared.

Survival of Workers and Sexuals. Test II

In April 1996, 3 colonies of S. richteri (2 Thelohania-infected and a healthy one)
were excavated in Moreno, Buenos Aires Province, and separated from the soil by flo
station (Banks et al. 1981). Sixty-four workers of different sizes separated at random
from each infected colony along with 13 female sexual were put individually in cells
of plastic rearing trays with moistened cotton as above. The trays were kept at room
temperature (21.8 + 1.4'C for workers and 20.7 + 1.6'C for sexuals. The workers and
sexual were held without food until death. Only a small piece of moistened cotton
was present in the cells as above. Workers of different sizes (n = 32) and sexual (n
18) separated from the healthy colony were used as controls. Once dead, the head
width of each worker was measured under an ocular micrometer. To confirm infected
and healthy individuals, the workers and sexual were crushed individually in a drop
of water placed on a microscope slide and checked under a phase-contrast microscope.
Mortality was recorded daily and survival was compared between confirmed infected
and healthy individuals. Arbitrarily, we considered minor workers those with head
widths from 0.6 to 0.8 mm, medium workers those with head widths from 0.9 to 1.1
mm, and major workers those with head widths from 1.2 to 1.5 mm.

September, 1997

Briano & Williams: Effect ofT. solenopsae

Weight of Workers

Sixty-seven infected and 65 healthy workers (not starved) of different sizes were
selected at random from the colonies used in Test II. They were weighed (live weight)
on an electronic balance (Precisa 120 A, PAG Oerlikon AG, Zurich, Switzerland),
killed in 70% ethyl alcohol and their head widths measured under an ocular microme
ter. Each worker was crushed on a microscope slide and examined under a phase-con
trast microscope to confirm the presence or absence of T solenopsae. The live weights
of infected and healthy workers were compared and correlated with worker size.

Statistical Analysis

Mortality rate was analyzed with the logrank method, an application of the Man
tel-Haenszel method (Mantel & Haenszel 1959). Longevity of colonies and survival of
individual ants was analyzed with 2-sample ttest. The simple linear regression model
was used to correlate survival of workers with their size, and the curvilinear (cubic)
model was used to correlate the live weight of workers with their size. Minitab Statis
tical Software (1991) was used for ttests and regressions. Means are reported + 1 SD.


Longevity of Colonies

Longevity of infected colonies was significantly shorter than in healthy colonies
(Fig. 1). The cumulative mortality during the first 21 d was 64% for infected colonies
and 24% for healthy colonies. After 3 mo, mortality was 92% for infected colonies and
49% for healthy colonies (Logrank method; 2 = 6.0; df = 1; P< 0.025). In most colonies
the queens died after the workers.
The different mortality rate between infected and healthy colonies suggests that T
solenopsae is lethal to stressed laboratory colonies of the black imported fire ant. Al
though mortality of healthy colonies is usually high under laboratory conditions, as
this test showed, clearly this pathogen exerted additional stress and increased mor
tality. This is consistent with results of field work that showed a detrimental effect of
this microspordium on native fire ant populations and individual colonies of S. richteri
(Briano et al. 1995a; 1995b). These results also agree with Knell et al. (1977) who re
ported that colonies of S. invicta infected with this microsporidium cannot be main
trained under laboratory conditions as long as healthy colonies.
In this experiment we actually compared residual longevity because queens and
workers were not newly closed when the test started. Comparisons are still valid be
cause this also happened for healthy colonies. The actual life span of infected colonies
compared to healthy colonies in the laboratory remains unknown and should be in
Egg-laying started at day 11 in 2 healthy colonies and at day 14 in one infected col
ony. At day 21, 72% of the surviving healthy colonies and 80% of the surviving infected
colonies showed worker brood production. The egg-laying rate of infected and healthy
queens was not compared and deserves further investigation.

Survival of Workers. Test I

Mortality rate of workers from infected colonies was higher than that of workers
from healthy colonies (Fig. 2). For small workers, after 3 d of starvation, mortality of

Florida Entomologist 80(3)

10" I00

0o .. .. ......................... ... ........ ........................................................................................ ........... ......... .

6 0 - -- . .... ..... ......... .. ................................................. ...................
S60 --------------

0 20 40 60 80 100

--f- Infected colonies --- Healthy colonies

Fig. 1. Mortality of fragmented (1 queen and 100 workers) infected and healthy col-
onies of S. richteri reared at 28.6C.

individuals from infected colonies was 75% and mortality of healthy ones was 43%. At
day 4, when all workers from infected colonies had died, 8% of healthy workers were
still alive (Logrank method; X2 = 4.5; df = 1; P< 0.05). On average, the survival of small
workers from infected colonies was 8.8% shorter than that of healthy ones. The mean
survival time was 3.1 + 0.2 d for workers from infected colonies and 3.4 + 0.7 d for
healthy ones (t= 2.691; df = 78; P= 0.0087).
For large workers, after 4 d of starvation, mortality of individuals from infected
colonies was 95% and mortality of healthy ones was 33% (Fig. 2). At day 6, when all
workers from infected colonies had died, 8% of the healthy workers were still alive
(Logrank method; X2 = 16.45; df = 1; P < 0.001). On average, large workers from in-
fected colonies lived 29.2% less than healthy workers. The mean survival time was 3.4
+ 0.7 d for workers from infected colonies and 4.8 + 1.3 d for healthy ones (t= 5.633;
df 78; P< 0.0001).
The difference in survival time both in small and large workers was underesti-
mated because some workers from infected colonies could have been actually healthy.
The difference was larger in large than in small workers (Fig. 2). It seems that so-
lenopsae affected large workers more than small workers. This is consistent with the
assumption that solenopsae, being a chronic disease, would affect more severely
those individuals with longer life span such as large workers. The tasks performed by
large workers in the colony (mound construction, foraging, territory defense, and
transport of sexual broods) would be affected more severely than the tasks performed
primarily by small workers. The actual impact that the high prevalence of infected

September, 1997

Briano & Williams: Effect ofT. solenopsae

S/ I

1 60

2 0 . .................................................... ............... ..... ........................ ..... ..........

0 1 2 3 4 5 6 7 8

--- Large infected workers -i- Large healthy workers

Small infected workers -- Small healthy workers

Fig. 2. Mortality of starved small and large workers from infected and healthy col
onies of S. richteri kept at 27.3C.

workers would have in field colonies remains unknown but is consistent with the find
ings reported by Briano et al. (1995a, 1995b).

Survival of Workers and Sexuals. Test II

The mortality rate of healthy workers was similar to that of infected workers (Fig.
3; logrank method; ,2 = 0.256; df = 1; P> 0.5). Although the mean survival time of in
fected workers was shorter than that of healthy workers, no statistically significant
differences were found. Infected minor workers survived 5.9 + 5.3 d and healthy mi
nor workers survived 6.5 + 5.0 d (t= 0.457; df = 76; P 0.648). Infected medium work
ers survived 9.0 + 9.3 d and healthy medium workers 10.0 + 8.8 d (t -0.278; df 44;
P 0.782). Infected major workers survived 11.5 + 9.1 d and healthy major workers
12.9 + 13.9 d (t= 0.313; df= 22; P= 0.756).
The regression of survival on worker size showed very low coefficients of determi
nation for both healthy workers (2 = 0.07) and infected ones ( = 0.05). The main rea
son for this was the high individual variability. Calabi & Porter (1989) also reported
a high scatter in regression of longevity on worker size for S. invicta in the United
States (2 0.02). They speculated that the scatter was due to the absence of queens,

Florida Entomologist 80(3)

I 8 0 -- .



0 10 20 30 40 50

Infected workers -- Healthy workers

Fig. 3. Mortality of starved infected and healthy workers of S. richteri kept at
21.8 C.

brood and/or intercolony differences. In our experiment, an extra source of variability
would be the undetermined age of the workers when the test started.
Considering the tests reported in this article, worker survival decreased about
60% when temperature increased from 22 to 27C. The validity of this comparison
may be questionable because the tests were conducted separately, the ants were col
elected in different locations and in different seasons. However, the information re
ported is consistent with studies conducted in the United States by Calabi & Porter
(1989) showing that workers of S. invicta had an 80% reduction in longevity when the
temperature increased from 17 to 30 C. It seems that at lower temperatures, the re
duced activity and metabolic rate of the workers can reduce the debilitating effects of
the infection. This should be investigated. We speculate that the detrimental effect of
T solenopsae could be more important in areas with warmer temperatures. According
to Tanada & Kaya (1993), temperatures higher than 30 C can limit the infectivity of
pathogens, but moderately high field temperatures accelerate the infectious process
and result in quicker mortality
The mortality rate of infected female sexual was not significantly different from
that of healthy ones (Fig. 4; logrank method; X2 0.45; df = 1; P > 0.5). However, the
mortality during the first 10 d was much higher for infected individuals (2 6.36; df
1; P < 0.025). This means that infected sexual females (future queens) died quicker
than healthy ones and might represent a negative effect of T solenopsae on the colony

September, 1997

Briano & Williams: Effect ofT. solenopsae






0 5 10 15 20 25 30 35 40 45 50 55 60

--- Infected female sexual Healthy female sexual

Fig. 4. Mortality of starved sexual females of S. richteri kept at 20.7C.

founding within infested areas. Again, this is consistent with field work showing a
detrimental effect of this pathogen on S. richteri (Briano et al. 1995a, 1995b).
On average, infected sexual survived 23.0 + 21.0 d and healthy ones survived 32.2
+ 15.5 d, but this difference was not statistically significant (t 1.413; df = 29; P
0.168). This was probably due to the small sample size and high individual variability.
Unfortunately, no more sexual were available when the test started. This test should
be replicated with larger sample size and at several temperatures.
As expected, mean survival time of sexual was longer than that of major workers.
This can be attributed in part to the fact that the ambient temperature was slightly
lower in the test with sexual, but the longer survival should be primarily attributed
to their larger body size and their extra energy source provided by the histolysis of
wing muscles. After 2-3 wk of starvation, all sexual lost their wings.

Weight of Workers

The live weight of infected workers was very similar to that of healthy workers. In
fected minor workers weighed 0.656 0.225 mg (range 0.3-1) and healthy minor
workers 0.653 0.246 mg (range 0.3-1.3). Infected medium workers weighed 1.636 +
0.362 mg (range 0.9-2.4) and healthy medium workers 1.628 0.386 (range 1-2.7). In
fected major workers weighed 3.665 + 0.824 (range 2.2-5.4) and healthy workers 3.311
0.747 (range 2.3-5).

.D .

* -- ----

i ..- .. .. ...... ' ..... . .. I. .
... .. . .. ..

-- - - - -- - -- - -- - - - -- - -- - - - -- - - - - -- - - -- - - - -- - -- - -

-- - -- - -- - -- - -- - - -- - -- - -- -- - -- -- - - - - -- - -- - -- - -- - -- - -- - -- - -

Florida Entomologist 80(3)

O/ /

0 tlllIIIii

0.5 0.7 0.9 1.1 1.3 15


-B- Infected workers *-Healthy workers

Fig. 5. Relationship between live weight and size (based on head width) of healthy
and infected workers of S. richteri.

As expected, the weight of workers was highly-positive correlated (cubic function)
with their size (Fig. 5). The regression equation for infected workers was y= 0.091 +
1.480 ( 0.93; F 857.1; df 1, 67; P< 0.0001) and for healthy workers was y
0.225 + 1.447 (T = 0.93; F 847.5; df 1, 65; P< 0.0001). This agrees with Porter
& Tschinkel (1985) who reported a similar relationship for workers of S. invicta in the
United States. There was not any evidence that the presence of T solenopsae
affected the weight of the workers. As suggested by Knell et al. (1977) for S. invicta,
we had speculated that the progressive destruction of the fat body produced by T so
lenopsae, would have an impact on body weight in workers of S. richteri. However, a
hypothetical loss of weight in infected workers could be balanced, at least in part, by
the weight of the cysts totally filled with masses of Thelohania spores. This deserves
further investigation.
We conclude that the microsporidium T solenopsae affected the mortality rate and
shortened the longevity of colonies of S. richteri reared under laboratory conditions.
Survival of starved workers, mainly large workers, and initial mortality of sexual fe
males was also affected. Temperature could be a regulating factor of this effect. These
laboratory findings are consistent with results of field work reported by Briano et al.
(1995a, 1995b), showing reduced mound volumes of infected colonies, less frequent
presence of sexual brood in infected colonies and decreased mound densities in a The
lohania infested area of Argentina.
The introduction of a complex of natural enemies into the United States has been
the ultimate goal of the imported fire ant control project. Among the several potential

September, 1997

Briano & Williams: Effect ofT. solenopsae

candidates, T solenopsae has been the first microorganism evaluated in South Amer
ica as a potential biological control agent. Still, important aspects of its life cycle, such
as the horizontal transmission and field propagation, remain unknown. After those
studies are completed, T solenopsae should be considered for the biological control of
the imported fire ants in the United States.


The authors thank James Becnel, Steven Valles (USDA-ARS Center for Medical,
Agricultural, and Veterinary Entomology, Gainesville, FL, USA), and Daniel Gandolfo
(USDA-ARS South American Biological Control Laboratory, Hurlingham, Buenos
Aires, Argentina) for reviewing this manuscript.


ALLEN, G. E., AND W. F. BUREN. 1974. Microsporidian and fungal diseases of S. invicta
Buren in Brazil. J. New York Entomol. Soc. 82: 125-130.
ALLEN, G. E., AND A. SILVEIRA GUIDO. 1974. Occurrence of microsporidia in Solenop
sis richteri and Solenopsis sp. in Uruguay and Argentina. Florida Entomol. 57:
WILLIAMS, D. P. WOJCIK, AND B. M. GLANCEY. 1981. Techniques for collecting,
rearing and handling imported fire ants. USDA. Sci. and Educ. Admin. Adv. in
Agric. Tech. AAT S-21.
BRIANO, J. A., R. S. PATTERSON, AND H. A. CORDO. 1995a. Relationship between col
ony size of Solenopsis richteri (Hymenoptera: Formicidae) and infection with
Thelohania solenopsae (Microsporida: Thelohaniidae) in Argentina. J. of Econ.
Entomol. 88: 1233-1237.
BRIANO, J. A., R. S. PATTERSON, AND H. A. CORDO. 1995b. Long-term studies of the
black imported fire ant (Hymenoptera: Formicidae) infected with a microspo
ridium. Environ. Entomol. 24: 1328-1332.
1995. Protozoan and fungal diseases in Solenopsis richteri and S. quinquecus
pis (Hymenoptera: Formicidae), in Buenos Aires province, Argentina. Florida
Entomol. 78: 531-537.
BRIANO, J. A., R. S. PATTERSON, J. J. BECNEL, AND H. A. CORDO. 1996. The black im
ported fire ant, Solenopsis richteri, infected with Thelohania solenopsae: intra
colonial prevalence and evidence for transovarial transmission. J. of Invertebr.
Pathol. 67: 178-179.
CALABI, P., AND S. D. PORTER 1989. Worker longevity in the fire ant Solenopsis in-
victa: ergonomic considerations of correlations between temperature, size and
metabolic rates. J. Insect Physiol. 35: 643-649.
JOUVENAZ, D. P. 1983. Natural enemies of fire ants. Florida Entomol. 66: 111-121.
JOUVENAZ, D. P. 1986. Diseases of Fire Ants: Problems and Opportunities, pp. 327
338 in C. S. Lofgren and R. K. Vander Meer [eds.], Fire Ants and Leaf Cutting
Ants: Biology and Management. Westview Press, Boulder, CO. 434 pp.
JOUVENAZ, D. P., W. A. BANKS, AND J. D. ATWOOD. 1980. Incidence of pathogens in fire
ants, Solenopsis spp. in Brazil. Florida Entomol. 63: 345-346.
JOUVENAZ, D. P., C. S. LOFGREN, AND W. A. BANKS. 1981. Biological control of im
ported fire ants: A review of current knowledge. Ann. Entomol. Soc. Am. 27:
KNELL, J. D., G. E. ALLEN, AND E. I. HAZARD. 1977. Light and electron microscope
study of Thelohania solenopsae n. sp. (Microsporida: Protozoa) in the red im
ported fire ant, Solenopsis invicta. J. Invertebr. Pathol. 29: 192-200.
MANTEL, N., AND W. HAENSZEL. 1959. Statistical aspects of the analysis of data from
retrospective studies of disease. J. Nat. Cancer Inst. 22: 719-748.

