(An International Journal for the Americas)
Volume 76, No. 4 December, 1993
TABLE OF CONTENTS
DRANEY, M. L.-The Subelytral Cavity of Desert Tenebrionids .................... 539
GHIDIU, G. M., AND J. T. ANDALORO-The Relationship Between Fall Ar-
myworm (Lepidoptera: Noctuidae) Instar and Susceptibility to Insecticides
Applied to Sweet Corn ..................................................................... 549
LANDOLT, P. J., AND B. LENCZEWSKI-Lack of Evidence for the Toxic Nectar
Hypothesis: A Plant Alkaloid Did Not Deter Nectar Feeding by Lepidoptera 556
WILLIAMS, M. L.-Toumeyella lignumvitae, A New Species of Scale Insect from
the Florida Keys (Homoptera: Coccidae) ........................................... 566
LANDOLT, P. J.-Suitability of Six Species of Commelinaceae as Larval Hosts of
Mouralia tinctoides (Lepidoptera: Noctuidae, Plusiinae) ...................... 572
PORTILLO, H. E., C. M. FELLAND, H. N. PITRE, AND R. P. PORTER-Pyre-
throid Resistance Levels in Soybean Looper (Lepidoptera: Noctuidae) in
M ississippi .................................................................................... 577
NATION, J. L., J. L. FOLTZ, W. N. DIXON, AND H. J. MCAUSLANE-Evalua-
tion of Loss of (+)-Disparlure from Gypsy Moth (Lepidoptera: Lyman-
triidae) Pheromone Dispenser Tapes Under Field Conditions in Florida ... 584
DEYRUP, M., AND T. H. ATKINSON-Survey of Evaniid Wasps (Hymenoptera:
Evaniidae) and Their Cockroach Hosts (Blattodea) in a Natural Florida
H habitat ......................................................................................... 589
LOUNIBOS, L. P., AND C. E. MACHADO-ALLISON-Field Test of Mosquito
Ovipositional Cues from Venezuelan Phytotelmata ............................. 593
RODRIGUEZ-DEL-BOSQUE, L. A. AND J. LOERA-GALLARDO-Influence of Corn
Phenology and Planting Date on Damage by the Black Cutworm (Lepidopt-
era: N octuidae) .............................................................................. 599
SCHEFFRAHN, R. H., AND J. KRECEK-Parvitermes subtilis, A New Subter-
ranean Termite (Isoptera: Termitidae) from Cuba and the Dominican
R public ........................................................................................ 603
ATKINSON, T. H.-A New Species of Pityophthorus Eichhoff (Coleoptera:
Scolytidae) from Southern Florida with a Key to the Florida Species ...... 608
OCHOA, M. G., M. C. LAVIN, F. C. AYALA, AND A. J. PEREz-Arthropods
Associated with Bromelia hemisphaerica (Bromeliales: Bromeliaceae) in
M orelos, M exico ............................................................................. 616
LANCIANI, C. A., AND R. EDWARDS-Effect of Photoperiod on Wing Area in
Anopheles quadrimaculatus (Diptera: Culicidae) ................................. 622
Continued on Back Cover
Published by The Florida Entomological Society
FLORIDA ENTOMOLOGICAL SOCIETY
OFFICERS FOR 1993-94
P resident t ....................................... ............................... J. E P e a
President-Elect ........................................ ................. E. M. Thorns
Vice-President ................................................... J. A. Coffelt
Secretary ........................................................... D. G. Hall
Treasurer ................................. .................... A. C. Knapp
Other Members of the Executive Committee
D. F. Williams N. D. Epsky L. A. Wood C. L. Bloomcamp
C. S. Lofgren M. Lindsey R. G. Goodson
C. S. Lofgren, USDA/ARS (Retired), Gainesville, FL ............................... Editor
Agricultural, Extension, & Regulatory Entomology
James R. Brown-Disease Vector Ecology & Control Ctr., NAS, Jacksonville, FL
Richard K. Jansson-Merck & Co., Inc., Three Bridges, NJ
Michael F. Hennessey-Subtropical Horticulture Res. Lab., USDA/ARS, Miami, FL
Geoffrey Zehnder-Alabama Coop. Ext. Serv., Auburn University, Auburn, AL
Stephen B. Bambara-North Carolina State University, Raleigh, NC
Biological Control & Pathology
Ronald M. Weseloh-Connecticut Agricultural Experiment Sta., New Haven, CT
John M. Brower-Stored Products Insect Res. Lab., USDA/ARS, Savannah, GA
J. Howard Frank-Dept. of Entomology, UF/IFAS, Gainesville
Chemical Ecology, Physiology, Biochemistry
R. K. Vander Meer-Med. & Vet. Entomol. Res. Lab., USDA/ARS, Gainesville, FL
Ecology & Behavior
Sanford D. Porter-Med. & Vet. Entomol. Res. Lab., USDA/ARS, Gainesville, FL
Gregory S. Wheeler-Ft. Lauderdale Res. & Ed. Ctr., UF/IFAS, Ft. Lauderdale, FL
Genetics & Molecular Biology
Sudhir K. Narang-Bioscience Research Laboratory, USDA/ARS, Fargo, ND
Medical & Veterinary Entomology
Arshad Ali-Central Florida Res. & Ed. Ctr., UF/IFAS, Sanford, FL
F. W. Howard-Ft. Lauderdale Res. & Ed. Ctr., UF/IFAS, Ft. Lauderdale, FL
Systematics, Morphology, and Evolution
Michael D. Hubbard-Dept. Entomol., Florida A&M University, Tallahassee, FL
Gary J. Steck-Florida State Collection of Arthropods, FDACS/DPI, Gainesville, FL
W. W. Wirth-Florida State Collection of Arthropods, FDACS/DPI, Gainesville, FL
Business M manager ....................................................................... A. C. Knapp
FLORIDA ENTOMOLOGIST is issued quarterly-March, June, September, and De-
cember. Subscription price to non-members is $30 per year in advance, $7.50 per copy;
institutional rate is $30 per year. Membership in the Florida Entomological Society,
including subscription to Florida Entomologist, is $25 per year for regular membership
and $10 per year for students.
Inquiries regarding membership and subscriptions should be addressed to the Busi-
ness Manager, P. O. Box 7326, Winter Haven, FL 33883-7326.
Florida Entomologist is entered as second class matter at the Post Office in DeLeon
Springs and in Winter Haven, FL.
Manuscripts from all areas of the discipline of entomology are accepted for consider-
ation. At least one author must be a member of the Florida Entomological Society.
Please consult "Instructions to Authors" on the inside back cover.
This issue mailed December 31, 1993
Draney: Subelytral Cavity 539
THE SUBELYTRAL CAVITY OF DESERT TENEBRIONIDS
MICHAEL L. DRANEY
Department of Entomology, University of Georgia
Athens, GA 30602-2603
Tenebrionid beetles are a conspicuous component of the world's hot deserts. Different
species exhibit considerable variability in physiology, morphology, behavior, and life
history; most aspects of their biology are influenced by the interrelated problems of
excess heat and insufficient moisture which all desert organisms face. This paper is a
review of research and thought concerning an adaptation characteristic of desert teneb-
rionids, an air space between fused elytra and the dorsum of the abdomen called the
The subelytral cavity may be seen as a hermetic seal which reduces transpiration
regardless of its size, and as a protected space within the beetle which, whether airtight
or not, allows the abdomen to expand to store food, water, or eggs. I hypothesize that
the cavity itself is probably an architectural constraint resulting from the need for
abdominal expansion within the heavily sclerotized, fused elytra that evolved primarily
as a water conservation adaptation.
Los escarabajos son un component conspicuo de los desiertos calientes del mundo.
Las diferentes species muestran variabilidad considerable en la fisiologia, la morfologia,
el comportamiento y el ciclo biol6gico. La mayoria de los aspects de su biologia estan
bajo infuencia de los problems vinculados de calor excesivo y de humedad no suficiente,
como es el caso con todos organismos del desierto. Este trabajo es una revista de las
investigaciones y los pensamientos sobre una adaptaci6n caracteristica de tenebrionidos
del desierto, eso es, un espacio de aire entire los elitros fusionados y el dorso del abdomen,
lo cual se llama la cavidad subelitral.
La cavidad subelitral se puede considerar como un sello herm6tico lo cual reduce la
transpiraci6n a pesar del tamaio, y como un espacio protectado adentro del escarabajo
lo cual, aunque hermetico o no, permit el abdomen estirarse para guardar comida,
agua, o huevos. Yo present el hip6tesis que la cavidad misma probablemente es un
constrefiimiento arquitectural resultando de la necesidad de expansion abdomenal dendro
de los elitros fusionados y fuertamente esclerotizados los cuales evolvieron primariamente
como una adapci6n para la conservation de agua.
Tenebrionid or darkling beetles form a significant portion of the fauna of hot deserts
throughout the world. Under some conditions, the biomass of tenebrionids living in a
desert area may exceed the combined biomass of the entire vertebrate community
(Thomas 1979). In such numbers, these beetles must exert a considerable ecological
influence in the desert ecosystem. Coleoptera seem to be one of the insect orders best
adapted to desert life (Cloudsley-Thompson & Chadwick 1964), and the Tenebrionidae
are probably the best-represented beetle family in deserts (Crawford 1981). Of the more
than 15,000 described species of tenebrionids, more are xerophilous than mesophilous
or hydrophilous. The family is at its most diverse in the Namib Desert in southwestern
Africa, where several hundred species are found representing over 90 genera (Koch 1962).