376 Florida Entomologist80(3) September, 1997

MINITAB STATISTICAL SOFTWARE. 1991. Reference manual, Release 8. Minitab Inc.
State College, PA.
MOSER, B. A. 1995. Comparative analysis of microsporidia of fire ants, Solenopsis
richteri and S. invicta. Ph.D. dissertation. University of Florida, Gainesville,
FL. 125 pp.
PORTER, S. D., AND W. R. TSCHINKEL. 1985. Fire ant polymorphism: the ergonomics of
brood production. Behav. Ecol. Sociobiol. 16: 323-336.
TANADA, Y., AND H. K. KAYA. 1993. Insect Pathology. Academic Press, San Diego, Cal
ifornia. 666 pp.
WILLIAMS, D. F., G. J. KNUE, AND J. J. BECNEL. 1977. Discovery of Thelohania sole
nopsaefrom the red imported fire ant, Solenopsis invicta, in the United States.
J. of Invertebr. Pathol. in press.
WOJCIK, D. P., D. P. JOUVENAZ, W. A. BANKS, AND A. C. PEREIRA. 1987. Biological con
trol agents of fire ants in Brazil, pp. 627-628 in J. Eder & H. Rembold [eds.],
Chemistry and Biology of Social Insects. Verlag J. Peperny, Munich. 757 pp.


Florida Entomologist 80(3)


'E.N.S.A.M., U.FR. d'Ecologie animal et de Zoologie agricole
2, Place Pierre Viala, 34060 Montpellier Cedex 1, France

Departamento de Zoologia, E.S.A.L.Q./U.S.P., 13418-900 Piracicaba-S.P, Brazil


Nine species of mites of the family Phytoseiidae are reported for the first time from
Guadeloupe and Martinique. Measurements of the specimens of each species collected
are given.

Key Words: French Caribbean Islands, predatory mites, Thrips palmi, Solanum mel


Nueve species de acaros de la familiar Phytoseiidae estan senaladas por primera
vez en Guadeloupe y Martinique. Medidas de los individuos de las species recolecta
das son presentadas.

This paper reports on phytoseiid mites from plants in Guadeloupe and Martin
ique, in collections made sporadically between 1985 and 1989. This is the first report
of phytoseiid mites from those Caribbean islands. Setal nomenclature is that of Row

September, 1997

Kreiter & De Moraes: Phytoseiid Mites

ell et al. (1978) and Chant & Hansell (1971) for dorsal and ventral surfaces, respect
tively All measurements are in micrometers. Except where indicated, the collector of
the specimens was J. Etienne, and information on world distribution of each species
was based on Moraes et al. (1986, 1991). The following abbreviations are used in this
paper: I.N.R.A. (Institut National de la Recherche Agronomique; Antilles Guyane);
E.N.S.A.M. (Ecole Nationale Superieure Agronomique de Montpellier).

Amblyseius aerialis (Muma)

Amblyseiopsis aerialis Muma 1955: 264; Garman 1958: 75.
Typhlodromus (Amblyseius) aerialis Chant 1959: 88; Muma 1961: 287.
Amblyseius aerialis DeLeon 1966: 91; Moraes & Mesa 1988: 71; Denmark & Muma
1989: 15.
Specimens Examined: GUADELOUPE -Petit-Bourg, Domaine Duclos I.N.R.A.,
IX 1988 to I-1989, on Solanum melongena L. infested with Thrips palmi Karny (Thys
Previous Records: Algeria, Bermuda, Brazil, Colombia, Cuba, Galapagos, Guyana,
Honduras, India, Jamaica, Mexico and USA.
Remarks: The measurements of the specimens collected agree well with those of
specimens from British Guyana (DeLeon 1966). The average measurements of 4 adult
females followed by the respective ranges (in brackets) are: dorsal shield length 374
(367-382), width 262 (245-284),jl 26 (24-29),j3 49 (46-53),j4 4 (2-5),j5 3 (2-5),j6 5,
J2 5 (5-6), J5 6 (5-6), z2 7 (5-9), z4 9 (6-10), z5 5, Z1 7 (6-7), Z4 145 (133-163), Z5 281
(257-318), s4 106 (94-118), S2 5, S4 10 (9-12), S5 10 (10-11), r3 10 (9-12), R1 10(7-13),
SgeI 44 (43-46), SgeII 38 (36-38), SgeIII 62 (58-70), StiIII 43 (38-48), SgeIV 136 (118
152), StiIV 94 (89-97), StIV 82 (72-89), ST1-ST3 71 (64-79), ST2-ST2 80 (77-82), ST5
ST5 84 (81-86), length of ventrianal shield 120 (118-121), width at ZV2 level 85 (77
89), width at anus level 83 (74-89), length of sclerotized proximal portion of cervix of
spermatheca 14 (12-17), length of unsclerotized distal portion of cervix of spermath
eca 20 (17-22), length of fixed digit 37 (36-38) with 12-14 teeth, length of movable digit
41 (41 42) with 4 teeth.

Iphiseiodes zuluagai Denmark & Muma

Iphiseiodes zuluagai Denmark & Muma 1972: 23.
Amblyseius zuluagai Moraes et al. 1991: 125.
Specimens Examined: GUADELOUPE -Petit-Bourg, Domaine Duclos I.N.R.A., II
1989, on Hibiscus sp., and S. melongena infested with T palmi.
Previous Records: Brazil, Colombia, Cuba, Panama and Puerto Rico.
Remarks: The measurements of the female specimens collected are very similar to
those of the holotype. The average measurements of 5 adult females followed by the
respective ranges (in brackets) are: dorsal shield length 357 (334-394), width 330
(312-356),jl 15 (13-19),j3 24 (23-28),j4 2 (1-3),j5 1 (1-3),j6 3 (1-3), J2 3, J5 4 (3-4),
z2 2 (1-3), z4 1 (1-3), z5 1, Z1 3 (3-4), Z4 4, Z5 90 (84-95), s4 97, S2 3 (3-4), S4 3, S5 3
(3-4), r3 4, R1 4, SgeI 51 (47-55), SgeII 31 (29-33), SgeIII 46 (41-51), StiIII 25 (24-27),
SgeIV 80 (77-85), StiIV 49 (44-51), StIV 33 (29-36), ST1-ST3 47 (44-51), ST2-ST2 72
(66-76), ST5-ST5 111 (104-118), length of ventrianal shield 108 (102-117), width at
ZV2 level 130 (126-140), width at anus level 117 (114-122), length of cervix of sper
matheca 14 (13-15), length of fixed digit 31 (29-34) with 11 teeth, length of movable
digit 35 (34-36) with 3 teeth.
The measurements of one of the 2 adult males collected are: dorsal shield length
287, width 217, jl 13,j3 27, j4, j5,j6 and J2 1, J5 4, z2, z4, z5 5, Z1 and Z4 1, Z5 62,

Florida Entomologist 80(3)

s4 (broken), S2, S4 and S5 1, r3 and R1 3, Sgel 38, SgeII 27, SgeIII 37, StiIII 23, SgeIV
61, StiIV 44, StIV 30, length of ventrianal shield 118, width at anterior corners 173,
length of shaft of spermatodactyl 28.

Neoseiulus anonymous (Chant & Baker)
Amblyseius anonymous Chant & Baker 1965: 21; McMurtry 1983: 254; Schicha &
Elshafie 1980: 32; Moraes & Mesa 1988: 76.
Neoseiulus anonymous Denmark & Muma 1973: 265.
Specimens Examined: GUADELOUPE -Matouba, VI 1988, on Fragaria sp., S. Si
mon leg.
Previous Records: Brazil, Colombia, Cuba, Guatemala, Honduras, Mexico and
Remarks: The measurements of a single adult female collected are: dorsal shield
length 312, width 144,jl 19,j3 38,j4 36,j5 41,j6 51, J2 53, J5 10, z2 46, z4 46, z5 42,
Z1 55, Z4 71,Z5 72, s4 58, S2 61,S4 51, S5 39, r3 36, R1 36, StIV 62, ST1-ST3 60, ST2
ST2 and ST5-ST5 58, length of ventrianal shield 101, width at ZV2 level 77, width at
anus level 72, length of cervix of spermatheca 12, length of fixed digit 22, length of
movable digit 24.

Fundiseius urquharti (Yoshida-Shaul & Chant), Comb. nov.
Amblyseius urquhartiYoshida-Shaul & Chant 1988: 2055.
Specimens Examined: GUADELOUPE -Petit-Bourg, Domaine Duclos I.N.R.A., II
1989, on Hibiscus sp. and S. melongena infested with T palmi.
Previous Records: Antigua (Type material -Yoshida-Shaul & Chant 1988).
Remarks: The measurements of the female specimens collected are similar to
those of the holotype (only known female specimen of this species to date), except for
the shorterj3 and longer Z4 and s4. The average measurements of 5 adult females fol
lowed by the respective ranges (in brackets) are: dorsal shield length 376 (344-423),
width 301 (286-324),jl 16 (13-18),j3 22 (19-25),j4 3 (1-6),j5 2 (1-5),j6 7 (6-9), J2 6 (5
6),J5 10 (811), z2 10 (8-13), z4 15 (14-17), z5 2 (13), Z1 7 (610), Z4 89, Z5 88 (86-89),
s4 82 (80-86), S2 14 (11-17), S4 11 (10-11), S5 9 (6-10), r3 17 (13-18), R1 13 (11-14),
StIV 39 (37-42), ST1-ST3 44 (42-46), ST2-ST2 71 (69-76), ST5-ST5 126 (121-130),
length of ventrianal shield 145 (140-152), width at ZV2 level 206 (184-225), width at
anus level 161 (130-178), length of cervix of spermatheca 11 (8-19), length of fixed
digit 34 (33-34) with 10-12 teeth, length of movable digit 38 (37-38) with 2 teeth. The
wide amplitude for the measured length of the cervix of the spermatheca relates to the
difficulty in determining its distal end, because the sclerotization of this structure di
finishes very slowly toward this portion.

Phytoseiulus macropilis (Banks)
Laelaps macropilis Banks 1905: 139.
Phytoseiulus speyeriEvans 1952: 398 (after Denmark & Muma 1973).
Phytoseiulus chantiEhara 1966: 135 (after Denmark & Muma 1973).
Phytoseiulus macropilis Muma et al. 1970: 30; McMurtry 1983: 259; Denmark &
Schicha 1983: 31.
Specimens Examined: GUADELOUPE -Pointe-a-Pitre, Domaine Duclos I.N.R.A.,
VIII-1985, on Phaseolus vulgaris L.
Previous Records: Angola, Barbados, Brazil, Canary Islands, Cook Islands, Colom
bia, Cuba, Fiji, Guatemala, Hawaii, Jamaica, Mexico, New Caledonia, Panama, Peru,
Puerto Rico and USA.

September, 1997

Kreiter & De Moraes: Phytoseiid Mites

Remarks: The measurements of the adult females collected are similar to those
provided by Denmark & Schicha (1983), except for the longer j5,j6, z4, s4 and r3. The
average measurements of 3 adult females followed by the respective ranges (in brack
ets) are: dorsal shield length 328 (310-344), width 213 (199-230),jl 24 (23-25),j3 41
(38-46),j4 52 (48-56),j5 75 (72-77),j6 154 (145-165), J5 5, z2 9 (8-9), z4 57 (55-60), z5
9 (6-13), Z1 107 (98-114), Z4 126 (118-135), Z5 109 (98-114), s4 174 (163-185), S5 28
(27-28), r3 and R1 24 (23-25), Sgel 47 (44-52), SgeII 31 (28-32), SgeIII 31 (30-32), Sti
III 28 (25-29), SgeIV 77 (72-79), StiIV 37 (34-41), StIV 109 (104 114), ST1-ST3 65 (64
67), ST2-ST2 74 (70-76), ST5-ST5 72 (70-75), length of ventrianal shield 91 (89-93),
width at JV2 level 67 (61 75), width at anus level 66 (64-70), length of proximal, in
flated portion of the cervix of spermatheca, 13 (11 14), length of remaining sclerotized,
distal portion of cervix 19 (18-22), length of fixed digit 22 (20-23) with 9 teeth, length
of movable digit 24 (23-24) with 3 teeth.
The measurements of 2 adult males collected are: dorsal shield length 241-267,
width 127 (201-226),jl 19-20,j3 29-42,j4 38-50,j5 56-60,j6 116-121, J5 45, z2 8-10,
z4 52-61, z5 10-11, Z1 75-83, Z4 89-95, Z5 80, s4 121-133, S5 28-29, r3 15-17, R1 18
20, SgeIII 25, StiIII 24, SgeIV 52, StiIV 27, StIV 74-79, length of ventrianal shield
105-116, width at anterior corners 152-163, length of shaft of spermatodactyl 15-17.

Proprioseiopsis cannaensis (Muma)

Amblyseiulus cannaensis Muma 1962: 4.
Amblyseius cannaensis Moraes & McMurtry 1983: 132; Moraes & Mesa 1988: 77;
Moraes et al. 1991: 126.
Proprioseiopsis cannaensis Muma et al. 1970: 38.
Specimens Examined: GUADELOUPE -Saint Francois, XI-1987, on S. melongena
infested with T palmi; Petit-Bourg, Domaine Duclos I.N.R.A., on S. melongena in
fested with T palmi.
Previous Records: Brazil, Colombia, Cuba, Ecuador, El Salvador, Guyana, New
Caledonia, Paraguay and USA.
Remarks: The measurements of the adult females collected agree well with those
of the holotype; however, similarly to specimens from Brazil (Moraes & McMurtry
1983), Colombia (Moraes & Mesa 1988) and Cuba (Moraes et al. 1991), S4 is shorter
than in the holotype. The average measurements of 5 adult females collected followed
by the respective ranges (in brackets) are: dorsal shield length 334 (316-343), width
264 (250-279),jl 25 (23-27),j3 67 (64-72),j4 5 (4-5),j5 5 (4-6),j6 10 (9-12), J5 9 (8-12),
z2 38 (36-42), z4 24 (19-26), z5 5 (46), Z1 23 (19-25), Z4 110 (95-114), Z5 88 (77-101),
s4 100 (95-112), S2 20 (13-23), S4 14 (14-15), S5 14 (13-17), r3 20 (18-24), R1 15 (13
17), SgeIII 27 (22-32), StilII 25 (19-24), SgeIV 53 (46-60), StiIV 35 (28-41), StIV 76
(70-83), ST1-ST3 52 (51-55), ST2-ST2 74 (70-76), ST5-ST5 95 (90-97), length of ven
trianal shield 107 (91 117), width at ZV2 level 116 (113-121), width at anus level 101
(104-110), length of cervix of spermatheca 17 (12-19), length of fixed digit 31 (28-33),
length of movable digit 32 (30-34).

Proprioseiopsis mexicanus (Garman)

Amblyseiopsis mexicanus Carman 1958: 75.
Amblyseius mexicanus Moraes & McMurtry 1983: 134.
Proprioseiopsis mexicanus Muma et al. 1970: 48.
Specimens Examined: Saint Francois, XI-1987, on S. melongena infested with T
palmi; Matouba, VI-1988, on Fragaria sp., S. Simon leg.

Florida Entomologist 80(3)

Previous Records: Australia, Brazil, Colombia, Cuba, Hawaii, Mexico, New
Zealand, Panama and USA.
Remarks: The measurements of the adult females collected are very similar to
those of the holotype, provided by Moraes & McMurtry (1983). The average measure
ments of 5 adult females collected followed by the respective ranges (in brackets) are:
dorsal shield length 335 (331 339), width 224 (212-241),jl 19 (15-22),j3 30 (24-34),j4
andj5 5 (4-7),j6 5 (5-6), J5 9 (9-10), z2 12 (11-14), z4 10, z5 4 (4-5), Z1 6 (5-7), Z4 74
(72-76), Z5 103 (97-110), s4 59 (56-65), S2 and S4 9 (8-10), S5 9 (9-12), r3 11 (9-14), R1
9 (8-10), SgeII 23 (20-24), SgeIII 24 (23-25), StiIII 22 (20-24), SgeIV 49 (48-51), StiIV
32 (27-36), StIV 56 (51-60), ST1-ST3 60 (58-62), ST2-ST2 68 (65-74), ST5-ST5 66 (64
72), length of ventrianal shield 108 (103 114), width at ZV2 level 92 (86-97), width at
anus level 85 (80-89), length of cervix of spermatheca 9 (6-10), length of fixed digit 33
(29-38), length of movable digit 31 (29-32). Fixed digit with 8 teeth; movable digit with
1 tooth.
The measurements of a single adult male collected are: dorsal shield length 279,
width 194,jl 17,j3 24,j4 5,j5 4,j6 5, J5 9, z2 and z4 11, z5 10, Z1 8, Z4 56, Z5 74, s4
43, S2, S4 and S5 9, r3 10, R1 8, SgeIII 19, SgeIV 32, StiIV 23, StIV 48, length of ven
trianal shield 103, width at anterior corners 121, length of shaft of spermatodactyl 18.

Phytoseius (Phytoseius) rexDeLeon

Phytoseius rex DeLeon 1967: 12.
Phytoseius (Phytoseius) rexDenmark & Muma 1975: 295.
Specimens Examined: GUADELOUPE -Saint Francois, XI-1987, on S. melongena
infested with T palmi.
Previous Records: Guyana, Puerto Rico and Trinidad.
Remarks: The measurements of the specimens collected are very similar to those
of the holotype, except for the slightly longer StIV The average measurements of 5
adult females collected followed by the respective ranges (in brackets) are: dorsal
shield length 285 (276-290), width 153 (149-156),jl 27 (26-29),j3 44 (41-48),j4 5 (5
7),j5,j6 and J5 6 (5-7), z2 11 (7-12), z3 24 (22-29), z4 8 (7-10), z5 6 (5-7), Z4 73 (70-77),
Z5 68 (62-72), s4 108 (101 110), s6 76 (72-79), r3 42 (41 46), SgeIV 21 (19-22), StiIV 48
(43-50), StIV 30 (29-31), ST1-ST3 60 (58-60), ST2-ST2 72, ST5-ST5 68 (62-70), length
of ventrianal shield 89 (84-96), width at ZV2 level 54 (50-58), width at anus level 53
(50-58), length of cervix of spermatheca 29 (24-36), length of inflated region of major
duct adjacent to atrium 13 (7-17), length of fixed digit 25 (24-26), length of movable
digit 26. Fixed digit with 3 teeth; movable digit with 1 tooth.
The measurements of 2 adult males collected are: dorsal shield length 214, width
127(120-134),jl 19,j3 24-29,j4,j5 andj6 5, J5 5-7, z2 7 10, z3 17, z4 and z5 5, Z4 38
41, Z5 34, s4 62-67, s6 43-48, r3 29-31, SgeIV 12, StiIV and StIV 19-22, length ofven
trianal shield 79-91, width at anterior corners 127, length of shaft of spermatodactyl

Phytoseius (Phytoseius) woodburyi DeLeon

Phytoseius (Phytoseius) woodburyi DeLeon 1965: 130.
Phytoseius (Dubininellus) woodburyi Denmark 1966: 64.
Phytoseius (Phytoseius) woodbury Muma and Denmark 1968: 236.
Specimens Examined: MARTINIQUE -Fort-de-France, 1-1987, on Hibiscus sp., B.
Hostachy leg.
Previous Records: Colombia, Hawaii, India, Jamaica, Puerto Rico and Trinidad.

September, 1997

Kreiter & De Moraes: Phytoseiid Mites

Remarks: The measurements of the specimens collected are similar to those of the
holotype. The average measurements of 6 adult females collected followed by the re
spective ranges (in brackets) are: dorsal shield length 286 (283-293), width 160 (156
163),jl 29 (26-31),j3 31 (29-31),j4 and j5 5 (5-7),j6 6 (5-7), J5 7, z2 12 (7-14), z3 31
(29-34), z4 11 (10-14), z5 5 (5-7), Z4 89 (82-94), Z5 69 (65-72), s4 121 (115-130), s6 81
(77-86), r3 44 (43-48), SgeIV 10 (10-12), StiIV 49 (48-50), StIV 25 (22-29), ST1-ST3 53
(48-55), ST2-ST2 61 (60-62), ST5-ST5 61 (58-65), length of ventrianal shield 87 (84
91), width at ZV2 level 35 (34-36), width at anus level 48 (46-50), length of cervix of
spermatheca 4 (2-5), length of fixed digit 23 (22-24), length of movable digit 21 (19-22).
Fixed digit with 3 teeth; movable digit with 1 tooth.


We are very grateful to J. Etienne (I.N.R.A. Antilles Guyane, Petit-Bourg, Guade
loupe, France), B. Hostachy (Service Regional de la Protection des V6egtaux, Fort-de
France, Martinique, France) and S. Simon (Centre International de Recherche
Agronomique pour le Developpement Dpartement Fruits, Legumes, Horticulture,
Capesterre Belle Eau, Guadeloupe, France) for kindly providing the specimens used
in this study
We thank J. A. McMurtry (California, U.S.A.), F Leclant, M. Martinez and P. Auger
(E.N.S.A./I.N.R.A., Montpellier, France) for valuable suggestions on the manuscript,
and CNP (Brazil) and E.N.S.A.M. (France) for partial support.


BANKS, N. 1905. Descriptions of some new mites. Proc. Entomol. Soc. Washington 7:
CHANT, D. A. 1959. Phytoseiid mites (Acarina: Phytoseiidae). Part I. Bionomics of
seven species in southern England. Part II. A taxonomic review of the family
Phytoseiidae, with descriptions of thirty-eight new species. Canadian Ento
mol., Supplement 12, 166 pp.
CHANT, D. A., AND E. W. BAKER 1965. The Phytoseiidae (Acarina) of Central America.
Mem. Entomol. Soc. Canada 41: 56 pp.
CHANT, D. A., AND R. I. C. HANSELL. 1971. The genus Amblyseius (Acarina: Phytosei
idae) in Canada and Alaska. Canadian J. Zool. 49: 703-758.
DELEON, D. 1965. Phytoseiid mites from Puerto Rico with descriptions of new species
(Acarina: Mesostigmata). Florida Entomol. 48: 121-131.
DELEON, D. 1966. Phytoseiidae of British Guyana with keys to species (Acarina: Me
sostigmata), in Studies on the fauna of Suriname and other Guyanas, 8: 81
DELEON, D. 1967. Some mites of the Caribbean Area. Part I. Acarina on plants in
Trinidad, West Indies. Allen Press Inc., Lawrence, Kansas, 66 pp.
DENMARK, H. A. 1966. Revision of the genus Phytoseius Ribaga, 1904 (Acarina: Phy
toseiidae). Florida Dept. Agr. Bull. 6: 1105.
DENMARK, H. A., AND M. H. MUMA. 1972. Some Phytoseiidae of Colombia (Acarina:
Phytoseiidae). Florida Entomol. 55: 19-29.
DENMARK, H. A., AND M. H. MUMA. 1973. Phytoseiid mites of Brazil (Acarina: Phy
toseiidae). Revista Brasileira de Biologia 33: 235-276.
DENMARK, H. A., AND M. H. MUMA. 1975. The Phytoseiidae (Acarina: Mesostigmata)
of Puerto Rico. J. Agric. Univ. Puerto Rico 59: 279-304.
DENMARK, H. A., AND M. H. MUMA. 1989. A revision of the genus Amblyseius Berlese,
1914 (Acari: Phytoseiidae). Occasional Papers Florida Sta. Coll. Arthropods 4:
149 pp.
DENMARK, H. A., AND E. SCHICHA. 1983. Revision of the genus Phytoseiulus Evans
(Acarina: Phytoseiidae). Internat. J. Acarol. 9: 27-35.

Florida Entomologist 80(3)

EHARA, S. 1966. Some mites associated with plants in the state of Sao Paulo, Brazil,
with a list of plant mites of South America. Japanese J. Zool. 15: 129-150.
EVANS, G. O. 1952. On a new predatory mite of economic importance. Bull. Entomol.
Res. 43: 397-401.
GARMAN, P. 1958. New species belonging to the genera Amblyseius and Amblyseiopsis
with keys to Amblyseius, Amblyseiopsis and Phytoseiulus. Ann. Entomol. Soc.
America 51: 69-79.
MCMURTRY, J. A. 1983. Phytoseiid mites from Guatemala, with descriptions of two
new species and redefinitions of the genera Euseius, Typhloseiopsis, and the
Typhlodromus occidentalis species group (Acari: Mesostigmata). Internat. J.
Acarol. 25: 249-272.
MORAES, G. J. DE, AND J. A. MCMURTRY. 1983. Phytoseiid mites (Acarina) of north
eastern Brazil with descriptions of four new species. Internat. J. Acarol. 9: 131
MORAES, G. J. DE, J. A. MCMURTRY, AND H. A. DENMARK. 1986. A catalog of the mite
family Phytoseiidae: references to taxonomy, synonymy, distribution and habi
tat. EMBRAPA-DDT, Brasilia, 353 pp.
MORAES, G. J. DE, AND N. C. MESA. 1988. Mites of the family Phytoseiidae (Acari) in
Colombia, with descriptions of three new species. Internat. J. Acarol. 14: 7188.
MORAES, G. J. DE, N. C. MESA, AND A. BRAUN. 1991. Some phytoseiid mites of Latin
America (Acari: Phytoseiidae). Internat. J. Acarol. 17: 117-139.
MUMA, M. H. 1955. Phytoseiidae (Acarina) associated with citrus in Florida. Ann. En
tomol. Soc. America 48: 262-272.
MUMA, M. H. 1961. Subfamilies, genera, and species of Phytoseiidae (Acarina: Mesos
tigmata). Florida St. Mus. Bull. Biol. Sci. 5: 267-302.
MUMA, M. H. 1962. New Phytoseiidae (Acarina: Mesostigmata) from Florida. Florida
Entomol. 45: 1-10.
MUMA, M. H., AND H. A. DENMARK. 1968. Some generic descriptions and name
changes in the family Phytoseiidae (Acarina: Mesostigmata). Florida Entomol
ogist 51: 229-240.
MUMA, M. H., H. A. DENMARK, AND D. DELEON. 1970. Phytoseiidae of Florida. Arthro
pods of Florida and Neighboring Land Areas, 6: 150 pp.
ROWELL, H. J., D. A. CHANT, AND R. I. C. HANSELL. 1978. The determination of setal
homologies and setal patterns on the dorsal shield in the family Phytoseiidae
(Acarina: Mesostigmata). Canadian Entomol. 110: 859-876.
SCHICHA, E., AND M. ELSHAFIE. 1980. Four new species of phytoseiid mites from Aus
tralia, and three species from America redescribed (Acari: Phytoseiidae). J.
Australian Entomol. Soc. 19: 2736.
YOSHIDA-SHAUL, E., AND D. A. CHANT. 1988. Descriptions of two unusual new species
in the genus Amblyseius Berlese (Acari: Phytoseiidae). Canadian J. Zool. 66:

September, 1997

Bertschy et al.: Attraction of cassava mealybugparasitoids 383


Institute of Plant Sciences, Applied Entomology,
ETH (Swiss Federal Institute of Technology), CH-8092 Zurich, Switzerland

'Centro Internacional de Agricultura Tropical, CIAT, Cali, Colombia

2Current address: Institut de Zoologie, Universit6 de Neuchatel,
CH-2007 Neuchatel, Switzerland

3To whom correspondences should be addressed

We investigated whether cassava plants that are infested by the cassava mealy
bug, Phenacoccus herreni (Pseudococcidae, Sternorrhyncha), emit attractants for the
encyrtid parasitoids Aenasius vexans Kerrich, Apoanagyrus (Epidinocarsis) divers
cornis Howard, and Acerophagus coccois Smith. Bioassays with a Y-tube olfactometer
showed for all three species that female wasps were most responsive and selective
when they were 1.5 to 2.5 days old. Females of these age groups were used to test their
ability to distinguish between the odor of plants with and without mealybugs. The
wasps were offered choices between infested cassava leaves vs. healthy ones, infested
leaves vs. clean air, and healthy leaves vs. clean air. A. vexans and A. diversicornis
were strongly attracted to infested leaves and preferred these over healthy ones. In
contrast, A. coccoiswas significantly attracted to either healthy or infested leaves, and
did not distinguish between the two. The results suggest that A. coccois, which has the
broadest known host range of the three, may be responsive only to general plant
odors, while A. vexans and A. diversicornis respond more specifically to odors associ
ated with mealybug infestation.

Key Words: Aenasius vexans, Apoanagyrus (Epidinocarsis) diversicornis, Acerophagus
coccois, cassava (Manihot esculenta), host location, semiochemicals

Se investig6 si las plants de yuca que son infestadas por el piojo harinoso, Phena
coccus herreni (STERNORRHYNCHA: Pseudococcidae), emiten sustancias atractivas
para los parasitoides Encyrtidae Aenasius vexans, Apoanagyrus (Epidinocarsis) diver
sicornis y Acerophagus coccois. Ensayos con un tubo olfact6metro en Y mostraron que
las tres species tienden a responder y seleccionar mas frecuentemente cuando tienen
de 1.5 a 2.5 dias de edad. Las hembras de esta edad fueron usadas para determinar
su capacidad de distinguir entire el olor de plants con y sin piojos. Se ofrecio a las
hembras olores de yuca infestadas o limpias, hojas infestadas o aire puro y hojas lim
pias o aire puro. A. vexans y A. diversicornis fueron atraidas fuertemente por las hojas
infestadas y presentaron preferencia por estas hojas contra las hojas limpias. En con
traste, A. coccois fue atraida de manera significant por hojas limpias u hojas infesta
das contra aire, y no pudo dintinguir entire ambos olores. Los resultados sugieren que
A. coccois, que tiene el mas alto rango de hu6spedes de los tres, puede responder solo
a los olores generals de las plants, mientras A. vexans y A. diversicornis responded
mas especificamente a los olores asociados con la presencia de los piojos harinosos.