Desert tenebrionids are not a phylogenetically distinct lineage within the family
Tenebrionidae. There are xerophilous forms from three tenebrionid subfamilies, and
540 Florida Entomologist 76(4) December, 1993
desert beetles belonging to dozens of different tribes occur in different parts of the
world. However, most desert tenebrionids are distinguished from most other tene-
brionids in that they are flightless, with fused elytra enclosing an air space called the
subelytral cavity. Although different species of tenebrionids display a diversity of
physiological, morphological, and behavioral adaptations to deal with various problems
of desert life, presence of a subelytral cavity is widespread.
The subelytral cavity, then, is a characteristic desert tenebrionid adaptation, a feature
common (with exceptions, especially among smaller species) to an otherwise diverse
group of coleopterans. For at least seventy years, there has been much speculation (and
some research) concerning the ecological function of this structure, i.e., what is its
importance to the beetle? This review provides a brief overview of the biology of desert
tenebrionids, and summarizes some interesting aspects of work done on a particularly
fascinating desert tenebrionid adaptation, the subelytral cavity.
I must mention at this point that generalizations are inevitable when reviewing the
biology of such a large and diverse group of organisms as desert tenebrionids. The
information has been culled from observations on a large number of species, and no one
species is "typical" of the group. Generic and specific names will not be given except
where necessary, since these names will not be meaningful to the non-specialist. In-
terested readers will find the names in the literature referenced.
BIOLOGY OF DESERT TENEBRIONIDS
Although they display considerable behavioral and morphological variation, desert
tenebrionids may generally be described as rather slow moving, black, flightless beetles.
Adults and larvae are considered detritivorous (Wallwork 1982), although adults have
also been observed feeding upon carrion (McKinnerney 1978) and dung (Buxton 1924),
and some larvae are root-feeders (Rafes 1960). Tenebrionids may play an important role
in detritus cycling in ecosystems where few microbial decomposers are active outside
the tenebrionid gut. This is most important on the vegetationless dunes of the Namib
Desert, where allochthonous detritus serves as the trophic base for the dune community
(Seely & Louw 1980).
All known desert tenebrionids are subterranean as larvae and may spend several
months developing in the soil. The number of instars is variable, often exceeding ten.
One, two, and three year life cycles have been identified. Adults or larvae overwinter
in the annual cycles, and both stages of a single species overwinter in the longer cycles
(Allsopp 1980). Adult longevity in different species varies from a few weeks up to five
years (Crawford 1981, 1990). Reproduction is iteroparous. Fecundity varies widely,
with females usually depositing a few eggs at a time (Wallwork 1982).
Central to the tenebrionid life strategy is the avoidance of extreme environmental
conditions by retreating to refuges during periods of inactivity. The egg, larval, and
pupal stages are always completed underground, and some dune-dwelling forms are
even "ultrapsammophilous," developing entirely in loose, shifting sand. Adults alternate
between active foraging on the ground surface (exceptionally, in vegetation (Crawford
1990)) and inactivity within a refuge. Type of refuge utilized varies with the edaphic
and vegetational characteristics of the beetle's habitat. Dune species may simply burrow
into the sand, while species on ordinary soil substrates may occupy abandoned rodent
burrows, ant nests, etc, or dwell beneath stones, wood, or debris (Crawford 1981).
Diel activity patterns are species, genus, and even tribe specific (Koch 1961). Tene-
brionids may be diurnal, nocturnal, or crepuscular. Also, the adults of most species are
active only during certain seasons of the year. Many long-lived species spend the re-
mainder of the year in a state of dormancy within their refuges. In the southwestern
United States, tenebrionid activity reaches a peak during the rainy season (Tanner &
Draney: Subelytral Cavity
Packham 1965). Different species, though, reach their peak of activity in different sea-
sons. This temporal separation of the detritivore niche, on both daily and annual scales,
may allow more species of tenebrionids to co-exist in a single habitat, although in-
terspecific competition has not been demonstrated in these beetles (Wise 1981). In any
case, limiting activity to certain parts of the day or year probably helps the animals to
minimize water loss and avoid extreme temperatures, and enables them to deal with
the variability and unpredictability of climatic factors inherent in desert ecosystems
(Ahearn 1971, Kramm & Kramm 1972, Wallwork 1982).
An important feature which distinguishes the two major desert tenebrionid sub-
families is the possession of abdominal defensive glands. These are present in the sub-
family Tenebrioninae and always lacking in the Tentyriinae (Doyen & Tschinkel 1982).
When disturbed, these glands emit a dark reddish secretion with a disagreeable odor.
The secretions differ chemically from species to species, but practically all contain vari-
ous quinones, and 1-alkenes and other hydrocarbons are common components as well
(Tschinkel 1975). Defensive secretions appear to make tenebrionids relatively unpalat-
able to many vertebrates (Eisner & Meinwald 1966) and invertebrates (Slobodchikoff
1979), although beetles possessing them still serve as food for a large set of vertebrate
and invertebrate predators (Allsopp 1980).
A few desert tenebrionids are so aberrant in morphology and habits that much of
this discussion probably does not apply to them. An example is the Old World genus
Cossyphus. These are exceedingly flat beetles which apparently mimic winged seeds
found in the vegetable debris they live in (Cloudsley-Thompson 1977).
More detailed information about the biology of desert tenebrionids can be found in
the comprehensive works of Crawford (1981, 1990) and Wallwork (1982).
THE DESERT ENVIRONMENT: THERMAL AND WATER BALANCE
Desert climates present organisms with the basic problems of avoidance of excessive
heat and dehydration (Hadley 1972). Extreme diurnal temperatures and intense solar
radiation may challenge an organism's thermal tolerance. Rainfall is low in absolute
amount, and may be seasonally distributed. High year-to-year and spatial variation of
rainfall is also typical in deserts, leaving some local areas much drier than others (W.
G. Whitford pers. comm.). Also, the heat, low relative humidity, and drying winds
combine to accelerate evaporation of what little water is available.
In extreme environments, such as deserts, organisms may be more limited by the
abiotic environment than by interactions with other organisms. Wise (1981) was unable
to demonstrate competition among several species of tenebrionids for a detritus re-
source. It seems likely that tenebrionids are limited less by the amount of available
detritus than by the harsh abiotic conditions of the desert environment. The high diurnal
temperatures limit the amount of time that the beetles can forage actively on the ground
surface, and lack of water may frequently reduce the beetles' metabolic and reproduc-
tive rates. In severe years, drought may also contribute directly to mortality. Probably
all aspects of tenebrionid life are directly or indirectly affected by the interrelated
problems of maintaining a favorable temperature and getting and keeping sufficient
Like other poikilotherms, tenebrionids can be active only when their body is warm
enough for efficient metabolism to proceed, yet below some lethal temperature. Desert
organisms may be selected for increased tolerance of high temperatures, which may
result in an ability to withstand high temperatures for relatively long periods without
perishing (higher "lethal temperature"), and/or a preference for relatively high temper-
Florida Entomologist 76(4)
Lethal temperatures of various species of tenebrionids have been measured in sev-
eral studies (e.g., Cloudsley-Thompson 1964, Edney 1971a, Zachariassen 1977, Seely et
al. 1988). Maximum lethal temperatures of tenebrionids in general vary between about
430 and 530 C. (Seely et al 1988). Beetles from mesic habitats fall towards the low end
of this range (about 43-450 C.). Cloudsley-Thompson (1964) measured similar upper
lethal temperatures for diurnal desert beetles, but other investigators documented
higher tolerances for desert beetles, especially for summer-active diurnal species
(Edney 1971a, Zachariassen 1977).
Measurements of temperatures that beetles experience in the field, or of tempera-
tures preferred by beetles in the laboratory, may be more ecologically meaningful than
upper lethal temperatures. Studies of temperature preference and field temperature
have been reviewed by Seely et al. (1988), who found trends similar to those seen with
lethal temperatures. Beetles from mesic and montane areas (Zachariassen 1977, Doyen
& Slobodchikoff 1984) often prefer temperatures below 200 C. Temperature prefer-
ences of desert species vary from about 200 C. for some nocturnal species up to about
430 C. in some Namib species which utilize a strategy called maxithermy: Maintenance
of body temperatures near the upper lethal temperature limit for as long as possible
(Hamilton 1975, Henwood 1975, Seely et al. 1988). Although desert beetles often have
higher temperature preferences than mesic beetles, there is much interspecific variation
in preference within desert species (El Rayah 1970, Henwood 1975, Slobodchikoff 1983).
Much of this variation may be due to specific adaptations which different beetles use to
optimize their thermal regime in an often inhospitable thermal environment.
The long legs of surface-active tenebrionids are a morphological feature that aids in
thermoregulation. Beetles modify their behavior by "squatting" or "stilting" to take
advantage of, or to avoid, the soil substrate which is usually warmer than the air
temperature (Henwood 1975). Minimal hemolymph circulation to the tarsi limits heat
conduction to the body.