Florida Entomologist 80(3)

Cassava mealybugs are among the most damaging pests of cassava in South
America and Africa (Vargas & Bellotti, 1984). The two most important species are
Phenacoccus herreni Cox & Williams and P manihoti Matile-Ferrero (Sternorrhyn
cha: Pseudococcidae), which both originate from South America (Cox & Williams,
1981; Bellotti et al., 1984). R herreni appeared as a problem rather suddenly in North
east Brazil in the mid 1970s and was then reported from Colombia, Venezuela and
Guyana (CIAT, 1984; 1987; 1988; 1990); it can cause root yield losses up to 80% (Bel
lotti et al., 1984; Bellotti, 1983). In Africa, the closely related P manihoti became a se
rious pest in most of the cassava growing regions after its accidental introduction in
the 1970s (Matile-Ferrero, 1977; Herren & Neuenschwander, 1991). For both pest spe
cies biological control programs have been developed. The encyrtid parasitoid
Apoanagyrus (Epidinocarsis) lopezi (De Santis) was successfully released in African
in the 1980s. It established and now maintains the mealybug population at an accept
able low-density in most regions (Herren & Neuenschwander, 1991; CIAT, 1992). For
the 5% of the African cassava fields where the parasitoid has not been effective in con
trolling the mealybug (Neuenschwander et al., 1991), alternative control agents were
investigated such as two strains of the coccinellid predator Hyperaspis notata Mul
sant (Staubli Dreyer et al., 1997a; 1997b; 1997c).
Natural enemies of P herreni have been systematically collected for the control of
the mealybug in South America, and laboratory colonies of three encyrtid parasitoids
were established at CIAT (Centro International de Agricultura Tropical), in Cali, Co
lombia. These parasitoids are Aenasius vexans Kerrich, Apoanagyrus (Epidinocarsis)
diversicornis Howard (asexual strain) and Acerophagus coccois Smith (CIAT, 1982;
1983; 1990). Knowledge on the biology of these insects is limited. Published informa
tion is mostly restricted to CIAT reports (1982-1992).
At the beginning of this century, studies showed that parasitic wasps use olfaction
to locate hosts and that they may first be attracted to the food that their hosts feed on
(Picard & Rabaud, 1914; Thorpe & Jones, 1937; Thorpe & Caudle, 1938). More re
cently, it was demonstrated that herbivore-damaged plants can play a key role in at
tracting enemies of insect herbivores (Dicke et al., 1990; Turlings et al., 1990; 1995;
Vet & Dicke, 1992). For example, lima bean plants under spider mite attack release
specific volatiles that are attractive to predatory mites (Dicke et al., 1990) and similar
volatile compounds released by caterpillar-infested maize plants are used by parasi
toids to locate caterpillars (Turlings et al., 1990).
Volatiles emitted by mealybug-infested plants are also suspected to attract natu
ral enemies of the mealybug (Nadel & van Alphen, 1987). Changes in chemicals pro
duced by the cassava plant due to P manihoti infestation have been reported by
Calatayud et al. (1994). Such changes could result in the emission of volatiles and ex
plain why A. lopezi and A. diversicornis (sexual strain) are attracted by P manihoti
infested cassava plants (Nadel & van Alphen, 1987; van Alphen et al., 1990). The feed
ing behavior of P herreni is very similar to that of P manihoti (Castillo & Bellotti,
1990), and it can be expected that they evoke similar reactions in the cassava plant.
However, studies with the asexual strain of A. diversicornis of South America by Hof
stee et al. (1993) showed no response by this parasitoid to the odor of P herreni in
fested cassava plants (var. Odungbo). A better understanding of the interactions
between cassava plants, mealybugs, and parasitoids requires more behavioral as well
as chemical studies.
In this paper, we report on olfactometer studies with the three encyrtid parasitoids
reared at CIAT, A. vexans, A. diversicornis (asexual strain), and A. coccois. The studies
were conducted to determine whether these parasitoids are attracted to odors that
may emanate from cassava plants infested by P herreni.

September, 1997

Bertschy et al.: Attraction of cassava mealybugparasitoids 385



CMC40 cassava stakes (20 cm long) were planted every week in pots and kept in
a screened compartment, where they were subjected to natural weather conditions at
Palmira, Colombia, though protected from rain. The plants were used in experiments
when they carried 10-30 leaves (approximately 6 weeks after planting).


The cassava mealybug, P herreni was reared at CIAT on potted cassava plants
(var. CMC40). Every week 30-40 cm high plants were infested with 15 mealybug
ovisacs, as described by van Driesche et al. (1987). The plants were separated in dif
ferent cages based on the age of the mealybugs they carried.
The parasitoids, A. vexans, A. diversicornis and A. coccois were continuously
reared at CIAT on mealybug-infested cassava plants (var. Mcol 1505). The colonies of
A. vexans and A. coccois were initiated with insects collected in Venezuela in 1990 and
the colony of A. diversicornis with insects from Colombia (1984). The colonies were
maintained in a greenhouse at 35'C and under natural light conditions.

The Olfactometer

A Y tube olfactometer similar to the one first described by Sabelis & van de Baan
(1983) was used in our experiments (Fig. 1). Two arms of a glass Y-shaped tube were
connected to glass chambers (6.5 cm diam.) in which odor sources could be placed. Ac
tivated charcoal filtered air at a rate of 400 ml/min was pushed into each glass cham-
ber. To avoid visual distractions and to diffuse the light, a wooden frame covered with
white cloth was placed around the Y-tube. A lamp (100 watt) was placed outside this
visual barrier opposite from the entrance where the insects were introduced. As these
parasitoids are attracted by light, the lamp helped to induce the insects to walk up
wind in the direction of the odor sources. When a female reached the center of the Y-
tube, where the three arms met, she could choose one of the offered odors.

Odor Sources

Every week ovisac-infested cassava plants were transferred into a greenhouse,
where they were kept in nylon cages for three weeks before being used for the Y tube
experiments. Control (healthy) plants were transferred weekly from the screened
compartment and enclosed in a nylon screen cage in the same greenhouse as the in
fested plants. Care was taken that no mealybugs came in contact with healthy plants.
To serve as an odor source, two leaves of either infested or healthy plants were cut off
and the cut ends were wrapped in wet cotton wool. The leaves were carefully placed
in one of the odor chambers. The infested leaves that were selected carried honeydew
and sooty mold, as well as mealybugs and exuviae.
Experimental Procedure

On the day of each experiment, parasitoid females were removed from their cage
and kept in a glass jar (400 ml) with some honey. Thejar was placed in the air-condi
tioned chamber (28-30'C) where the experiments would take place. The insects were
left one or two hours in their new environment to become adjusted. Before each olfac
tometer test, female parasitoids were allowed to parasitize a mealybug on a cassava
leaf. An infested cassava leaf was placed upside down in a petri dish and several fe

Florida Entomologist 80(3)

with valves

Odor chamber

J C,* -- Zone 4
Zone 3

'"/\ -Zone 2

/ Zone 1

No choice area

Fig. 1. Diagram of the olfactometer set-up. Drawing by Urs Lengwiler.

males were introduced and observed until they had parasitized, or at least stung a
mealybug. The parasitoids were given this experience as it may increase their respon
siveness to host-related odors (Turlings et al., 1993; Vet et al., 1995, Steinberg et al.,
1992). The parasitoids were then captured in a gelatin capsule and kept there for 10
to more than 60 minutes. Before each Y tube test, the gelatin capsule was opened and
inserted at the base of the Y tube. Females were introduced and were observed indi



September, 1997

Bertschy et al.: Attraction of cassava mealybugparasitoids 387

vidually in the olfactometer and used only once. The odor sources were reversed each
time three wasps had been tested.

Evaluation of Choices
A stopwatch was started when the insect left the gelatin capsule. The female was
allowed 5 minutes to walk up the no-choice-area (Fig. 1) to reach the center of the ol
factometer, which is the area where the three arms meet. If a female did not reach this
center within 5 minutes, she was counted as a "no-choice". For the other females, the
observation was stopped 5 minutes after they had made it to the center, or after they
had reached the end of one of the arms. Each arm, was divided into four zones (Fig.
1), which measured 8, 6, 6, and 3 cm, respectively.
A female had to enter at least zone 2 (Fig. 1) to be considered to have made a
choice. A few females switched arms after reaching zone 2. In those cases, females
were considered to choose the arm which they entered the furthest. For statistical
analyses, a chi-square test was applied, using the total number of females that made
a choice for a particular odor (a = 0.05).


The Effect of Wasp Age

It has been shown that the responsiveness to odors may change when parasitoids
get older (e.g. Thorpe & Caudle, 1938; Steinberg et al., 1992). To determine the opti
mal age of our parasitoids for olfactometer bioassays, parasitoid females of different
ages were tested. Newly emerged wasps were isolated daily at about noon and trans
ferred to Plexiglas" cages in which they were provided honey and water. The insects
remained in the cage until they had reached a certain age. Six different age classes
were tested, varying from 0.5 to 6.5 days after emergence. Each wasp was given an
oviposition experience, and then introduced into the olfactometer, in which they had
a choice between the odors of infested and healthy cassava leaves.
Responsiveness, i.e. proportion of females that made a choice, did not decrease
with increasing age of females. Overall it was high for A. diversicornis with an aver
age of 73% and medium to low for both A. vexans and A. coccois with an average of re
spectively 49 and 48% of the responding females.
Preference for an odor source changed in two of the three species (Fig. 1-3). In A.
vexans and A. diversicornis, the preference for the odor of infested leaves over odor of
healthy leaves was age dependent and significant for young females only. Of the
younger (1.5-2.5d old) A. vexans females, 80% preferred infested cassava leaf odors (x2
7.2, P < 0.01). The youngest A. diversicornis (0.5-1.5d) showed the clearest prefer
ence (82.6%) for the odor of infested leaves over the odor of healthy leaves (2 = 9.78,
P < 0.005), but 17.4% of the females that made a choice switched between arms before
making a final decision. The 1.5 to 2.5-day-old A. diversicornis switched arms much
less (3.8%), but exhibited a weaker preference for odors of infested leaves (69.2%, X2
3.85, P < 0.05). The older wasps showed no significant preference. All age classes of A.
coccois did not differentiate between infested and healthy plant odors. Like A. diver
sicornis, A. coccois walked a lot in the olfactometer, often switching between arms
(26.3% of the choosing females).

The Role of Plant Odors
In a subsequent series of experiments we more specifically determined the relative
attractiveness of healthy and infested cassava leaves. Based on the results of the pre

Florida Entomologist 80(3)

September, 1997


a n
00) Na

- 0 0 ^ ^^ If) Nn N 0) ^

.E E
*a- a' s^a .

C D :3 N n cco

S~~~~~a~~ o- o ni ^ S

t ^ -Ut ^ C-

ac o a co -3

---------- :--

i B a-

in Mr c

oC Qr o g, ^ 5

O 0c M

4 NA CO CA o LiS g
o> c 0 C


a) C )
LD L 6 h to to a



>,s Z
) 6 C-c. au
CU -
f0 C 0 0
, ~C "

a,00.0 E 0
O r C2 S
-r C -

Bertschy et a.: Attraction of cassava mealybugparasitoids 389


- O No
ii Y~ C7 C 5


~4 LnYI mn "n
'50~" X om

S S S S .5 s

LO 4).

pg -c

o C


U) LO 0 in
.s1 .s m-c

*S0~~L rt Ct c o co d S-
0 0 II I II II H II0 0 ~ 2 >

r VN Q.I
-o I ts s ao
- 1 CM O- '9 (9 r- *o--

-?'S S 3

G)HH Mn in in
oH 0; -a cvQ 't s

*~ ~~ S O

BO n co
"g ^ -s

.^ U r C c

---------L J - s I j i
w~~ ~ ~ a Mi ci ded"
^ O /
u7 y ? r
______________ If ^ ^
go |1|^

Florida Entomologist 80(3)

vious experiments, only females that were 1.5 to 2.5 days old were used. On a given day
three different pairs of odor sources were tested, namely "Infested vs. Healthy", "In
fested vs. Blank", and "Healthy vs. Blank". In the case of "Blank", one arm introduced
clean air that had passed through an odor chamber withjust a piece of wet cotton wool.
For each pair of odor sources, 4 to 6 insects per day were individually tested in the Y
tube. Occasionally, another series of 6 insects per odor source was tested the same day.
A. vexans females were significantly attracted to infested cassava leaves compared
to healthy ones or a blank (Fig. 5a). Healthy leaf odors were less attractive, since only
64.5% of the females responded in the "Healthy vs. Blank" test without showing a sig
nificant preference for one of the two odor sources (x2 2.5, P > 0.05).
A. diversicornis females were significantly attracted to infested and healthy cas
sava leaves when offered against a blank. They also showed a significant preference
for infested cassava plant odors over healthy ones (x2 6.08, P < 0.025).
Only 51.7 to 58.3% of A. coccois females made a choice, but these were significantly
attracted by healthy and infested plant odors when offered against a blank (2 = 7.53,
P < 0.01 and 2 11.65, P < 0.001). In the "Infested vs. Healthy" test, the choosing fe
males very often switched sides before going up one arm, and they showed no signifi
cant preference for either odor source (x2 0.26, P > 0.05) (Fig. 5c).