It appears that the most important thermal adaptations in tenebrionids are be-
havioral. Many species avoid high diurnal temperatures by limiting periods of activity
to cooler seasons (Ahearn 1971) and/or cooler times of the day (Cloudsley-Thompson
1964). Day-active beetles may retreat to refuges during the hottest part of the day
(Hadley 1970, Edney 1971b), or may take temporary refuge in the shade of vegetation
(Edney 1971a) or dunes (Koch 1961), or even atop quartz pebbles (Hamilton 1975).
It is uncertain what accounts for increased thermal tolerance in desert versus mesic
tenebrionids. Evaporative cooling is insignificant in desert tenebrionids. They are so
effective at conserving water that what little is transpired has a negligible cooling effect
(Edney 1971b). Higher thermal tolerance may be due in part to the presence of more
stable enzyme systems which denature at higher temperatures (Hadley 1972).
In general, water is the most important limiting factor of desert ecosystems, because
free water is not commonly available. An organism must be able to obtain sufficient
water where and when it does exist, and it also must be able to minimize water loss so
that its water requirement does not exceed its supply.
Desert tenebrionids' water requirements are not great. Both larvae and adults feed
on detritus which is often low in moisture content. Oxidation water may satisfy most
of their water needs (Hadley 1972). Cloudsley-Thompson & Chadwick (1964) state that
"these insects are able to live on dry food and exist without water." The authors imply
that the beetles can complete their entire life cycle without access to free water, a
notion which to my knowledge has been neither demonstrated nor refuted.
Regardless of their ability to do without, it is clear that tenebrionids will drink water
when it is available, often consuming large quantities after a rainfall (Slobodchikoff &
Wisman 1981). Between rain events, dew is an important source of water for species
in some deserts (Hamilton & Seely 1976, Broza 1979).
Draney: Subelytral Cavity 543
Although desert tenebrionids are efficient at obtaining water and may have some
resistance to desiccation, their ability to avoid water loss may be the most important
factor involved in maintaining a favorable water balance. They exhibit some of the
lowest rates of water loss of all arthropods (Edney 1971b). Water is lost to insects by
two routes: Excretion, which includes feces, defensive secretions, and eggs; and trans-
piration via the tracheal system or directly through the cuticle.
Feces are an insignificant source of water loss in desert tenebrionids, since efficient
resorption of water by the hindgut produces fecal pellets low in moisture (Ahearn 1970).
Release of repugnant abdominal (and occasionally oral) secretions results in a relatively
large loss of water (Ahearn 1970). Production of eggs certainly involves a loss of water
for female beetles, but the implications (in terms of survivorship and life-history tactics)
Water loss via transpiration is limited in desert beetles both by passive morpholog-
ical features and active physiological control. The dominant morphological features are
the thick fused sclerites which limit cuticular transpiration, and the fused elytra making
up the subelytral cavity, which will be discussed in greater detail below. Active
physiological control of cuticular and/or spiracular transpiration is evidenced by the fact
that freshly killed beetles lose water much more quickly than identical living specimens
In addition to thick, fused sclerites, the low rates of cuticular transpiration of desert
tenebrionids result from epicuticular lipids which are water impermeable at higher
temperatures than the lipids of mesic species (Ahearn 1970).
Transpiratory water loss in insects occurs through the spiracles. Tenebrionids pos-
sess two large spiracles laterally on the thorax (in the membrane between the prothorax
and the mesothorax) and several metathoracic and abdominal spiracles which open into
the subelytral cavity. As discussed below, the subelytral cavity plays a large role in
limiting spiracular water loss.
The rate of transpiratory water loss, as well as the relative importance of cuticular
versus spiracular transpiration, is largely determined by temperature. Thus, heat and
water loss are interrelated problems which desert organisms must deal with in tandem.
Transpiration increases as temperatures increase (Ahearn & Hadley 1969, Ahearn 1970,
Zachariassen et al. 1987). This may be partly due to epicuticular breakdown above a
species-specific transition temperature. There is also a general increase in spiracular
transpiration due to increased metabolic rate (and hence, oxygen demand) as tempera-
tures increase. Zachariassen et al. (1987) found that the metabolic rate of carabids and
tenebrionids is "the single determinant" of transpiratory water loss, between species
and at different temperatures.
Ahearn (1970) reports that at relatively low temperatures some desert tenebrionids
exhibit very low loss rates, and these losses are dominated by cuticular transpiration.
As temperatures approach the beetle's transition temperature (about 40 C.), increased
metabolic rates cause a breakdown of spiracular control, and spiracular transpiration
becomes the dominant component of water loss.
THE SUBELYTRAL CAVITY
Koch (1962) states that 98% of tenebrionids in the Namib Desert are flightless, the
few winged species being restricted to pockets of mesic habitat, such as river-beds or
pans. Buxton (1923) hypothesized that winglessness is solely an adaptation to excessive
windiness in deserts, pointing out that wingless beetles are also found on mountain tops
and windy islands. Schmoller (1970) gave credence to this hypothesis fairly recently in
his review of desert arthropods. However, evidence will be reviewed below which
Florida Entomologist 76(4)
indicates that winglessness is an indirect result of the formation of a hermetically sealed
air space beneath the elytra. Consequently, the elytra must be fused together, preclud-
ing the use of wings. The vestigial wings beneath the sealed elytra display varying
degrees of atrophy. In different genera they may be reduced to small, slender tubes,
tiny scales, or they may be completely absent (Fiori 1977).
Fiori (1977) describes the morphology of the tenebrionid subelytral cavity in some
detail. The elytra are usually fused permanently at the elytral suture with a "dovetail"
or "double-dovetail" joint. The elytra are fused with the mesonotum towards the front
and with the meso- and metaepimera and metaepisternum at the sides. The elytra may
simply abut these parts, or they may lie in furrows or grooves, or there may be a
permanent dovetail joint. The joints may be permanent at all points, or the permanent
joints may alternate with temporary (non-fused) connections. Fringes of hairs on the
elytra or urosterna may assist in making an effective seal in places where the joint is
not fused. The overall effect is a hermetically sealed air space underneath the elytra.
A small area at the posterior end can be voluntarily opened by the beetle in order to
ventilate the abdominal spiracles, which open into the subelytral cavity. The floor of
the subelytral cavity consists of the meso- and metanotum and the first seven abdominal
uroterga. The uroterga are reduced to soft membranes in beetles with subelytral
cavities. A small proportion of the cavity may contain any wing remnants the species
may possess; the remainder of the cavity is simply an air space.
The subelytral cavity is not unique to the Tenebrionidae, though it probably serves
different functions in different groups. Dytiscids (Coleoptera) possess a clearly defined
subelytral cavity, as do some cave-dwelling beetles belonging to the families Carabidae
and Anisotomidae (Fiori 1977). Desert-dwelling carabids also generally possess subely-
tral cavities (Zachariassen et al. 1987), which probably serve the same general function
as the tenebrionid cavities.
The subelytral cavity is so widespread among desert tenebrionids and is such a
prominent anatomical feature that it probably performs an important function or func-
tions for the beetle. Biologists have suggested at various times that the cavity helps
the beetle deal with many problems it encounters, from handling excessive heat or wind
to obtaining or conserving food and water. It may even enhance reproductive success.
Testing any given hypothesis has proven difficult for several reasons. First, the cavity
may serve several functions, yet it may not serve any unique function. For example,
the beetles obviously have many different adaptations for managing the heat and water
problems of their desert environment. Second, cavity size and shape varies continuously
from beetles with massive subelytral cavities to winged forms with no cavity at all, and
there is no clear-cut line between xerophilous and mesophilous species. Many species
occupy intermediate habitats. Lastly, the subelytral cavity can be visualized as a volume
of air surrounded by a beetle. The cavity and the structures that form it are such an
integral part of tenebrionid anatomy that it may be difficult to imagine and impossible
to test what the beetle would be like without it.
An early and still persistent hypothesis about the function of the subelytral cavity
is that it protects the beetle from excess heat. The basic idea is that the elytra absorb
radiation by acting as a cover over the beetle. The subelytral cavity acts as an insulating
dead-air space, retarding inward conduction of heat to the beetle. Hadley (1970) found
that in direct sunlight, the temperature of the air in the subelytral cavity was 2-8 C.
warmer than body temperature or outside air temperature. It should be noted that
beetles were unable to maintain an equilibrium temperature, and would eventually die
if kept in direct sunlight. However, the buffering capacity of the cavity may delay heat
build-up, allowing the beetles to make short foraging trips in the sun.
Hadley (1970) also suggested the more subtle hypothesis that the subelytral cavity
increases convective cooling. He cited Bolwig's (1957) laboratory study of thermal toler-
Draney: Subelytral Cavity
ance in tenebrionids. Bolwig found that at about 400 C., one beetle species would at-
tempt to cool itself by exposing its genitalia and ventilating its subelytral cavity. The
latter behavior resulted in fluctuations in subelytral cavity temperature. However, it
must be noted that this behavior may be an artifact of the laboratory conditions, which
included the use of an electric heater as a heat source.