The preference of female wasps to plant odors in the olfactometer was age depen
dent for A. vexans and A. diversicornis. The younger age classes of both these species
significantly preferred the odor of infested leaves, while older females showed no par
ticular preference. The preferences exhibited by young A. vexans and A. diversicornis
may have been due to the experience that the wasps received with an infested leaf just
before their introduction into the olfactometer. During such an experience the females
may learn to respond to the odors that they encounter through a process of association
(Turlings et al., 1993; Vet et al., 1995), which may be age dependent. Some parasitoids
only learn as young adults (Kester & Barbosa, 1991), which could explain why older
wasps did not make a distinction in our tests. It is possible that if these older wasps
had been given an experience at a younger age, they would have shown a preference
as well. In the subsequent experiments only younger females were used.
For A. coccois, the lack of preference of females of any age class may be due to the
particular choice offered. This species obviously did not distinguish between infested
and healthy cassava leaves. An alternative choice, such as between plants and a blank
might have revealed a similar age dependency of the response as found for the two
other species.
All three species distinguished between plant material and clean air (blank). A.
vexans showed only a marginal attraction to healthy leaves, but was strongly at
tracted to infested leaves. A. diversicornis was attracted to healthy leaves, but pre
ferred the odor of infested leaves. A. coccois was also attracted to both healthy and
infested leaves, but did not distinguish between these two odor sources. These differ
ences in response of the three encyrtid parasitoids suggest that they may employ dif
ferent foraging strategies. A. vexans and A. diversicornis recognized odors that are
specifically associated with mealybug infestation. A. coccois, on the other hand, ap
peared to respond only to general cassava plant odors. It remains unknown if A. vex
ans and A. diversicornis are attracted to odors emanating directly from the mealybugs
or if the infested plants emit the attractive odors.
In the petri dish, where females were experienced by giving them the opportunity
to walk over a cassava leaf and sting a mealybug, A. vexans walked slower, but showed
a more direct orientation towards mealybugs. This slower, but directed searching be

September, 1997

Bertschy et al.: Attraction of cassava mealybugparasitoids 391

-c .a a;

c c -o

-7 q |


tO ~ ~ ~ ~ ~ ~ i (D i c o -'
r- s coj c', ff r- ^ c c d 9

cr to r -m
tt~ cvi & cc
IO~~~" <3 0 -h ~0 ,

~- c r i~ i E -Q i S

-- CI (0D OD t

----------- 2 U 1 C-

^." "k'. *^ ) co

rr I |1||a


N- 0o (0 0o r- N COC, *S
CJ ~~

'-'~~ -^ a

_h _. ~ ^
---- --------------- m a --S

u, q I w Uo U)
c N M c dO
0D LA N tq ^ Lb
6 oij .q L

o C
A -

Ur Qt
U Uo
/ 5

Lo -R


C *

Florida Entomologist 80(3)

September, 1997

vi (
, ,


S- 0
-P d e

S V, c 0 M 0'

s? ~ i i

S rr m m i e oac
0 C
C mm mrs .
Cu C i C

I g
m i .



il| ^L
c^- C
s^ ^
s i

o CC

0t i M ?
C| >

0 CC -4 C
>-^ CO y0c

8 r u
-.CZ a,
S : -S-

^^ c
^- :a-

s ^ s
C3C';2 *3
.--c C S

-c 2 I -C

* ;~ co -

y-i C- 0 i

nS nS S,

"" C
So w E

- C
CC -S 5~ -

n*c c ?
C1 C
o a C-
" l-K-

St- COC -

> C- -4^ C-
on~ C ~o:
i~ cf-



IIc'Nc- oJ-
c* H

Bertschy et al.: Attraction of cassava mealybugparasitoids 393

havior was also observed in the olfactometer. A. vexans was clearly attracted to the in
fested cassava plants, but not to the odors of healthy plants. After it found a
mealybug, this solitary parasitoid needed only a few seconds to parasitize it. A. coccois
is gregarious and took up to an hour to parasitize a host. It spent a lot of time walking
rapidly around the petri dish and had fewer encounters with mealybugs. Also in the
olfactometer, this species walked much faster and in more different directions than
the other species, particularly when the females were given the choice between in
fested and healthy plant odors. This fast moving species did not readily distinguish
between the odors of infested and healthy leaves.
The reported host preference of these parasitoids may explain their behavior in
the olfactometer to some extent. A. vexans prefers P herreni over a related species,
Phenacoccus_. .. (= madeirensis) (CIAT, 1990). It has been most frequently recov
ered from P herreni on cassava, but its host range does include other Phenacoccus
species on different plants (Noyes & Ren, 1995). Pijls & van Alphen (1996) studied the
specificity of a sexual strain of A. diversicornis on cassava. It appears to be specific to
P herreni and P manihoti. An asexual strain from Venezuela has been shown to prefer
P herreni over P. ..- (= madeirensis) (Van Driesche et al., 1987). A. coccois seems
to be the most polyphagous of the three. It parasitizes Pseudococcidae species of dif
ferent genus such as Oracella acuta (Homoptera: Pseudococcidae) on loblolly pine (Pi
nus taeda L.) (Clarke et al., 1990). On cassava plants, it parasitizes P herreni and P
madeirensis more or less equally (CIAT, 1990). As a generalist, A. coccois may be more
responsive to general plant odors, while the more specialized wasps, A. vexans and A.
diversicornis, may have adapted to exploit odors that are specifically associated with
the presence of mealybugs on cassava.
It remains to be determined if cassava volatiles play an important role in the spe
cific attraction to infested plants, or if the mealybug and its by-products emit odors
that are attractive. It is known that some herbivores induce reactions in plants that
make them highly attractive to some parasitic wasps (Turlings et al., 1995). Nadel &
van Alphen (1987) found evidence that mealybug-infested cassava plants also release
odors that are attractive to the parasitoid A. lopezi. The odors probably come from the
plant itself, as the parasitoid was not attracted by the mealybug and its by-products.
Van Alphen et al. (1990) also found an attraction to P manihoti infested cassava
plants by A. diversicornis. Unlike our results, the females were not attracted by
healthy cassava plants but showed a clear attraction to uninfested leaves taken from
a partially infested plant, which suggests that the infested plant emits attractants.
Little is known about the exact source and identity of parasitoid attractants. Our on
going experiments aim to determine the exact role of the cassava plant in the foraging
success of the parasitoids in order to consider and exploit this role in further control
measures against the cassava mealybug.


We thank Carlos Nanez for maintaining and rearing mealybugs and parasitoids;
Jair L6pez for his collaboration in the preparation of the experiments. We also thank
Letizia Mattiacci and Lincoln Smith and three anonymous reviewers for their critical
comments on the manuscript. The project was supported by a grant from the Swiss
Center for International Agriculture, ETH Zurich.


of Epidinocarsis lopezi and E. diversicornis: A possible explanation for the

Florida Entomologist 80(3)

failed introduction of E. diversicornis against cassava mealybug Phenacoccus
manihoti into Africa. Med. Fac. Landbouww. Rijksuniv. Gent 55: 276-287.
BELLOTTI A. C. 1983. More on the mealybug: a major cassava pest. Cassava newslet
ter. 7: 14.
BELLOTTI, A. C., J. A. REYES, AND A. M. VARELA. 1984. Observations on cassava mea
lybugs in the Americas; their biology, ecology and natural enemies. Plant pro
tection. Symposium of the International Society for Tropical Root Crops, 6th
Lima, 1983. Lima, Peru, CIP p. 339-352.
CASTILLO, J., AND A. C. BELLOTTI. 1990. Caracteres diagnosticos de cuatro species de
piojos harinosos (Pseudococcidae) en cultivos de yuca (Manihot esculenta) y ob
servaciones sobre algunos de sus enemigos naturales. Revista Colombiana de
Entomologia. 16: 33-43.
CALATAYUD, P. A., M. TERTULIANO, AND B. LE RU. 1994. Seasonal changes in second
ary compounds in the phloem sap of cassava in relation to plant genotype and
infestation by Phenacoccus manihoti (Homoptera: Pseudococcidae). Bull. Ent
Res. 84: 453-459.
CIAT (Centro Internacional de Agricultura Tropical). 1982. Annual Report, CIAT,
Cali, Colombia
CIAT (Centro Internacional de Agricultura Tropical). 1983. Annual Report, CIAT,
Cali, Colombia.
CIAT (Centro Internacional de Agricultura Tropical). 1984. Annual Report, CIAT,
Cali, Colombia.
CIAT (Centro Internacional de Agricultura Tropical). 1987. Annual Report, CIAT,
Cali, Colombia.
CIAT (Centro Internacional de Agricultura Tropical). 1988. Annual Report, CIAT,
Cali, Colombia.
CIAT (Centro Internacional de Agricultura Tropical). 1990. Annual Report, CIAT,
Cali, Colombia.
CIAT (Centro Internacional de Agricultura Tropical). 1992. Cassava Program 1987
CLARKE, S. R., G. L. DEBARR, AND C. W. BERISFORD. 1990. Life history of Oracella
acuta (Homoptera: Pseudococcidae) in Loblolly pine seed orchards in Georgia.
Environ. Entomol. 19: 99-103.
Cox, J. M., AND D. J. WILLIAMS. 1981. An account of cassava mealybugs (Hemiptera:
Pseudococcidae) with a description of a new species. Bull. Ent. Res. 71: 247-258.
Plant strategies of manipulating predator-prey interactions through allelochem
icals: prospects for application in pest control. J. Chem. Ecol. 16: 3091-3118.
preferences of two encyrtid parasitoids for the Columbian Phenacoccus spp. of
cassava mealybugs. Entomol. Exp. Appl. 43: 261-266.
HERREN, H. R., AND P. NEUENSCHWANDER 1991. Biological control of cassava pests in
Africa. Annu. Rev. Entomol. 36: 257-83.
HOFSTEE, S. K., J. W. A. M. PIJLS, AND J. J. M. VAN ALPHEN. 1993. The attractiveness
of Cassava infested with different Phenacoccus- (Cassava Mealybug) Species to
two Epidinocarsis-species. Med. Fac. Landbouww. Univ. Gent. 58: 543-549.
KESTER, K. M., AND P. BARBOSA. 1991. Postemergence learning in the insect parasi
toid Cotesia congregate (Say) (Hymenoptera: Braconidae). J. Insect Behav. 4:
MATILE-FERRERO, D. 1977. Une cochenille nouvelle nuisible au manioc en Afrique
6quatoriale, Phenacoccus manihoti n. sp. (Homoptera, Coccoidea, Pseudococ
cidae). Ann. Soc. Ent. Fr. (N.S.). 13: 147-152.
NADEL, H., AND J. J. M. VAN ALPHEN. 1987. The role of host and host-plant odors in
the attraction of a parasitoid, Epidinocarsis lopezi, to the habitat of its host, the
cassava mealybug, Phenacccus manihoti. Entomol. Exp. Appl. 45: 181-186
BOKONONGANTA, R. ALLOMASSO, AND I. OKON. 1991. Biological control of the

September, 1997

Bertschy et al.: Attraction of cassava mealybugparasitoids 395

cassava mealybug Phenacoccus manihoti (Horn., Pseudococcidae) by Epidi
nocarsis lopezi (Hym., Encyrtidae) in West Africa, as influenced by climate and
soil. Agric. Ecoysyt. Environ. 32: 39-55.
NOYES, J. S., AND H. REN. 1995. Encyrtidae of Costa Rica (Hymenoptera: Chalci
doidea): the genus Aenasius Walker, parasitoids of mealybugs (Homoptera:
Pseudococcidae). Bull. Nat. Hist. Mus.Lon. Entomol. 64: 124-125
PICARD, F., AND E. RABAUD. 1914. Sur le parasitisme externe des Braconides. Bulletin
de la Societe Entomologique de France. 83: 266-269.
PIJLS, J. W. A. M, AND J. J. M. VAN ALPHEN. 1996. On the coexistence of the cassava
mealybug parasitoids Apoanagyrus diversicornis and A. lopezi in their native
South America. Bull. Ent. Res. 86: 5159.
SABELIS, M. W., AND H. E. VAN DE BAAN. 1983. Location of distant spider mite colo
nies by phytoseiid predators: Demonstration of specific kairomones emitted by
Tetranychus urticae and Panonychus ulmi. Ent. exp. Appl. 33: 303-314.
Survival, development and reproduction of Hyperaspis notata Mulsant strains
(Coleoptera, Coccinellidae) under different food supply. J. Appl. Entomol. (in
The functional responses of two Hyperaspis notata strains to their prey, the cas
sava mealybug Phenacoccus manihoti. Mitt. Schweiz. Ent. Ges. (in press).
DORN. 1997c. Biology of Hyperaspis notata Mulsant (Coleoptera, Coccinel
lidae), an introduced predator of the cassava mealybug in Africa: Life tables at
different temperatures. Ent. Exp. App. (in press).
STEINBERG, S., M. DICKE, L. E. M. VET, AND R. WANNINGEN. 1992. Response of the
braconid parasitoid Cotesia (= Apanteles) glomerata to volatile infochemicals:
effects of bioassay set-up, parasitoid age and experience and barometric flux.
Entomol. Exp. Appl. 63: 163-175.
THORPE, W. H., AND H. B. CAUDLE. 1938. A study of the olfactory responses of insect
parasites to the food plant of their host. Parasitology 30: 523-528.
THORPE, W. H., AND F. G. W. JONES. 1937. Olfactory conditioning in a parasitic insect
and its relation to the problem of host selection. Proc. Royal Soc. B. 124: 56-81.
TURLINGS, T. C. J., J. H. TUMLINSON, AND W. J. LEWIS. 1990. Exploitation of herbi
vore-induced plant odors by host-seeking parasitic wasps. Science. 250: 1251
1993. Learning of host-finding cues by hymenopterous parasitoids, pp. 51 78 in
D. R. Papaj and A. C. Lewis, [ed.]. Insect Learning: Ecological and Evolutionary
Perspectives. Chapman and Hall, New York.
TUMLINSON. 1995. How caterpillar-damaged plants protect themselves by at
tracting parasitic wasps. Proc. Natl. Acad. Sc. USA. 92: 4169-4174.
VARGAS, O., AND A. C. BELLOTTI. 1984. Perdidas en rendimiento causadas por Phen
acoccus herreni Cox & Williams en dos clones de yuca. Revista Colombiana de
Entomologia. 10: 41-46.
VET, L. E. M., AND M. DICKE. 1992. Ecology of infochemical use by natural enemies in
a tritrophic context. Annu. Rev. Entomol. 37: 141-172.
VET, L. E. M., W. J. LEWIS, AND R. T. CARDE. 1995. Parasitoid foraging and learning,
pp. 65-101 in W. J. Bell and R. T Carde, [ed.]. Chemical Ecology of Insects.
Chapman and Hall, New York.

Florida Entomologist 80(3)


Department of Entomology, Louisiana State University Agricultural Center
Louisiana Agricultural Experiment Station, Baton Rouge, LA 70803, USA


Population density estimates were used to determine the effectiveness of a com
mercial formulation of Bacillus thuringiensis Berliner subspp. kurstaki and aizawai
(Agree 50 WP) and host plant resistance in three cabbage cultivars against the dia
mondback moth, Plutella xylostella (L.). Cabbage plots treated with Agree 50 WP" had
significantly fewer larvae per plant compared with untreated ones. The ranking from
most to least susceptible of the three main cabbage cultivars grown in Jamaica was
KY Cross' > 'Early Jersey' > Tropicana'. These findings provide evidence that a new
cabbage hybrid, Tropicana', and products containing effective strains of B. thuring
iensis may be successfully used for P. ,' ... 'i I 1,1 I: ,I-. in Jamaica.

Key Words: P.... r. ',. :,ee 50 WP plant resistance, Jamaica


Fueron usados estimados de la densidad poblacional para determinar la efectivi
dad de una formulacidn commercial de Bacillus thuringiensis Berliner subspp. kurstaki
y aizawai (Agree 50 WP ), y la resistencia de tres cultivares de col a la polilla Plutella
xyllostella (L.). Las parcelas de col tratadas con Agree 50 WP" tuvieron significativa
mente menos larvas por plant que las no tratadas. El rango de susceptibilidad en or
den creciente de los tres cultivares mas usados en Jamaica fue "KY Cross" > "Early
Jersey" > "Tropicana". Estos hallazgos muestran que un nuevo hibrido de col, "Tropi
cana", y products conteniendo cepas efectivas de B. thuringiensis podrian ser exito
samente usados para el manejo de P xylostella en Jamaica.

Cruciferous vegetables grown in Jamaica and other Caribbean islands are suscep
tible to damage by many insect pests: armyworms, Spodoptera spp., cabbage looper,
Trichoplusia ni (Hubner), cabbage white butterfly, Pieris rapae L., and diamondback
moth, Plutella xylostella (L.). A complex of these pests occurs whenever these crops
are grown for commerce, and their control is a prerequisite for meeting quality stan
dards for damage and pest-free produce. Populations of P xylostella frequently ac
count for 75% of the insect pest population and cause crop loss of up to 90%, making
it the key insect pest from an economic standpoint (Salinas 1986, Alam 1992).
Historically, farmers have relied primarily on multiple applications of broad-spec
trum insecticides for control of P xylostella in Jamaica. Alam et al. (1987) recom-
mended an action threshold of six larvae per plant at the post-transplanting stage of
cabbage. Over 18 different insecticides have been used since 1972 (Walton 1989), and
between 20-22 insecticide applications over a growing season are not uncommon. As
a result, many insecticides from the organophosphate, carbamate, and pyrethroid
groups are now ineffective because of insecticide resistance (Alam 1992, Robinson et

September, 1997

Ivey & Johnson: Efficacy ofB. thuringiensis

al. 1995). Reports of low efficacy of Biotrol" and Thuricide" (products containing the
kurstaki strain of Bacillus thuringiensis Berliner) in controlling P xylostella, presum
ably due to insecticide resistance, led Alam (1992) to question their reliability. Be
cause of pest management problems, environmental degradation, and occupational
and public health risks associated with insecticides, it is imperative to find an inte
grated pest management (IPM) approach for P xylostella management which utilizes
tactics such as host plant resistance and microbial control.
The utility of host plant resistance as an insect pest management tactic is well es
tablished (Painter 1951, Lim 1992). Dickson et al. (1984, 1986) reported the release of
four cabbage breeding lines possessing resistance to P xylostella. Results of genetic
and other studies indicated that the host plant resistance exhibited by these cabbage
breeding lines was associated with the glossy dark-green leaf found in the cauliflower
Plant Introduction (PI) 234599 (Dickson et al. 1990, Stoner 1990, Eigenbrode & Shel
ton 1992). However, P xylostella-resistant cabbage cultivars are not generally com-
mercially available. Use of microbial agents for controlling P xylostella has been most
successful with B. thuringiensis; certain strains of this bacteria are highly effective
against early instars (Hofte & Whiteley 1989). Furthermore, B. thuringiensis is envi
ronmentally benign and non-toxic to beneficial organisms, many of which are natural
enemies of P xylostella. Because the kurstaki strain of B. thuringiensis is already of
questionable reliability in Jamaica (Alam 1992), and field resistance in P xylostella
has been reported in the Philippines (Kirsch & Schmutterer 1988), Hawaii (Tabash
nik et al. 1990), Malaysia (Syed 1992), mainland USA (Shelton et al. 1993), and in
Central America (Perez & Shelton 1997), it is unwise for farmers to rely solely on this
microbial insecticide. To thwart the development of resistance to this insecticide and
preserve its longevity and effectiveness, a more integrated approach should be devel
oped. Combining the use of B. thuringiensis and host plant resistance for P xylostella
control is plausible. A new product, Agree 50 WP" wettablee powder), containing both
kurstaki and aizawai strains of B. thuringiensis and a new cabbage cultivar, 'Tropi
cana', reputed resistant to P xylostella, has recently become available to farmers. Be
fore the advent of this new cultivar, two hybrids (KY Cross' and KK Cross') and an
open-pollinated cultivar ('Early Jersey') were available to growers in Jamaica. In
1995, marketing of Tropicana' began by the leading distributor of agricultural inputs
with claims of resistance to P xylostella. However, the relative resistance of these cul
tivars to P xylostella, under local conditions, has not been empirically studied. There
fore, toward achieving our long range goal of IPM of P xylostella, the objective of this
study was to determine the efficacy of Agree 50 WP" and evaluate the relative resis
tance of cabbage cultivars grown in Jamaica to P xylostella.


In September 1995, a field experiment was initiated on the farm of the College of
Agriculture, Science, and Education, Port Antonio, Portland, Jamaica. Three cabbage
cultivars, Tropicana', 'Early Jersey' (Petoseed, Saticoy, CA), and KY Cross' (Takii
Seed, Kyoto, Japan), were subjected to two different treatment regimes: Agree 50 WP"
(Ciba-Geigy, Greensboro, NC) applied weekly and untreated controls. Cultivars and
treatments were replicated four times in a completely randomized experimental de
sign with a split-plot treatment arrangement. Main plots were cultivar and sub-plots
were Agree 50 WP" and untreated controls. Individual plots were 2.73 m x 4.45 m and
were separated by a distance of 1.52 m. Four-week-old seedlings of the three cabbage
cultivars, raised in outdoor seedbeds, were planted 0.45 m apart on raised beds spaced
0.91 m apart. Standard cultivation practices for cabbage production were employed.
Plots were sampled for P xylostella once per week for seven weeks, between 18 No

Florida Entomologist 80(3)

vember and 30 December, by counting larvae on the leaves of 10 plants in each plot.
Agree 50 WP" [3.8% (AI) (25,000 IU per ml) (B. thuringiensis subspp. kurstaki and
izirnl la)]. was applied to the appropriate plots once per week, at the rate of 83 g in 15
liters of water using a Solo 475 Knapsack Sprayer (Solo Incorporated, Newport News,
VA). The data were subjected to analysis of variance using PROC GLM (SAS Institute


P xylostella larval population density per plant was significantly affected by cab
bage cultivar (F= 20.94; df= 2, 18; P= 0.0001) and insecticide treatment (F= 52; df
1, 18; P= 0.0001). The ranking of cultivars from most to least susceptible to P xy-
lostella was KY Cross' > 'Early Jersey' > Tropicana'. Plots treated weekly with Agree
50 WP" had significantly (F= 52; df= 1, 18; P= 0.0001) fewer larvae per plant com
pared with untreated plots (Table 1).
The interaction between cabbage cultivar and insecticide treatment was margin
ally significant (F= 3.38; df= 2, 18; P = 0.0567). Further examination of this interac
tion was done using PROC PLOT (SAS Institute 1989). The cell means for the levels
of the two factors, insecticide treatment and cabbage cultivar, as shown in Table 1, in
dicated that the Tropicana' cultivar supported fewer larvae per plant across insecti
cide treatments compared with the other two cultivars.


Based on larval population density estimates, the Tropicana' cultivar was supe
rior to the other two cultivars in resisting attack from P xylostella. To our knowledge,
before this study, cabbage cultivars available to growers in Jamaica had never been
evaluated under local conditions regarding their relative susceptibility to P xylos
tella. The existence of host plant resistance to P. xylostella in a commercially culti
vated cabbage cultivar, such as Tropicana', is significant because of the economic
importance of this insect pest related to the intractable problem of its widespread de
velopment of resistance to conventional insecticides. Several authors have reported
on breeding and the potential for using resistant cabbage cultivars for P xylostella
management (Dickson et al. 1984, 1986, 1990, Stoner 1990, Eigenbrode & Shelton
1992), but host plant resistance in a commercially available cabbage cultivar has hith
erto not been reported.
It appears that the ability of the Tropicana' cabbage cultivar to resist P xylostella
is based on leaf texture; the epidermis of its leaves is relatively thicker, especially as
plants approach the heading stage, compared with those of the other two cultivars.
Importantly, there have not been any reports of complaints from consumers regarding
the texture of Tropicana'. Because first instars of P xylostella are leafminers and
must tunnel into the leaf to feed, the thicker epidermis of Tropicana' may have been
too great a challenge for the mandibles of neonate larvae. These neonates may starve
to death, desiccate, drown or be washed from leaves, and be vulnerable to predators.
Eigenbrode and Shelton (1992) found that neonate P xylostella larvae had greater
movement on resistant cabbage breeding lines than on susceptible ones. Also, Tanton
(1962) found that leaf texture affects the number of nibbles and subsequent leaf area
of Brassica rapa L. consumed by Phaedon cochleariae F In addition, Iheagwam (1981)
reported that penetrability of the leaf tissue of Brassica oleraceae L. influences the de
gree of exploitation by Aleyrodes brassicae Walker. And, Martin et al. (1975) found a
negative correlation between internode hardness of Saccharum officinarum L. and
susceptibility to attack by neonate Diatraea saccharalis (F) larvae.

September, 1997

Ivey & Johnson: Efficacy ofB. thuringiensis


No. larvae per plant
Overall cultivar
Cultivar Agree 50 WP" Untreated means + SEM2

Tropicana 0.30 + 0.21 1.55 +0.39 0.93 +0.31
Early Jersey 0.88 +0.25 4.23 +0.33 2.55+ 0.66
KY Cross 2.23 +0.70 4.95 + 0.40 3.59+ 0.64
Overall treatment means + SEM' 1.13 + 0.34 3.58 + 0.48

Treatment means significantly different (ANOVA Ftest; a = 0.05).
Cultivar means significantly different (Waller Duncan K-ratio t test; a = 0.05).

Agree 50 WP" proved to be effective in controlling P xylostella. This was probably
due primarily to the presence of the aizawai strain of B. thuringiensis. There are dif
ferences in the crystal protein toxin profiles of B. thuringiensis subsp. kurstaki and
subsp. aizawai; the former produces Cry IA (a), Cry IA(b), Cry IA(c), Cry IIA, and Cry
IIB, whereas the latter produces Cry IA (a), Cry IA(b), Cry IC, Cry ID, and Cry IIB
(Hofte & Whiteley 1989). Shelton et al. (1993) found differential responses between R
xylostella populations treated with two formulations containing B. thuringiensis
subsp. kurstaki (Javelin WG" and Dipel 2X ) and populations treated with B. thuring
iensis subsp. aizawai (ZenTari). Commercial formulations of B. thuringiensis subsp.
kurstaki, for example, Biotrol" and Thuricide have been used in Jamaica for many
years to control P xylostella but their reliability is now questionable (Alam 1992).
However, formulations of B. thuringiensis subsp. aizawai have only recently begun to
be more widely used. So, P xylostella populations have no prior exposure to this strain
of B. thuringiensis. In fact, B. thuringiensis subsp. aizawai was first used against P
xylostella in Jamaica in 1995.
The existence of a significant interaction makes it necessary to exercise caution
when making statements about the main effects (cabbage cultivar and insecticide
treatments), even though both were statistically significant with Pvalues of 0.0001
(Freund & Wilson 1993). The consistently lower numbers of larvae on the 'Tropicana
cultivar, compared with'Early Jersey' and 'KY Cross', across plots treated with Agree
50 WP" and in untreated plots, clearly show that the 'Tropicana' cultivar was superior
to the other two cultivars in resisting P xylostella. Also, from the interaction between
cabbage cultivar and insecticide treatment, it can be inferred that the 'Tropicana' cul
tivar may be compatibly combined with use of Agree 50 WP" for successful manage
ment of P xylostella.
Regarding the effect of these two control tactics on armyworms and cabbage
looper, past experience has shown that tactics which are successful in controlling P
xylostella simultaneously also controlled armyworm and cabbage looper populations.
Usually, insecticides are more effective against other insects in the crucifer pest com-
plex than P xylostella. In fact, for the duration of the study, armyworms and cabbage
loopers were not encountered.
Considering the low efficacy and tenuous reliability of commercial formulations of
B. thuringiensis subsp. kurstaki against P xylostella in Jamaica (Alam 1992), the ef
fectiveness of Agree 50 WP" (B. thuringiensis subspp. kurstaki and aizawai) seen in
this study makes continued use of toxins of B. thuringiensis for controlling this insect
a viable option, especially when combined with the 'Tropicana' cabbage cultivar. How

Florida Entomologist 80(3)

ever, we do not recommend that formulations containing both the kurstaki and aiza
wai strains of B. thuringiensis, such as Agree 50 WP be used extensively as this may
allow for the development of resistance in P xylostella to the aizawai strain without
losing its resistance to the kurstaki strain. It might be a better strategy to alternate
both strains.


We thank Lloyd Bailey of the College of Agriculture, Science, and Education, Port
Antonio, Portland, Jamaica for help with the field work, and Terrence Thomas of the
Environmental Foundation of Jamaica from which this project received substantial fi
nancial support.
Approved for publication by the Director, Louisiana Agricultural Experiment Sta
tion as Manuscript number: 97-17-0029.


ALAM, M. M., J. REID, AND I. H. GIBBS. 1987. Diamondback moth and its control. Car
ibbean Agricultural Research and Development Institute, Factsheet No. PP-F/
ALAM, M. M. 1992. Diamondback moth and its natural enemies in Jamaica and some
other Caribbean Islands, pp. 233-243 in N. S. Talekar [ed.], Diamondback Moth
and Other Crucifer Pests, Proceedings of the Second International Workshop.
AVRDC, Tainan, Taiwan. 603 pp.
and NYIR 8329 lepidopterous pest-resistant cabbage breeding lines. Hort
Science 19: 311-312.
DICKSON, M. H., C. J. ECKENRODE, AND J. LIN. 1986. Breeding for diamondback moth
resistance in Brassica oleracea, pp. 137-143 in N. S. Talekar and T D. Griggs
[eds.], Diamondback Moth Management, Proceedings of the First International
Workshop. AVRDC, Shanhua, Taiwan. 471 pp.
1990. Selection for resistance to diamondback moth (Plutella xylostella) in cab
bage. HortScience. 25: 1643-1646.
EIGENBRODE, S. D., AND A. M. SHELTON. 1992. Resistance to diamondback moth in
Brassica: Mechanisms and potential for resistant cultivars, pp. 65-74 in N. S.
Talekar [ed.], Diamondback Moth and Other Crucifer Pests, Proceedings of the
Second International Workshop. AVRDC, Tainan, Taiwan. 603 pp.
FREUND, R. J., AND W. J. WILSON. 1993. Statistical Methods. Academic Press. San Di
ego, CA. 644 pp.
HOFTE, H., AND H. R. WHITELEY. 1989. Insecticidal crystal proteins of Bacillus thur
ingiensis. Microbiol. Rev. 53: 242-255.
IHEAGWAM, E. U. 1981. The relationship between weight of insect, age, hardness and
nitrogen content of cabbage leaves and fecundity of the cabbage whitefly, Aley
roides brassicaeWalker (Homoptera: Aleyroididae). Z. Ang. Ent. 91: 349-354.
KIRSCH, K., AND J. SCHMUTTERER 1988. Low efficacy of a Bacillus thuringiensis (Ber.)
formulation in controlling the diamondback moth, Plutella xylostella (L.), in
the Philippines. J. Appl. Entomol. 105: 249-255.
LIM, G. 1992. Integrated pest management of diamondback moth: Practical realities,
pp. 565-576 in N. S. Talekar [ed.], Diamondback Moth and Other Crucifer
Pests, Proceedings of the Second International Workshop. AVRDC, Tainan, Tai
wan. 603 pp.
MARTIN, F. A., C. A. RICHARD, AND S. D. HENSLEY. 1975. Host resistance to Diatraea
saccharalis (F.): Relationship of sugarcane internode hardness to larval den
sity. Environ. Entomol. 4: 687-688.