Other experimental evidence provides little support for the hypothesis of a thermal
function for the cavity. Cloudsley-Thompson (1964) found no difference in temperature
between the subelytral cavity and the outside air. Also, the lethal temperature of
another species did not change after a piece of elytra had been removed, although
elytral removal causes increased transpiration, which may cool the animal. Fiori (1977)
points out that several studies have found lower cavity temperature in white-elytra
than in black-elytra beetles (Bolwig 1957, Edney 1971a). This suggests that any thermal
effect the subelytral cavity may exert is too small to mask the effect of elytral color.
The early desert ecologist P. A. Buxton (1923) formed his "winglessness as an adap-
tation to excessive wind" hypothesis after rejecting the hypothesis of the subelytral
cavity as a heat insulator. He may have been the first to note that not only diurnal, but
also many nocturnal tenebrionids have large subelytral cavities (Buxton 1923, 1924).
Koch (1961) states that increases and reductions in subelytral cavity volume are to be
found in both nocturnal and diurnal forms. It might also be mentioned that although
arid-mountainous and a few mesic species have well developed subelytral cavities, they
exhibit no more heat tolerance than winged mesic species without subelytral spaces.
The small size of arthropods may ultimately preclude any effective adaptations (other
than behavioral adaptations) for regulating temperature. Although desert tenebrionids
are generally fairly large beetles (often exceeding 1 cm), they are still too small to
possess much thermal inertia. Tenebrionids' internal body temperatures generally
parallel black bulb temperatures (Seely et al. 1988). Thus, the beetles act as physical
bodies in terms of heat gain; they quickly equilibrate with external temperatures (Edney
1971b, Henwood 1975). Without active (physical or physiological) cooling mechanisms,
beetles must avoid prolonged exposure to temperatures above their tolerance limit.
It is now generally believed that one of the major functions of the subelytral cavity
is to conserve water by reducing transpiration. Although Buxton (1923) did not propose
a water-conserving function for the cavity, a statement he wrote was prophetic of
research and thought to come:
"The extremely close fit between the margins of the elytra and the sternite is
remarkable: it would appear to hinder the ventilation of the subelytral space into
which the spiracles open."
The importance of this airtight cavity has since been demonstrated by modifying it, so
that it is no longer airtight (Dizer 1955). Pieces of elytra were removed from various
species of beetles, some from the Turkmenistan Desert with well developed subelytral
cavities, and some lightly sclerotized and winged forms from the Russian steppe. After
elytral removal, the beetles with (formerly) hermetic subelytral cavities showed a great
increase in rate of water loss, while the water loss of winged species increased only a
little. Similar results were obtained independently some nine years later by Cloudsley-
Thompson (1964), who found Dizer's work after his experiments were completed. In
three short pages, Cloudsley-Thompson simultaneously rejected the idea of a heat func-
tion for the cavity (discussed above) and demonstrated via elytral removal experiments
the efficacy of the cavity in reducing transpiration. These finding were reconfirmed by
Ahearn & Hadley (1969). These authors hypothesized that transpiration is decreased
because the abdominal spiracles open into the relatively humid subelytral cavity rather
than the outside air. The spiracles are protected from the drying effects of wind, and
Florida Entomologist 76(4)
the high humidity within the cavity reduces the vapor pressure gradient between the
tracheae and their exterior environment, reducing evaporation via the spiracles.
Ahearn (1970) presented evidence that the airflow with the beetle's tracheal system
is unidirectional, with the thoracic spiracles serving only to take in air. Although the
thoracic spiracles are not protected by the subelytral cavity, little water should be lost
from them since air does not flow outward through them. All moist exhalent air empties
into the subelytral cavity, increasing its humidity and slowing evaporation even more.
Fiori (1977) argued that the subelytral cavity reduces transpiration mainly by slow-
ing evaporation from the membranous abdominal uroterga and their lateral and inter-
segmental membranes, which constitute the floor of the cavity. He noted that the
abdominal spiracles are imbedded in atria and can be closed, indicating that they may
already be adequately protected. Probably, the cavity prevents potential loss from both
sources. In any case, water that would otherwise be lost is conserved, enabling the
beetle to survive longer dry periods.
For the subelytral cavity to effectively reduce transpiration, it must be airtight.
Provided a hermetic seal is achieved, the size of the cavity should not affect its water
conserving efficiency. However, great interspecific variation exists in the size (volume)
of the cavity in relation to the size of the beetle. There is no clear correlation between
cavity size and size of beetle, habitat, or activity times.
Cloudsley-Thompson (1971) maintained that cavity size has no direct effect on indi-
vidual beetle fitness, and cited variability in cavity size as an example of non-adaptive
In his morphological study of the subelytral cavity, Fiori (1977) observed that the
size of the cavity functions to allow abdominal expansion in these heavily sclerotized
beetles. The abdomens of these beetles can only expand via extension of intersegmental
membranes between the uroterga which make up the floor of the cavity. The subelytral
cavity thus provides room for the uroterga to raise up with increase in abdominal
volume. Fiori noted that the ability to expand the abdomen may be important to beetles
which must feed on large quantities of their detrital food, which generally is of low
nutritional value. He observed that after feeding, the uroterga may expand to occupy
almost the entire subelytral space.
Slobodchikoff & Wisman (1981) elaborated on the abdominal expansion hypothesis,
although Fiori (1977) was not cited. They hypothesized that the abdominal expansion
into the subelytral cavity allows tenebrionids to drink and store large quantities of
water. This may be important in enabling the beetles to take advantage of ephemeral
water sources, such as rain puddles, which are present only for a short time following
the occasional rain event. Slobodchikoff & Wisman's (1981) observations on changes of
cavity depth as the beetle's water status changed supports their hypothesis.
Fiori (1977) also invokes his abdominal expansion hypothesis to explain the sexual
dimorphism in subelytral cavity size seen in many species. Females generally possess
cavities which are proportionally larger than those of males. Fiori suggested that
females need a larger subelytral cavity to accommodate additional abdominal expansion
caused by the growth of eggs in the ovarioles and egg-calyces prior to oviposition.
The subelytral cavity may be seen as a hermetic seal which reduces transpiration
regardless of its size, and as a protected space within the beetle which, whether airtight
or not, allows the abdomen to expand to store food, water, or eggs. Its effect as a heat
absorber has not been clearly demonstrated, and may not be significant.
I hypothesize that the cavity itself is probably an architectural constraint resulting
from the need for abdominal expansion within the heavily sclerotized, fused elytra that
evolved primarily as a water conservation adaptation. The cavity may represent the
same type of architectural constraint as the triangular spandrels which support the
domes of cathedral ceilings (Gould & Lewontin 1979). The spandrels are beautiful in
Draney: Subelytral Cavity
their own right, containing artwork seemingly perfectly suited to presentation on a
triangular space. However, the spandrels were not designed to bear artwork-they are
simply necessary architectural by-products of mounting a dome on rounded arches. So
too may the actual cavity of desert tenebrionids be an architectural by-product of having
a well sclerotized and hermetically sealed abdomen. The real adaptation here is the set
of elytral joints described by Fiori (1977) which serves to create an airtight abdominal
exoskeleton. The cavity is present simply to give the beetle needed "wiggle room"
within its exoskeleton. Any other use the beetle has for the cavity (such as the possible
insulating effects of the air space) may be seen as secondary uses in the same way that
the cathedral spandrels were put to use bearing artwork.
Although the subelytral cavity is an intriguing "place," little recent work has been
done on it (C. S. Crawford pers. comm.). It appears that research on the subelytral
cavity follows close behind new suggestions concerning its function. Most of the inves-
tigations into the subelytral cavity seem to have been sparked by ideas in the seminal
papers of Cloudsley-Thompson (1964) and Fiori (1977). Our understanding of the sub-
elytral cavity is far from complete. Research will continue when someone forwards a
new hypothesis, a reason to question our current notions.
I thank C. S. Crawford (University of New Mexico) for making some important but
difficult to find literature available to me; V. L. Medland (University of Georgia) for
review of the paper and for much patience; D. J. Moellenbeck (Louisiana State Univer-
sity) and several anonymous reviewers for thorough editorial comments; and M. R.
Conley and W. G. Whitford (New Mexico State University) for teaching me about the
AHEARN, G. A. 1970. The control of water loss in desert tenebrionid beetles. J. Exp.
Biol. 53: 573-595.
AHEARN, G. A. 1971. Ecological factors affecting population sampling of desert teneb-
rionid beetles. Am. Midl. Nat. 86: 385-406.
AHEARN, G. A., AND N. F. HADLEY. 1969. The effects of temperature and humidity
on water loss in two desert tenebrionid beetles, Eleodes armata and Crypto-
glossa verrucosa. Comp. Biochem. Physiol. 30: 739-749.
ALLSOPP, P. G. 1980. The biology of false wireworms and their adults (soil-inhabiting
Tenebrionidae) (Coleoptera): A review. Bull. Entomol. Res. 70: 343-379.
BOLWIG, N. 1957. Experiments on the regulation of the temperature of certain teneb-
rionid beetles. J. Entomol. Soc. S. Afr. 20: 454-458.
BROZA, M. 1979. Dew, fog, and hygroscopic food as a source of water for desert ar-
thropods. J. Arid Env. 2: 43-49.