September, 1997

Ivey & Johnson: Efficacy ofB. thuringiensis

PAINTER, R. H. 1951. Insect Resistance in Crop Plants. University of Kansas Press,
PEREZ, C. J., AND A. M. SHELTON. 1997. Resistance of Plutella xylostella (Lepidoptera:
Plutellidae) to Bacillus thuringiensis Berliner in Central America. J. Econ. En
tomol. 90: 8793.
ROBINSON, D. E., K. M. DALIP, AND A. MANSINGH. 1995. Integrated Management of
Pests and Pesticides in the Caribbean. Department of Zoology, University of the
West Indies, Mona, Kingston. The Jamaica National Commission for UNESCO.
Kingston. 78 pp.
SALINAS, P. J. 1986. Studies on diamondback moth in Venezuela with reference to
other Latin American countries, pp. 17-24 in N. S. Talekar and T D. Griggs
[eds.], Diamondback Moth Management, Proceedings of the First International
Workshop. AVRDC, Shanhua, Taiwan. 471 pp.
SAS INSTITUTE. 1989. SAS/STAT user's guide: statistics, 5th ed. SAS Institute, Cary,
PREISLER, W. T. WILSEY, AND R. J. COOLEY. 1993. Resistance of diamondback
moth (Lepidoptera: Plutellidae) to Bacillus thuringiensis subspecies in the
field. J. Econ. Entomol. 86: 697-705.
STONER, K. A. 1990. Glossy leaf wax and plant resistance to insects in Brassica oler
acea under natural infestation. Environ. Entomol. 19: 730-739.
SYED, A. R. 1992. Insecticide resistance in diamondback moth in Malaysia, pp. 437
442 in N. S. Talekar [ed.], Diamondback Moth and Other Crucifer Pests, Pro
ceedings of the Second International Workshop. AVRDC, Tainan, Taiwan. 603
TABASHNIK, B. E., N. L. CUSHING, N. FINSON, AND M. W. JOHNSON. 1990. Field devel
opment of resistance to Bacillus thuringiensis in diamondback moth (Lepi
doptera: Plutellidae). J. Econ. Entomol. 83: 1671-1676.
TANTON, M. 1962. The effect of leaf "toughness" on the feeding of larvae of the mus
tard beetle, Phaedon cochleriae F. Ent. Exp. & Appl. 5: 7478.
WALTON, W. C. 1989. Problems associated with the control of diamondback moth in
Douglas Castle, St. Ann. Department of Crop Science, University of the West
Indies, St. Augustine, Trinidad. 60 pp.

Florida Entomologist 80(3)


United States Department of Agriculture, Agricultural Research Service

'Yakima Agriculture Research Laboratory, 5230 Konnowac Pass Road
Wapato, WA 98951

2Center for Medical, Agricultural and Veterinary Entomology, 1700 SW 23rd Dr.
Gainesville, FL 32604


Male tobacco budworm, Heliothis virescens Fab., moths released into a field cage
were recaptured in traps baited with aged 10% jaggery, a palm sugar extract. Both
male and female tobacco budworm moths were attracted to aged 10% jaggery in a
flight tunnel, exhibiting oriented flights ending in contact with the bait. Although the
bait was initially not attractive either to females in a flight tunnel or to males in a
field cage, it subsequently became attractive after one week and increased in attrac
tiveness for up to 24 days after it was made.

Key Words: attractant, feeding, Heliothis virescens, trap, behavior, sugar


Machos adults del gusano del tabaco, Heliothis virescens Fab., liberados en una
jaula de campo fueron recapturados en trampas cebadas con 10% de jaggery enveje
cido, un extract de azfcar de palma. Tanto hembras como machos adults fueron
atraidos por el cebo en un tfnel de vuelo, y ambos mostraron vuelos orientados term
nando en el contact con el cebo. A pesar de que el cebo inicialmente no fue atractivo
a las hembras en el tfnel de vuelo, o a los machos en lajaula de campo, este se torn
atractivo despues de una semana e increments su atractividad hasta los 24 dias de
haber sido preparado.

The tobacco budworm, Heliothis virescens (Fab)., is a pest of several agricultural
crops in North America, including tobacco and cotton. The principal means of moni
touring the presence of tobacco budworm is a female sex pheromone blend that is at
tractive to males (Sparks et al. 1979, Ramaswamy et al. 1985). However, such a
method is ineffectual in fields treated with female sex pheromone as a mating disrup
tant. Also, the relationship between numbers of males captured in pheromone traps
and either population levels or crop damage is not clear. Additional attractants, par
ticularly if effective for females, would be useful under such circumstances.
A variety of moths are attracted to sweet materials, presumably as a source of
adult nutrition. Sugar-rich concoctions often are used by moth and butterfly collectors
(Holland 1903, Sargent 1976). Molasses solutions have been used to trap oriental fruit
moth, Grapholita molesta (Busck), (Frost 1926) and codling moth, Cydia pomonella
(L), (Eyer 1931) in tree fruit orchards. Many noctuid moths are attracted to sugar

September, 1997

Landolt & Mitchell: Tobacco Budworn Food Attractant 403

baits (Norris 1936), although documentation of which species are attracted is lacking.
Frost (1928) captured 23,574 noctuid moths in 300 pails containing sugar baits set out
for oriental fruit moth, but did not identify them below the family level. Poisoned
sweet baits were used for control of corn earworm, Helicoverpa zea (Boddie), during
the 19th century (Ditman & Cory 1933 and references therein). The grass looper,
Mocis latipes Guenee, can be captured in traps baited with solutions of molasses or
jaggery (Landolt 1995). Jaggery is an unrefined sugar made from palm sap, used asa
cooking sweetener in some areas of subtropical and tropical Asia. Landolt (1995) also
reported the capture of 13 additional species of Noctuidae in glass traps baited with
jaggery or molasses solutions in Florida.
There are no reports of captures of tobacco budworm moths in traps with baits con
training sugars or sugar-based materials. However a great number of species of Noc
tuidae likely are attracted to sugar baits (Norris 1936), and most Noctuidae captured
in traps baited with sugar-rich materials have not been identified (e.g., Frost 1928).
The tobacco budworm moth feeds at flowers, extrafloral nectaries, artificial sugar
sources, and grass heads (Lingren et al. 1977, Ramaswamy 1990) and may be at
tracted to sugar baits.
We report here the attraction of male and female tobacco budworm moths to solu
tions ofjaggery, and we also determined the optimum age of the bait for attractiveness
to moths in a flight tunnel. This work demonstrates the upwind orientation of tobacco
budworm in response to food baits and provides a convenient assay system for pursu
ing the isolation and identification of attractive volatile chemicals emanating from
sugar baits. It is hoped that such compounds may be useful as attractants for tobacco
budworm as well as other pestiferous species of moths.


Insect Rearing and Handling

Tobacco budworm pupae were obtained from the laboratory colony maintained at
the Gainesville, Florida, United States Department of Agriculture, Agricultural Re
search Service laboratory. Pupae were sorted by sex and were held in screened cages
(25 x 25 x 25 cm) for adult emergence. Pupae were moved to new cages daily to provide
moths of discrete age groups. A waterjar was placed on the top of each cage, and each
cage was provisioned with a 60 ml paper cup containing water-soaked cotton balls.
Males and females were held in separate environmentally-controlled rooms at 24'C,
60-80% RH and a 14:10 (L:D). photoperiod with lights off at 0800 and on at 1800 hours

Field Cage Bioassay

An initial test of tobacco budworm moth response tojaggery (Indian Kolhapur Jag
gery, House of Spices Inc., Jackson Heights, NJ) was conducted in 2 large cylindrical
cages, each 2.2 m in height and 2.7 m in diameter (Calkins & Webb 1983), which were
set up in an area of lawn largely beneath the shade of a live oak tree. Pairs of glass
McPhail traps (Newell 1936) were hung from the ceiling of each cage, about 0.5 m
north and 0.5 m south of the center of the field cage. Traps were hung by a wire with
the trap opening 20 cm below the cage ceiling. One of each pair of traps in a cage was
baited with 200 ml of 10% jaggery in deionized water (5 to 16 days old) and the other
trap was baited with 200 ml of deionized water. From 20 to 30 male tobacco budworm
moths (3 to 5 days of age) were released into a field cage in late afternoon, and the
numbers of moths captured in traps were counted the following morning. Jaggery bait

Florida Entomologist 80(3)

was 5 to 16 days old when placed in the field cages. This assay was conducted 20 times,
with jaggery bait reused for replicates. Jaggery-baited and control traps were alter
nated in position with each assay replicate for both field cages. Mean trap catch data,
combined for all bait ages, were analyzed by Students t-test to determine if the
catches of moths in treatment and control traps were significantly different. Catch
data forjaggery-baited traps were also evaluated in comparison to bait age by regres
sion analysis.

Flight-Tunnel Bioassays

Two experiments were conducted using a flight tunnel to evaluate tobacco bud
worm moth responses to jaggery. The flight tunnel and room were described by
Landolt and Heath (1987). Moths were tested during the 3rd through 5th hours of the
10 h scotophase, and they were placed in the flight tunnel room 30 min before the bio
assays. Moths were tested individually. They were released from a plastic vial near
the center of the downwind end of the flight tunnel and were given 2 min to respond
to test materials placed at the upwind end of the flight tunnel. Moths were scored for
upwind oriented flights within the odor plume and for proximity or contact with the
odor source following plume tracking. The baits tested were 10% solutions of jaggery
made up as 400 ml batches and placed in open glassjars in a laboratory fume hood un
til used. For flight tunnel assays, 200 ml of solution were poured into a 9 cm plastic
petri dish supported by a ring and ring stand. A paper towel was hung into the middle
of the dish to act as a wick, increasing the surface area of the solution.
The first flight tunnel experiment was a demonstration of male and female tobacco
budworm attraction to aged jaggery. Three to four-day old unfed females were tested
to either a200 ml batch of aged (12 to 28 days) 10% jaggery in water or to 200 ml of
water alone. Ten female moths were sequentially tested for a response to water, fol
lowed by ten females sequentially tested for response to jaggery This experiment was
conducted on five different days, with water presented first in 2 trials, and jaggery
presented first in 3 trials. This experiment was repeated with males, but on 7 different
days. The treatment sequence (water and jaggery) was also alternated between repli
cates in this experiment. Attraction response data (attraction is upwind oriented
flight and contact with the bait) were analyzed by Student's t-test, after transform
tion to percentages of moths tested within each data set.
Because microbial fermentation may be a determining factor in the attractiveness
of food baits to many lepidopterans (Norris 1936), a second flight tunnel experiment
was conducted to evaluate the effect of the age of thejaggery bait on its attractiveness
to female tobacco budworm. It is expected that colonization and growth of microbes in
baits, and resultant changes in odorants released from baits, occur over time. Attrac
tiveness to bait may then increase with time, as microbes and their metabolic byprod
ucts increase in abundance. This experiment was conducted as two separate series of
bait ages: a short series and a long series. The short series consisted of baits held for
0, 3, 6, 9, and 12 days in a fume hood in the laboratory. Baits were made every 3 days
and bioassays were conducted on 6 different days when baits of all age cohorts were
available simultaneously The long series consisted of baits held for 0, 6, 12, 18, and
24 days. Similarly, these were made every 6 days and bioassays were conducted on 6
different days when baits of all age cohorts were available simultaneously. Every time
a series of bait ages was tested in the flight tunnel, five females were tested per treat
ment (bait age). Thirty females were tested per treatment over the course of the rep
licates. The treatment sequence was altered daily over the 6 days that the test was
replicated. Response data for the long series was subjected to regression analysis for
relationship between bait age and moth response.

September, 1997

Landolt & Mitchell: Tobacco Budworn Food Attractant 405


Male tobacco budworm moths were captured only in traps baited with 10%jaggery
placed in field cages. Mean numbers of released males captured in glass McPhail
traps baited with 10% aged jaggery (4.45 + 1.5 moths per trap per day) were signifi
cantly greater than those captured in traps baited only with water (no moths cap
tured) (t = 3.0, df = 19, p = 0.008). There was a significant linear regression of bait age
versus numbers of male tobacco budworm captured in traps baited withjaggery (r2
0.65, t 2.58, df 18, p 0.03, Y 9.26+ 1.42X) (Fig. 1).
Both sexes of tobacco budworm were attracted to 10% solutions of jaggery pre
sented in the flight tunnel. Twenty-four percent of females flew upwind and contacted
the jaggery bait, compared to no females responding to the control (water only) (t
4.71, df = 4, p = 0.009). Twenty-seven percent of males tested flew upwind and con
tacted the jaggery bait compared to no males responding to the control (t = 3.14, df
6, p 0.022).
In a direct comparison of the attractiveness of jaggery bait of different ages, no fe
male tobacco budworm moths were attracted to bait that was freshly made or was 3
or 6 days old. Nine-day old jaggery was essentially non-attractive as well. Female re
sponse to jaggery increased with bait age from 12 through 24 days old (Fig. 2), with
the highest response (40%) obtained with 24 day old bait. The relationship between
bait age and moth response for the long series was significant by regression analysis
(r2 0.948, t 5.14, df 4, p0.014, Y 6.3+ 1.73X).

0 0
u 6

S5 6 7 8 9 10 11 12 13 14 15 16

Fig. 1. Numbers of male tobacco budworm moths released into a field cage and cap
tured in traps baited with 10% jaggery of different ages. February-March 1996.

Florida Entomologist 80(3)


i- 50

4 40
s 30

u. 20
0 10
U 0
M 0 3 6 9 12 15 18 21 24

Fig. 2. Percentages (+ SEM) of female tobacco budworm moths attracted to contact
a pan containing 200 ml of 10% jaggery in a flight tunnel, for different ages ofjaggery
The short series (solid bars) included 0, 3, 6, 9, and 12-day old baits. The long series
(cross hatched bars) included 0, 6, 12, 18, and 24-day old baits.


These results demonstrate that both sexes of tobacco budworm are attracted to
fermented bait made from 10%jaggery. Both females and males were attracted tojag
gery (exhibited upwind oriented flights from the release dispenser and contacted the
bait) in the flight tunnel. The recapturing of male tobacco budworms in baited traps
in a field cage also indicates an ability to orient to the source of odors from such baits.
This is the first report of orientation responses to food baits by H. virescens. Adult
tobacco budworms feed at materials that are sugar rich, including flowers, extrafloral
plant nectaries, and grass florets (Lingren et al. 1977, Ramaswamy 1990). There are
also reports of corn earworm moths feeding at sweet baits (Ditman and Cory 1933)
and at fungal-infected grass florets (Beerwinkle et al 1993), with the assumption that
they are attracted to such materials. Tobacco budworm attraction to odors emanating
from fermenting sugar solutions is likely a mechanism to locate the sources of such
odors in order to feed.
The significant regressions of bait age versus males captured in traps in a field
cage and bait age versus female response in a flight tunnel support the assumption of
Norris (1936) that microbial activity is a critical factor in moth attraction to sweet
baits. The grass looper, M. latipes, also responds optimally to sweet baits that are aged
(Landolt 1995). However, 3 day old jaggery or 3-day old molasses was most effective
as a trap bait for the grass looper, compared with 16 or 24 day old jaggery for the to
bacco budworm. Perhaps the grass looper moths and tobacco budworm moths respond
to different sets of odorants emanating from baits of different ages.

September, 1997

Landolt & Mitchell: Tobacco Budworn Food Attractant 407

The positive results using the flight tunnel provide a convenient bioassay technique
for pursuing the isolation and identification of odorants from solutions of jaggery that
are attractive to tobacco budworm. The liquid bait and trap used in these experiments
are too limited and inconvenient to use as a monitoring method for female tobacco bud
worms. The trap is heavy and fragile, and trap maintenance is time-consuming. The
trap also holds a limited number of captured moths, and captured specimens may be
difficult to identify if allowed to remain and decompose. For these reasons, it is consid
ered that a formulated blend of synthetic chemicals that are attractive to tobacco bud
worm can be adapted to a cheaper and easier trap design for field use.


Technical assistance was provided by O. Molina and S. Lovvorn. Moths used in bio
assays were provided by F Adams, T Nguyen, C. Greene, and N. Loman. S. Clement,
K. Ward, T W. Phillips, and D. M. Jackson critiqued an early draft of the manuscript.


BEERWINKLE, K. R., T. N. SHAVER, AND J. D. LOPEZ, JR 1993. Field observations of
adult emergence and feeding behavior of Helicoverpa zea (Lepidoptera: Noctu
idae) on dallisgrass ergot honeydew Environ. Entomol. 22: 554-558.
CALKINS, C. O., AND J. C. WEBB. 1983. A cage and support framework for behavioral
tests of fruit flies in the field. Florida Entomol. 66: 512-514.
DITMAN, L. P., AND E. N. CORY. 1933. The response of corn earworm moths to various
sugar solutions. J. Econ. Entomol. 26: 109-115.
EYER, J. R. 1931. A four year study of codling moth baits in New Mexico. J. Econ. En
tomol. 24: 998-1001.
FROST, S. W. 1926. Bait pails as a possible control for the oriental fruit moth. J. Econ.
Entomol. 19: 441-450.
FROST, S. W 1928. Continued studies of baits for oriental fruit moth. J. Econ. Ento
mol. 21: 339-348.
HOLLAND, W. J. 1903. The moth book. A guide to the moths of North America. Dover
Publications Inc., New York, 479 pp.
LANDOLT, P. J. 1995. Attraction of Mocis latipes (Lepidoptera: Noctuidae) to sweet
baits in traps. Florida. Entomol. 78: 523-530.
LANDOLT, P. J., AND R. R. HEATH. 1987. Role of female-produced sex pheromone in be
havioral reproductive isolation between Trichoplusia ni (Hubner) and
Pseudoplusia includes (Walker) (Lepidoptera: Noctuidae: Plusiinae). J. Chem.
Ecol. 13: 1005-1018.
1977. Nocturnal behavior of four lepidopteran pests that attack tobacco and
other crops. Ann. Entomol. Soc. America 70: 161-167.
NEWELL, W. 1936. Progress report on the Key West (Florida) fruit fly eradication
project. J. Econ. Entomol. 29: 116-120.
NORRIS, M. J. 1936. The feeding habits of the adult Lepidoptera Heteroneura. Trans.
Roy Entomol. Soc. London. 85: 61 90.
RAMASWAMY, S. B. 1990. Periodicity of oviposition, feeding, and calling by mated fe
male Heliothis virescens in a field cage. J. Insect Behav. 3: 417-427.
RAMASWAMY, S. B., S. A. RANDLE, AND W. F. MA. 1985. Field evaluation of the sex
pheromone components of Heliothis virescens (Lepidoptera: Noctuidae) in cone
traps. Environ. Entomol. 14: 293-296.
SARGENT, T. D. 1976. Legion of night. The underwing moths. Univ. Massachusetts
Press, Amherst, MA 222 pp.
G. MULLINIX. 1979. Field responses of male Heliothis virescens to pheromonal
stimuli and traps. Bull. Entomol. Soc. America 25: 268-274.

Florida Entomologist 80(3)


Ft. Lauderdale Research and Education Center, University of Florida
Institute of Food and Agricultural Sciences, Ft. Lauderdale, Florida 33314

During April 1996, we were informed by the Department of Agriculture and Con
summer Services of a possible new infestation of the Formosan subterranean termite,
Coptotermes formosanus Shiraki, in a commercial building at the northwest corner of
Highway 1 and State Road 836 in Miami, Florida (Fig. 1). This infestation, which is 2
blocks west of the Port of Miami, is about 10 km south of the currently known distri
bution of C. formosanus in southeastern Florida, and about 1.5 km south of the site of
another introduced subterranean termite, Heterotermes species (Scheffrahn & Su
1995). A large number of alates swarmed in the front office of this building, and nu
merous foraging tubes similar to those of C. formosanus were found on the garage
walls. Workers and soldiers were also collected from a nearby tree. A close examine

Browafd Bvld.

GriflinRd -

Shendan St -- ---
Hollywood Bvld.
SB/ i .. C. havilandi
Hallendale B. Bvid.
A Helerofermes sp.

Fig. 1. The first infestation of C. havilandi was found in a commercial building at
the northwest corner of Highway 1 and State Road 836 in Miami (solid star); about 10
km south of the currently known distribution of C. formosanus in southeastern Flor
ida (shaded area), and about 1.5 km south of the site of another introduced subterra
nean termite, Heterotermes species (solid triangle). The second C. havilandi
infestation was found in a church at the northeast corner of I-95 and State Road 836;
about 1 km west of the first find (solid star).

September, 1997

Scientific Notes

tion of alates revealed that the infestation belonged to another destructive species of
subterranean termite, Coptotermes havilandiHolmgren. This is the first record of this
species in the continental United States. Voucher specimens were deposited at the
University of Florida termite collection in Ft. Lauderdale.
Alates of C. havilandi are readily distinguishable from C. formosanus by the dif
ferential dorsal and ventral coloration. Head and abdominal dorsal tergites of C. havi
land alates are dark brown, and the ventral surfaces of their heads and abdomens
are light yellowish brown. Alates of C. formosanus are entirely light yellowish brown.
The presence of white, halfmoon-shaped "antennal spots" in front of each occelus (Fig.
2) is also characteristic of C. havilandi alates (Ahmad 1965). Alates of C. formosanus
lack such antennal spots. A consistent diagnostic characteristic that distinguishes sol
diers of C. havilandi from C. formosanus is the single pair of setae projecting dorso
laterally from the base of the fontanelle. Soldiers of C. formosanus have two pairs of
such setae (Scheffrahn et al. 1990).
Coptotermes havilandi is a destructive pest of structural wood and agricultural
crops in Thailand, Malaysia, and Indonesia (Ahmad 1965, Gay 1967, Roonwal 1979).
Like C. formosanus, C. havilandi is considered native to the Orient (Araujo 1970,
Grasses 1984) and has been widely exported. First introduced to Brazil in 1923, C.
havilandi is currently considered the major structural pest in the city of Sao Paulo
(Lelis 1995). This termite species was first found in the West Indies on Barbados (Ad
amson 1938). Current distribution of C. havilandi in the West Indies also includes An
tigua, Cayman Islands, Cuba, Isla de la Juventud, Jamaica, Montserrat, and Turks
and Caicos Islands (Scheffrahn et al. 1994). Other regions of known C. havilandi dis
tribution are Madagascar and Mauritius (Edwards & Mill 1986).
Coptotermes havilandi is mostly found in the tropics whereas C. formosanus is dis
tribute primarily in subtropical and temperate regions (Su & Tamashiro 1987). Both
C. havilandi and C. formosanus cause devastating damage to structures wherever
they occur. Records showed that C. havilandi in southeast Asia attacks dead and dy
ing trees of various species, construction timber, furniture, structural wood, plastics,
and synthetic fibers (Roonwal 1979). Like C. formosanus, C. havilandi in Sao Paulo,
Brazil also construct aerial nests in high rise buildings (Lelis 1995). No data are avail
able for the overall economic impact by C. havilandi in the Caribbean, but it is a se
rious pest of structures in Little Cayman, and Providenciales and Turk (Su &
Scheffrahn, unpublished data).
Numerous timbers in the one-story concrete building infested by C. havilandi in
Miami were so severely damaged that they had to be replaced. The infestation was no
ticed by the occupants 5 years ago, but commercial pest control firms contracted for
treatment have mistaken the infestation for native subterranean termites, Reticuli
terms species. Soil termiticides have been applied annually for the last 5 years. The
most recent soil termiticide treatment was done in April 1996. Despite the annual ap
plication of soil termiticide, infestation by this C. havilandi colony continues.
According to the occupant, alate swarming was observed 3 years ago. Because it
generally takes 3-5 years for a colony to be mature enough for alate production, and
because of the treatment history, C. havilandi was probably introduced to Miami
about 10 years ago. Leisure-crafts infested by C. havilandi have been found in Carib
bean and Florida waters (Scheffrahn et al. 1990). The close proximity of this infesta
tion to the Port of Miami suggests a maritime introduction. Our suspicion that C.
havilandi is not limited to this site was confirmed when, in August, another infesta
tion was found in a church at the northeast corner of I-95 and State Road 836; about
1 km west of the first find (Fig. 1). Damage potential and behavior of C. havilandi is
similar to that of C. formosanus, but these two pest species are geographically sepa
rated because of their different climatic adaptations. This is an unprecedented inci

Florida Entomologist 80(3)

sp o C o m p o u n d


C. havilandi


C. formosanus
Fig. 2. Alates of C. havilandi is distinguishable from C. formosanus by the presence
of white, halfmoon-shaped "antennal spots" in front of each occelus.

dent in which both C. havilandi and C. formosanus are found within such a short
distance, and their interaction will be closely monitored.
We are grateful to Frank Valdes and Mike Petrozzino (Department of Agriculture
and Consumer Services) who first collected C. havilandi in Miami, and Jan Krcek and
Ted Center for their critical reviews. Florida Agricultural Experimental Station Jour
nal Series No. R-05322.


The first introduction record of the subterranean termite, Coptotermes havilandi
Holmgren, into the continental United States is reported. Thus far, two infestations

September, 1997

Scientific Notes

have been recorded in Miami, Florida. The infestation history suggests that C. havi
landiwas probably introduced to Miami about 10 years ago through maritime trans
portation. Coptotermes havilandi is found primarily in tropical regions such as
southeast Asia, Brazil, and the Caribbean, and its damage potential is similar to that
of the Formosan subterranean termite, C. formosanus Shiraki.


ADAMSON, A. M. 1938. Notes on termites destructive to buildings in Lesser Antilles.
Trop. Agric., Trinidad 15: 220-224.
AHMAD, M. 1965. Termites (Isoptera) of Thailand. Bull. American Mus. Nat. Hist. 131:
ARAUJO, R. L. 1977. Catalogo dos Isoptera do Novo Mundo. Acad. Brasileira de Cien-
cias, Rio de Janeiro, RJ. 92 pp.
EDWARDS, R., AND MILL, A. E. 1986. Termites in buildings. Their biology and control.
Rentokil ltd., W Sussex.
GAY, F. J. 1967. A world review of introduced species of termites. CSIRO Melbourne,
Australia, Bull. 286: 188.
GRASSE, P.-P. 1984. Termitologia Tom II: Fondations des society Construction, Chap.
V pp. 162-208, Masson, Paris.
LELIS, A. T. 1995. A nest of Coptotermes havilandi (Isoptera: Rhinotermitidae) off
ground level, found in the 20th story of a building in the city of Sao Paulo, Bra
zil. Sociobiology, 26: 241-245.
ROONWALD, M. L. 1979. Termite life and termite control in tropical South Asia. Scien
tific Publishers, Jodhpur.
SCHEFFRAHN, R. H., AND N.-Y. Su. 1990. Native, introduced, and structure-infesting
termites of the Turks and Caicos Islands, B.W.I. (Isoptera: Kalotermitidae, Rhi
notermitidae, Termitidae). Florida Entomol. 73: 622-627.
SCHEFFRAHN, R. H., AND N.-Y. Su. 1995. A new subterranean termite introduced to
Florida: Heterotermes Frogatt (Isoptera: Rhinotermitidae: Heterotermitinae)
established in Miami. Florida Entomol. 78: 623-627.
1994. Termites (Isoptera: Kalotermitidae, Rhinotermitidae, Termitidae) of the
West Indies. Sociobiology, 24: 213-238.
Su, N.-Y., AND M. TAMASHIRO. 1987. An overview of the Formosan subterranean ter
mite (Isoptera: Rhinotermitidae) in the world, pp. 3-15 in M. Tamashiro and N.
Y. Su [eds.], Biology and control of the Formosan subterranean termite. College
of Trop. Agric. Human Res., Univ. of Hawaii, Honolulu, Hawaii.


MORON, M. A., B. C. RATCLIFFE, AND C. DELOYA. 1997. Atlas de los escarabajos de
Mexico. Coleoptera: Lamellicornia. Vol. I Familia Melolonthidae. CONABIO and So
ciedad Mexicana de Entomologia; Mexico, xvi + 280 p. ISBN 9680-7801 00-X. Paper
back. 21 x 27 cm. Available from: Sociedad Mexicana de Entomologia, km 2.5 antigua
carretera a Coatepec, Apartado Postal 63, 91000 Xalapa, Veracruz, Mexico, for US
$50.00 (check or international money order) including packing and certified airmail.

The English word beetle means a member of the order Coleoptera, and the word
scarab means a member of the family Scarabaeidae. The Spanish word escarabajo
does double duty, meaning in its broadest sense beetle, but in its most restricted
sense scarab. The authors of this book use the word escarabajo to mean a member
of the evocatively-named superfamily Lamellicornia (having laminate antennae). The
name Lamellicornia in most modern works has been replaced by Scarabaeoidea, fol
lowing recommendation 29G of the International Code of Zoological Nomenclature.
This is one of two books, the other still in preparation, which deal with the scara
baeoid fauna of Mexico. This one deals with the family Melonthidae, and the forth
coming one will deal with the families Trogidae, Scarabaeidae, Lucanidae, and
Passalidae. Most readers in America north of Mexico will not be familiar with the fam-
ily name Melolonthidae, because the retrograde classification used in the USA recog
nizes only 3 families of Scarabaeoidea (Lucanidae, Passalidae and Scarabaeidae) and
thus includes Melolonthidae and Trogidae as subfamilies of Scarabaeidae. In the class
sification used in this book, the family Melolonthidae includes subfamilies Rutelinae,
Dynastinae, Trichiinae, Valginae, Cetoniinae, and Melolonthinae.
In English, some of the vernacular names for members of this family
Melolonthidae are chafer, May beetle, or June beetle. The name chafer conjures up an
image of a chunky, thumb-nail sized beetle with bright pattern or metallic coloration
seen feeding on flowers of the family Umbelliferae, whereas the name May beetle or
June beetle evokes a picture of a cylindrical brown beetle about the size of the last seg
ment of a little finger, and sometimes attracted to electric lights in surprising num
bers. Those names do not do justice to the magnificent Mexican beetles described and
illustrated in this book. If you have not seen Dynastes hercules (70-130 mm) or Megas
oma elephas (51-120 mm) alive you have missed something.
It is unfortunate that the name dung beetle is promoted in the USA for members
of the Scarabaeidae. These Melolonthidae are not dung beetles: their larvae feed on
roots of plants or in decaying wood, and their adults feed on flowers or foliage of
This book has a diagnosis of each of the 110 genera of Melolonthidae known from
Mexico. It includes a brief description and notes on habitat, distribution, and in some
instances behavior, for 253 of the 1,040 known species. The distribution (by Mexican
state) of the remaining 787 species is given in tables. There are 61 black and white il
lustrations of adults: most of them are drawings and many of them are of superb qual
ity. Remarkably, there are 32 plates containing 253 color photographs, some showing
living larvae or pupae, some showing adults in nature, most showing pinned speci
mens. A preface by Gonzalo Halffter, and prologue and introduction by Miguel Moron
set the background (and show that the states with by far the highest recorded diver
sity of species are Chiapas, Oaxaca, and Veracruz). A bibliography, and systematic
and thematic indices complete the book which is well-printed on glossy paper.

Book Reviews 413

I look forward to seeing Volume II of this ground-breaking work. Volume I should
serve as an inspiration and challenge to entomologists in Mexico (and elsewhere) to
match its quality in publications on other families of insects in their country. Ento
mologists north of the border should note the event of its publication and buy a copy
while supplies last. The next generation of the work must include keys to adults and
genitalic illustrations to allow identification to the species level, but this will require
a great increase in number of pages and will drive the price much higher.
J. H. Frank
Entomology & Nematology Dept.
University of Florida
Gainesville, FL 32611 0630

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