BUXTON, P. A. 1923. Animal life in deserts-a study of fauna in relation to environ-
ment. Edward Arnold & Co., Ltd. 176 pp.
BUXTON, P. A. 1924. Habits of some tenebrionid beetles. Entomol. Mon. Mag. 60: 3-7.
CLOUDSLEY-THOMPSON, J. L. 1964. On the function of the sub-elytral cavity in desert
Tenebrionidae (Col.). Entomol. Mon. Mag. 100: 148-151.
CLOUDSLEY-THOMPSON, J. L. 1971. Non-adaptive variation and group selection. Sci.
Prog., Oxf. 59: 243-254.
CLOUDSLEY-THOMPSON, J. L. 1977. The genus Cossyphus (Col., Tenebrionidae): A
striking instance of protective resemblance. Entomol. Mon. Mag. 113: 151-152.
CLOUDSLEY-THOMPSON, J. L., AND M. J. CHADWICK. 1964. Life in Deserts. Dufour,
Philadelphia, PA. 218 pp.
548 Florida Entomologist 76(4) December, 1993
CRAWFORD, C. S. 1990. Scorpiones, Solifugae, and associated desert taxa, pp. 421-476
in D. L. Dindal, [ed.] Soil Biology Guide. John Wiley & Sons, NY, NY. 1349 pp.
CRAWFORD, C. S. 1981. Biology of Desert Invertebrates. Springer-Verlag, NY, NY.
DIZER, Y. B. 1955. On the physiological role of the elytra and sub-elytral cavity of
steppe and desert Tenebrionidae [in Russian]. Zool. Zh. S. S. R. 34: 319-322.
DOYEN, J. T., AND W. R. TSCHINKEL. 1982. Phenetic and cladistic relationships
among tenebrionid beetles (Coleoptera). Systematic Entomol. 7: 127-183.
EDNEY, E. B. 1971a. The body temperature of tenebrionid beetles in the Namib De-
sert of southern Africa. J. Exp. Biol. 55: 253-272.
EDNEY, E. B. 1971b. Some aspects of water balance in tenebrionid beetles and a
thysanuran from the Namib Desert of southern Africa. Physiol. Zool. 44: 61-76.
EISNER, T., AND J. MEINWALD. 1966. Defensive secretions of arthropods. Science
EL RAYAH, E. A. 1970. Some reactions of two desert beetles Adesmia antiqua and
Pimelia grandis (Tenebrionidae) to temperature. Entomol. Exp. Appl. 13: 286-
FIORI, G. 1977. La cavita sottoelitrale dei tenebrionidi apomorfi. Redia 60: 1-112.
GOULD, S. J., AND R. C. LEWONTIN. 1979. The spandrels of San Marco and the
Panglossian paradigm: a critique of the adaptationist programme. Proc. R. Soc.
Lond. 205: 581-598.
HADLEY, N. F. 1970. Micrometeorology and energy exchange in two desert ar-
thropods. Ecology 51: 434-444.
HADLEY, N. F. 1972. Desert species and adaptation. American Scient. 60: 338-347.
HAMILTON, W. J. III. 1975. Coloration and its thermal consequences for diurnal de-
sert insects, pp. 67-89 in N. F. Hadley, [ed.] Environmental physiology of desert
organisms. Dowden, Hutchinson, and Ross, Stroudsberg, PA.
HAMILTON, W. J. III, AND M. K. SEELY. 1976. Fog basking by the Namib Desert
beetle, Onymacris unguicularis. Nature (London) 262: 284-285.
HENWOOD, K. 1975. A field-tested thermoregulation model for two diurnal Namib
Desert tenebrionid beetles. Ecology 56: 1329-1342.
KOCH, C. 1961. Some aspects of abundant life in the vegetationless sand of the Namib
Desert dunes-positive psammotropism in tenebrionid beetles. J. SW. Afr. Sci.
Soc. 15: 9-33.
KOCH, C. 1962. The Tenebrionidae of southern Africa XXXI. Comprehensive notes on
the tenebrionid fauna of the Namib Desert. Ann. Transvaal. Mus. 24: 61-103.
KRAMM, R. A., AND K. R. KRAMM. 1972. Activities of certain species of Eleodes in
relation to season, temperature, and time of day at Joshua Tree National Monu-
ment (Coleoptera: Tenebrionidae). Southwest. Nat. 16: 341-355.
MCKINNERNEY, M. 1978. Carrion communities in the northern Chihuahuan Desert.
Southwest. Nat. 23: 563-576.
RAFES, P. M. 1960. The life forms of insects inhabiting the Naryn sands of the
semidesert Transvolga region. Entomol. Rev. 38: 19-31.
SCHMOLLER, R. R. 1970. Terrestrial desert arthropods: Fauna and ecology. The
Biologist 52: 77-98.
SEELY, M. K., AND G. N. Louw. 1980. First approximation of the effects of rainfall
on the ecology and energetic of a Namib Desert dune ecosystem. J. Arid Env.
SEELY, M. K., C. S. ROBERTS, AND D. MITCHELL. 1988. High body temperature of
Namib dune tenebrionids-why? J. Arid Env. 14: 135-143.
SLOBODCHIKOFF, C. N. 1979. Utilization of harvester ant debris by tenebrionid bee-
tles. Environ. Entomol. 8: 770-772.
SLOBODCHIKOFF, C. N. 1983. Water balance and temperature preferences, and their
role in regulating activity times of tenebrionid beetles. Oikos 40: 113-119.
SLOBODCHIKOFF, C. N., AND K. WISMAN. 1981. A function of the subelytral chamber
of tenebrionid beetles. J. Exp. Biol. 90: 109-114.
TANNER, V. M., AND W. A. PACKHAM. 1965. Tenebrionidae beetles of the Nevada
test site. Brigham Young Univ. Sci. Bull., Biol. Ser. 6: 1-44.
Ghidiu & Andaloro: Fall Armyworm Instar and Susceptibility 549
THOMAS, D. B. JR. 1979. Patterns in the abundance of some tenebrionid beetles in the
Mojave Desert. Environ. Entomol. 8: 568-574.
TSCHINKEL, W. R. 1975. A comparative study of the chemical defensive system of
tenebrionid beetles. Chemistry of the secretions. J. Insect Physiol. 21: 753-783.
WALLWORK, J. A. 1982. Desert soil fauna. Praeger Pub. NY, NY. 296 pp.
WISE, D. H. 1981. A removal experiment with darkling beetles: Lack of evidence for
interspecific competition. Ecology 62: 727-738.
ZACHARIASSEN, K. A. 1977. Ecophysiological studies on beetles from arid regions in
east Africa. Norw. J. Entomol. 24: 167-170.
ZACHARIASSEN, K. A., J. ANDERSON, G. M. O. MALOIY, AND J. M. Z. KAMAU.
1987. Transpiratory water loss and metabolism of beetles from arid areas in east
Africa. Comp. Biochem. Physiol. 86A: 403-408.
THE RELATIONSHIP BETWEEN FALL ARMYWORM
(LEPIDOPTERA: NOCTUIDAE) INSTAR AND
SUSCEPTIBILITY TO INSECTICIDES APPLIED TO SWEET
GERALD M. GHIDIU' AND JOHN T. ANDALORO2
'Dept. of Entomology, Rutgers University
New Brunswick, NJ 08903
2DuPont Experiment Station, Stine-Haskell Building
Newark, DE 19711
Toxicities of 2 concentrations of fenvalerate and methomyl to second, fourth and
sixth instar fall armyworms, Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noc-
tuidae), were determined in laboratory and field tests. In laboratory tests, the LC50 for
both methomyl and fenvalerate increased as larval age increased from second to fourth
to sixth instar. Fenvalerate was more toxic than methomyl to second instars but less
toxic than methomyl to sixth instars. In the 1986 field tests with whorl stage sweet
corn, all treatments effectively reduced whorl damage caused by second, fourth and
sixth instars compared with the untreated. Both rates of methomyl were significantly
more effective than either rate of fenvalerate in reducing whorl damage caused by
fourth or sixth instar fall armyworm. In 1987, both rates of methomyl and the 0.224 kg
AI/ha rate of fenvalerate significantly reduced second and fourth instar fall armyworm
whorl damage compared with the untreated, but only the 1.0 kg AI/ha rate of methomyl
significantly reduced sixth instar whorl damage. All insecticide treatments resulted in
significantly fewer second instars recovered from the whorls, but only the methomyl
treatments resulted in significantly fewer fourth instars compared with the untreated.
Greater whorl damage was caused by fourth and sixth instars compared with second
instars. Results indicate that timing of application is as important as selection of pes-
ticide for management of fall armyworm on whorl stage sweet corn.
Key Words: Spodoptera frugiperda, corn, whorl damage, fenvalerate, methomyl.
Se determinaron on pruebas de laboratorio y de campo la toxicidad de dos concen-
traciones de fenvalerato y metomilo a los segundos, cuartos y sextos instares del qusano
550 Florida Entomologist 76(4) December, 1993
cogollero del maiz, Spodoptera frugiperda (J. E. Smith) Lepidoptera: Noctuidae). En
las pruebas de laboratorio, el LC, para metomilo y fenvalerato se aument6 a media
que se aument6 la edad de las larvas desde el segundo al cuarto y al sexto instar.
Fenvalerato fu4 mas t6xico que metomilo a los segundos instares pero menos t6xico que
metomilo a los sextos instares. En las pruebas de campo de 1986 con maiz dulce en la
etapa del verticilo, todos los tratamientos efectivamente redujeron los daios a los ver-
ticilos causodos por segundos, cuartos y sextos instares al compararse con el testigo.
Ambas concentraciones de metomilo fueron significativamente mAs efectivos que cual-
quiera de las dos concentraciones de fenvalerato en reducir dafios al verticilo causado
por gusanos cogolleros del maiz de cuarto o sexto instar. En 1987, ambas concen-
traciones de metomilo y la concentraci6n de 0.224 kg AI/ha do fenvalerato significativa-
ment redujeron dafos de segundo y cuarto instar al verticilo comparado con el testigo,
pero solamente la concentration de 1.0 kg AI/ha de metomilo significativamente redujo
dafios causados por el sexto instar. Todos tratamientos resultaron en significativemente
menos segundos instares recuperados de los verticilos, pero solamente los tratamientos
de metomilo resultaron en significativamente menos cuartos instares comparado con el
testigo. El dafo mas grande al verticilo fud causado por cuartos y sextos instares com-
parado con segundo instares. Los resultados indican que escoger la ocasi6n precise de
alicaci6n de plaguecidas es tan important como la selecci6n de la plaguecida en el
control del gusano cogollero de maiz sobre maiz dulce en la etapa del verticilo.
The fall armyworm (FAW), Spodopterafrugiperda (J. E. Smith) is a serious pest of
sweet corn (Zea mays L.) in the mid-Atlantic region. Adult moths migrate into the
region from southern states during early to mid-August, and can infest sweet corn at
nearly all crop stages. Control of this insect pest throughout Central and North America
is obtained primarily with insecticides, including fenvalerate and methomyl.
With many insects, susceptibility to insecticides generally decreases with increasing
larval age. Ahmad & Forgash (1975) reported that susceptibility of gypsy moth larvae
to carbaryl and diazinon decreased with older instars, and Kuhr & Hessney (1977)
demonstrated that fifth instar European corn borers were less susceptible to methomyl
than third or fourth instars. Yu (1983) reported that, in laboratory tests, the LD,5
values for methomyl, diazinon and permethrin increased with older instars.
Laboratory and field data are often difficult to compare (Bagent 1964), and reports
of the efficacy of insecticides against fall armyworm in laboratory tests have not been
consistent with field trials. Bass (1978) reported poor control of fall armyworm with
permethrin and fenvalerate in field trials, but Wood et al. (1981) found permethrin to
be more toxic to fall armyworm in laboratory tests than several other insecticides.
The objective of our study was to conduct laboratory and field tests to determine
the susceptibility of different FAW instars to two commercially available insecticides,
fenvalerate and methomyl. Results of insecticide bioassays with laboratory-reared fall
armyworm were compared with field experiments to evaluate the effect of insecticides
on fall armyworm mortality and feeding damage to sweet corn.
MATERIALS AND METHODS
In 1988, FAW were reared from eggs on a wheat germ-agar artificial diet (Bio-Serv
Corn Earworm Rearing Media, Bio-Mix #9394, Bio-Serv, Frenchtown, NJ) in the lab-
oratory at the E. I. du Pont de Nemours' Stine-Haskell Research Farm, Newark,
Delaware (USA). Larvae were maintained on diet in rearing chambers at 60% R.H.
and 26.7 C before testing.
Ghidiu & Andaloro: Fall Armyworm Instar and Susceptibility 551
Fall armyworm larvae (5 second, fourth or sixth instars) were placed in flat, plastic
237-ml cups (11.4 cm diam) containing an artificial wheat germ-agar diet for each treat-
ment replicate. Cups were placed on a moving belt sprayer that passed beneath a single
stationary nozzle (8001E) calibrated to deliver 935.4 liters/ha at 1.2 kPa. Concentrations
of technical methomyl and fenvalerate in ppm were respectively: 12.5, 25, 50, 100, 200,
400 and 6.25, 12.5, 25, 50, 100, 200. An acetone control treatment was included. Treat-
ments were replicated 4 times. All larvae were maintained in temperature-controlled
rearing chambers for 48 h after application. Data from laboratory tests were corrected
for control mortality with Abbott's (1925) formula. Log dosage-probit lines were deter-
mined using probit analysis (SAS Institute 1985).
'Silver Chief cultivar sweet corn was planted on 1 July 1986 and 15 June 1987 in a
Sassafras sandy loam field at the Rutgers Research & Development Center, Bridgeton,
New Jersey (USA). In southern New Jersey, sweet corn planted in mid- to late June
would be in the mid-whorl stage of development before infestation by fall armyworm.
Treatment rows were 7.62 m long and 0.8 m wide. Treatments were replicated 4 times
in a split-plot experimental design. The main plots received the following rates of insect
icide: fenvalerate (Pydrin 2.4EC) at 0, 0.112 or 0.224 kg AI/ha and methomyl (Lannate
1.8L) at 0, 0.50 or 1.00 kg AI/ha. Subplots were infested with different fall armyworm
instars (second, fourth or sixth). Each subplot consisted of 4 rows: the middle 2 rows
were treatment rows and the outer 2 rows were nontreated buffer rows. Twenty plants
from the center 2 rows of each subplot were infested on 14 August 1986 and 17 July
1987 with either first, third or fifth instars, and marked with a 0.3 m garden stake
placed in the soil at the base of the plant. First instars were applied with a Bio-Serv
#9040 "bazooka" dry grit applicator (approximately 25 larvae per whorl), and third and
fifth instars were applied with soft forceps (4 larvae per plant whorl, or 80 larvae per
subplot). Foliar sprays were applied 48 h after infestation using a self-propelled, high-
clearance sprayer (John Deere 6000 Hi-Cycle) with a single nozzle (TeeJet 8010) cen-
tered over each of the middle 2 rows. Sprayer application volume was 224.4 liter/ha at
275.8 kPa pressure and a sprayer ground speed of 3.2 km/h. Dissection of infested,
untreated plants each year showed that at the time of application of the foliar sprays,
or shortly thereafter, larvae had aged by one instar. Infested plants were evaluated on
22 August 1986 and 21 July 1987 for FAW whorl damage according to a modified method
from Carvalho (1970): (0) no damage; (1) slight damage, 0-10% whorl feeding; (2) moder-
ate damage, 10-25% whorl feeding; (3) heavy damage, 25-50% whorl feeding; (4) severe
damage, 50-75% whorl feeding, emerging tassel damaged and (5) whorl destroyed. Ten
whorls from each subplot were dissected in the field and the numbers of live FAW
larvae were recorded. The remaining plants were harvested on 7 September 1986 and
18 August 1987. The ears were husked and weighed and FAW damage evaluated as
"marketable" (clean or less than 3.6 cm of tip damage) or "cull" (tip damage beyond 3.6
cm and/or side or bottom damage).
The data from individual plots were averaged to obtain a plot mean for each insect-
icide and FAW instar treatment. Data for each year and parameter were subjected to
analyses of variance (ANOVA) (SAS 1985). Means were separated by Duncan's (1955)
new multiple range test.
Fenvalerate was significantly more toxic to second instar larvae than methomyl, but
methomyl was significantly more toxic to sixth instars than fenvalerate (Table 1). LCs5's
Florida Entomologist 76(4)
TABLE 1. LABORATORY TOXICITY DATA FOR METHOMYL AND FENVALERATE
APPLIED TO 2ND, 4TH AND 6TH INSTAR FALL ARMYWORM, NEWARK,
Insecticide Instar LCo50 95% Fiducial Limits
Methomyl 2nd 103.9 79.9- 147.1
Methomyl 4th 157.7 105.0- 265.5
Methomyl 6th 189.4 134.4- 299.9
Fenvalerate 2nd 35.8 31.6- 40.3
Fenvalerate 4th 136.6 112.7- 174.4
Fenvalerate 6th 823.0 341.9-7130.0
'ig/fall armyworm larva.
increased as instar increased from second to fourth to sixth for both methomyl and
fenvalerate. In addition, sixth instars were 1.7- and 10.8-fold more tolerant of methomyl
and fenvalerate, respectively, than were 2nd instars. These results are in agreement
with those of Yu (1983) who reported that LD5o's for methomyl, diazinon and permethrin
increased with increasing FAW larval age.
In each year, ANOVA demonstrated significant (P<0.05) insecticide (whole plots),
instar (sub-plots) and insecticide X instar interactions for both whorl damage ratings
and recovered larvae (Table 2). When the data from 1986 and 1987 were combined, both
rates of methomyl were significantly more effective than the 0.112 kg AI/ha rate of
fenvalerate in reducing whorl damage caused by FAW (Table 3); no significant differ-
ences between either rate of fenvalerate were observed. Methomyl (1.0 kg AI/ha) re-
sulted in significantly fewer larvae recovered from whorls than all other treatments
Yields. No significant treatment differences were observed for mean ear weight,
total number of ears per 20 stalks, or ear damage caused by FAW in 1986 or 1987 (yield
data not shown). Ghidiu & Drake (1989) reported that no significant loss in marketable
sweet corn ears occurred when plants were infested by FAW larvae during the pre-
whorl to mid-whorl stages of plant development.
Field Trials 1986.
Whorl Damage Ratings. The whorl damage caused by fall armyworm significantly
increased as larvae aged from second to fourth instar in untreated plots (Table 4). All
TABLE 2. ANALYSIS OF VARIANCE STATISTICS FROM FIELD EXPERIMENTS.
Whorl Damage' No. Larvae Recovered1
Year Variable df F F
1986 Insecticide 5 21.84 35.31
Instar 2 93.39 295.41
Insecticide x Instar 10 5.08 9.61
1987 Insecticide 5 5.04 13.03
Instar 2 122.80 100.83
Insecticide x Instar 10 2.99 4.51
P> 0.05 for analysis of all variables in both years.
Ghidiu & Andaloro: Fall Armyworm Instar and Susceptibility 553
TABLE 3. EFFECT OF INSECTICIDES ON FALL ARMYWORM IN WHORL STAGE
SWEET CORN, BRIDGETON, NEW JERSEY'.
Mean No. Live
Rate Whorl Larvae Per
Treatment (kg AI/ha) Damage2.3 20 Corn Whorls3
Untreated 0.0 2.3 c 19.6 b
Fenvalerate 0.112 1.7b 10.5 a
Fenvalerate 0.224 1.4 ab 8.8 a
Methomyl 0.5 1.0a 6.3 a
Methomyl 1.0 0.9 a 6.5 a
'1986 and 1987 data combined.
2Rated 0-5 (0 = no damage, 5 = whorl destroyed).
'Numbers in a column with a letter in common are not significantly different (Duncan's multiple range test;
treatments significantly reduced whorl damage by all 3 instars compared with the un-
treated. Both rates of methomyl resulted in significantly less whorl damage by fourth
and sixth instar FAW than did either rate of fenvalerate.
Recovery of FAW. All treatments resulted in lower recovery of second instars com-
pared with the untreated control (Table 4). Both rates of methomyl resulted in signifi-
cantly fewer fourth instars recovered than did either rate of fenvalerate. Although
fewer sixth instars were recovered in corn plots treated with both rates of methomyl
and the 0.224 kg AI/ha rate of fenvalerate, the differences were not significant compared
with untreated corn.
Field Trials 1987.
Whorl Damage Ratings. Both rates of methomyl and the 0.2 kg AI/ha rate of fenval-
erate significantly reduced whorl damage caused by second or fourth instars (Table 5)
compared with the untreated. However, there were no significant differences among
insecticide treatments in whorl damage caused by second or fourth instars. Only the
1.0 kg AI/ha rate of methomyl resulted in significantly less whorl damage caused by
sixth instar FAW.
Recovery of FAW. The 1.0 kg AI/ha rate of methomyl resulted in significantly fewer
second instars recovered compared with all other treatments, although all treatments
resulted in significantly (P2-0.05) fewer second instars than did the untreated control
TABLE 4. EFFECT OF INSECTICIDES ON VARIOUS FALL ARMYWORM INSTARS IN
THE WHORL OF SWEET CORN, BRIDGETON, NJ, 1986.
Whorl Damage' Larvae Recovered'
Rate Instar Instar
Treatment (kg AI/ha) 2nd 4th 6th 2nd 4th 6th
Untreated 1.1 b 2.6 d 3.6 c 26 c 43 b 12 ab
Fenvalerate 0.112 0.3 a 1.8 c 2.5 b 9 ab 32 b 14 ab
Fenvalerate 0.224 0.3 a 1.7 c 1.8 b 6 a 39 b 5 a
Methomyl 0.5 0.3 a 1.1 b 1.0 a 10 ab 17 a 4 a
Methomyl 1.0 0.3 a 0.9 ab 1.0 a 9ab 21 a 5 a
'Numbers in a column with a letter in common are not significantly different. (Duncan's multiple range test;
554 Florida Entomologist 76(4) December, 1993
TABLE 5. EFFECT OF INSECTICIDES ON VARIOUS FALL ARMYWORM INSTARS IN
THE WHORL OF SWEET CORN, BRIDGETON, NJ, 1986.
Whorl Damage' Larvae Recovered'
Rate Instar Instar
Treatment (kg AI/ha) 2nd 4th 6th 2nd 4th 6th
Untreated 1.4 bc 1.4 bc 2.7 b 105 b 35 bc 23 ab
Fenvalerate 0.112 1.1 ab 1.1 ab 3.3 b 70 c 22 ab 14 a
Fenvalerate 0.224 0.9 a 1.0 a 2.7 b 67 c 25 ab 11 a
Methomyl 0.5 0.8 a 0.9 a 2.6 b 53 bc 19 a 14 a
Methomyl 1.0 0.5 a 0.6 a 1.8 a 33 a 15a 8 a
'Numbers in a column with a letter in common are not significantly different (Duncan's multiple range test;
(Table 5). Both rates of methomyl resulted in significantly fewer fourth instars com-
pared with the untreated control, although there were no significant differences in
recovered fourth instars among the insecticide treatments. Further, there were no
significant differences among any treatments in the numbers of sixth instars recovered
from the whorls.
Laboratory studies indicated that fenvalerate was more toxic to second instars than
methomyl (based on LCo5 values). However, field studies demonstrated that fenvalerate
and methomyl were equally toxic to second instars. In the laboratory, fenvalerate and
methomyl were equally toxic to fourth instars, but methomyl was more toxic to fourth
instars than fenvalerate in the field experiments (1986). Methomyl was more toxic to
sixth instars than fenvalerate in the laboratory, but in field studies this trend was
observed only in whorl damage and not with numbers of larvae recovered. These find-
ings agree with previous research demonstrating the difficulty in making meaningful
comparisons between laboratory and field results (Bagent 1964).
Laboratory studies also showed that LC5 values increased for both fenvalerate and
methomyl as test fall armyworm aged from second to sixth instars. This agrees with
Yu (1983) who reported that the LD5o values of permethrin and methomyl increased as
the fall armyworm larvae aged.
In each year of the field study, the whorl damage caused by fall armyworm increased
as larval age (size) increased. Mulder & Showers (1986) showed that in field corn,
defoliation increased as armyworms (Pseudaletia unipuncta Haworth) aged from fourth
through sixth instars.
Our field study results showed that methomyl and fenvalerate were equally effective
in reducing whorl damage by second instars during each year, but methomyl was more
effective than fenvalerate in reducing whorl damage by fourth and sixth instars in 1986.
In 1986, both rates of fenvalerate and methomyl significantly reduced whorl damage
caused by sixth instars compared with untreated corn, but only the high rate of
methomyl significantly reduced sixth instar whorl damage in 1987. Guillebeau & All
(1991) reported that fenvalerate gave more variable control of fall armyworm on sweet
corn than did methomyl.
Both rates of fenvalerate and methomyl significantly (P< 0.05) reduced the numbers
of live second instars recovered in the whorls each year, and both rates of methomyl
significantly reduced the numbers of fourth instars recovered each year, compared with
untreated corn. However, none of the fenvalerate or methomyl treatments significantly
Ghidiu & Andaloro: Fall Armyworm Instar and Susceptibility 555
reduced the numbers of sixth instar FAW recovered in the whorls in either year com-
pared with untreated corn.
Greater numbers of second instar FAW were recovered from the whorls in 1987
than in 1986. This may have been due to rainfall [0.51 cm (0.21 inches)] on 17 August
1986, the day after the spray application. The rain may have helped carry the spray
residue deeper into the whorl, and thus increased the amount of toxicant at the larval
feeding site. No rainfall occurred within 1 wk of the 1987 spray application.
Control of fall armyworm may depend not only on selection of an effective insecticide,
but also on timing of application. An application of fenvalerate or methomyl would be
less effective against older larvae (sixth instar) compared with younger larvae. A delay
in application, either due to poor weather conditions or inaccurate timing, may result
in reduced effectiveness of the insecticide, rectified only by additional pesticide applica-
tions or increasing dosage. Such a delay would be counterproductive to a sound fall
armyworm management program. Pesticide applications should be directed against the
younger instars to obtain best results. A pest management program should, therefore,
utilize larval instar data, as well as percentage infestation (number of plants infested
during the whorl stage of plant development) to make more precise spray decisions and
maximize crop protection.
New Jersey Agricultural Experiment Station publication No. D-08120-05-89 sup-
ported by State funds and by the United States Hatch Act. The authors thank E. I. du
Pont de Nemours & Co. for technical equipment and assistance.
ABBOTT, W. S. 1925. A method of computing the effectiveness of an insecticide. J.
Econ. Entomol. 18: 265-267.
AHMAD, S., AND A. FORGASH. 1975. Toxicity of carbaryl and diazinon to gypsy moth
larvae: changes in relation to larval growth. J. Econ. Entomol. 68: 803-806.
BAGENT, J. L. 1964. The response of two species of Heliothis to certain insecticides
in field and laboratory tests. M.S. Thesis, Louisiana State Univ. 45pp.
BASS, M. H. 1978. Fall armyworm: evaluation of insecticides for control. Agric. Exp.
Stn., Auburn Univ. Leaflet 93: 7.
CARVALHO, R. P. L. 1970. Danos, flutuacao da populacao control e comportamento
de Spodopterafrugiperda (J. E. Smith) e susceptibilidae de diferentes genotipos
de milho em condicoes de campo. Piracicaba, Brasil, ESALQ, 1970. Ph.D. disser-
DUNCAN, D. B. 1955. Multiple range test and multiple F tests. Biometrics 11: 1-42.
GHIDIU, G. M., AND G. E. DRAKE. 1989. Fall armyworm (Lepidoptera: Noctuidae)
damage relative to infestation level and stage of sweet corn development. J.
Econ. Entomol. 82(4): 1197-1200.
GUILLEBEAU, L. P., AND J. N. ALL. 1991. Use of pyrethroids, methomyl and chlor-
pyrifos to control fall armyworm (Lepidoptera: Noctuidae) in whorl stage field
corn, sweet corn and sorghum. Florida Entomol. 74: 261-270.
KUHR, R. J., AND C. W. HESSNEY. 1977. Toxicity and metabolism of methomyl in
the European corn borer. Pestic. Biochem. Physiol. 7: 301-308.
MULDER, P. G., AND W. B. SHOWERS. 1986. Defoliation by the armyworm (Lepidopt-
era: Noctuidae) on field corn in Iowa. J. Econ. Entomol. 79: 368-373.
SAS INSTITUTE. 1985. SAS User's guide: statistics. SAS Institute, Cary, N.C.
WOOD, K. A., B. H. WILSON, AND J. B. GRAVES. 1981. Influence of host plant on
the susceptibility of the fall armyworm to insecticides. J. Econ. Entomol. 74: 96-
Yu, S. J. 1983. Age variation in insecticide susceptibility and detoxification capacity
of fall armyworm (Lepidoptera: Noctuidae) larvae. J. Econ. Entomol. 76: 219-
Florida Entomologist 76(4)
LACK OF EVIDENCE FOR THE
TOXIC NECTAR HYPOTHESIS: A PLANT ALKALOID
DID NOT DETER NECTAR FEEDING BY LEPIDOPTERA
P. J. LANDOLT AND B. LENCZEWSKI
Insect Attractants, Behavior, and Basic Biology Research
Laboratory, Agricultural Research Service
U.S. Department of Agriculture, Gainesville, FL 32604
Floral nectars of many plant species contain alkaloids and other allelochemics that
might deter butterfly visitors and promote flower constancy by specialized pollinators.
The pyrrolizidine alkaloid, monocrotaline, has been implicated as such a feeding inhibitor
in nectar, in support of this toxic nectar hypothesis. We tested this hypothesis by
evaluating monocrotaline for deterrence of nectar feeding in the tobacco budworm moth,
Heliothis virescens (Fab.), the cabbage looper moth, Trichoplusia ni (HUbner), and the
gulf fritillary butterfly, Agraulis vanilla (L.). Concentrations of monocrotaline added
to aqueous sucrose solutions did not reduce consumption in these three species, even
with near saturation concentrations of monocrotaline. Also, gulf fritillary butterflies did
not alter their preference for visiting orange artificial flowers when given the choice
between yellow flowers with a sugar solution and orange flowers with a monocrotaline
and sugar solution. Patterns of flower visitation by Lepidoptera are likely due to a
combination of factors, such as attractants, feeding stimulants and deterrents, visual
stimuli, flower morphology, and ecological factors that control nectar availability.
Key Words: Insecta, Trichoplusia ni, Heliothis virescens, Agraulis vanilla, deterrent,
Los nectares florales de muchas species de plants contienen alcaloides y otros
aleloquimicos los cuales pueden impedir a los visitantes lepid6pteros y promover la
constancia floral por polinizadores especializados. El alcaloide pirrolizidino, mono-
crotalino, se ha sido implicado como un inhibidor en el nectar, en apoyo de este hip6tesis
sobre el nectar t6xico. Nosotros hicimos una prueba de este hip6tesis por evaluar el
monocrotalino como un refrenamiento a la alimentaci6n de nectar en el cogollero del
tabaco, Heliothis virescens (Fab.), el falso medidor del repollo, Trichoplusia ni
(Hiibner), y la mariposa "gulf fritillary," Agraulis vanilla (L.). Concentraciones de
monocrotalino afiadidas a soluciones acuosas de sucrosa no redujeron el consume en
estas tres species, ni hasta con concentraciones de monocrotalino casi saturadas.
Ademas, las mariposas "gulf fritillary" no alteraron su preferencia por visitar flores
anaranjadas artificiales, dado un surtido entire flores amarillas con una soluci6n de azucar
y flores anaranjadas con una soluci6n de monocrotalino y azucar. Los patrons de visita-
ci6n de flores por lepid6pteros probablemente se debena a una combinaci6n de factors,
tales como los atrayentes, estimulantes y refrenamientos de alimentarse, estimulantes
visuales, morfologia de flores, y factors ecol6gicos los cuales controlan la disponibilidad
A toxic nectar hypothesis has been put forward that offers an explanation for pat-
terns of insect visitation at flowers (Baker & Baker 1975, Masters 1991). The hypothesis
proposes that many floral nectars contain toxic secondary plant chemicals that deter
Landolt & Lenczewski: Toxic Nectar Hypothesis 557
feeding by non-adapted flower visiting insects. Baker & Baker (1975) and Baker (1978)
discovered a widespread occurrence of alkaloids and phenolics in tree flowers. Both of
these classes of secondary plant chemicals are known to be toxic or distasteful to insects.
They suggested that these chemicals could function to protect plants against "nectar-
robbing" by inefficient pollinators. A similar idea was put forth by Janzen (1977) as an
explanation for why ants generally do not visit or raid flowers for nectar. Additional
examples of such chemicals in nectar were given by Deinzer et al. (1977). Rhoades &
Bergdahl (1981) expanded the idea to suggest that the exclusion of generalist flower
visitors should promote flower constancy of adapted species, thereby increasing pol-
Subsequently, two examples were reported of deterrence or inhibition of insect
feeding by nectar or nectar constituents. Stephenson (1981) demonstrated feeding de-
terrence in ants and in a butterfly species by Catalpa speciosa (Warder ex Barney)
nectar. Chemicals in Catalpa nectar evidently intoxicated these insects, thus limiting the
amount of nectar consumed. Masters (1991) concluded that monocrotaline, a pyrrolizidine
alkaloid found in several plant families, inhibits consumption of sugar solution by the
gulf fritillary butterfly, Agraulis vanilla (L.). His experiments suggested that this
alkaloid in nectar would reduce both consumption and visits by A. vanilla to flowers
with nectar containing monocrotaline. Masters' (1991) studies provide the first experi-
mental support for the contention that the widespread occurrence of alkaloids in floral
nectars functions to deter nectar feeding in Lepidoptera, as suggested by Baker &
We tested the hypothesis of Baker & Baker (1975), that alkaloids are widespread in
floral nectars to deter Lepidoptera flower visitors, using two noctuid moth species: the
cabbage looper, Trichoplusia ni (Hiibner) and the tobacco budworm, Heliothis virescens
(Fab.). The cabbage looper feeds as an adult at flowers of a number of species of plants,
including Aster spinosa Benth. (Shorey et al. 1962), Abelia grandiflora (Grant 1970),
Araujia sp. (Cantelo & Jacobson 1979) and Cestrum nocturnum (L.) (Heath et al. 1992).
The cabbage looper and the tobacco budworm both require sugar sources as adults for
maximum reproductive development (Lukefahr 1960, Lukefahr & Martin 1964). Mono-
crotaline was selected as a hypothetical toxic nectar feeding deterrent because it is a
pyrrolizidine alkaloid, is somewhat soluble in water, is readily available in a purified
form, and had already been implicated as a feeding deterrent in nectar (Masters 1991
using A. vanillae. We included A. vanilla in these studies to provide a direct compari-
son to our results with the noctuid moths.
We report here the negative results of a series of experiments testing for evidence
of deterrence of feeding on solutions of sucrose and monocrotaline by adults of T. ni,
H. virescens, and A. vanilla. A feeding deterrent, as defined by Nordlund (1981), is
a chemical that inhibits feeding in a place where an organism would, in its absence,
feed. We expected to find widespread deterrence of these flower-visiting Lepidoptera
by monocrotaline added to sucrose solutions, but found no evidence of deterrence.
MATERIALS AND METHODS
All T. ni and H. virescens used in feeding trials were reared in the laboratory using
the methods of Guy et al. (1985). Pupae were sorted by sex and only unmated female
moths were used. Adults were held in aluminum frame and plastic screen cages (30 x
30 x 30 cm) in environmental chambers at 24 1C and 60 + 10% RH under a reversed
14:10 (L:D) photoperiod with lights off at 0630 hours and on at 1630 hours (EST). Cages
of moths were supplied with a 50-ml cup of a 3 to 1 sucrose-honey solution on cotton as
food after emergence and an inverted jar filled with water was placed on paper toweling
on top of the cage. The sucrose-honey solution was removed 48 hours prior to feeding
558 Florida Entomologist 76(4) December, 1993
trials. Pupae were moved every day to provide adults of known ages. All moths used
were 5 days old. Feeding experiments on moths were conducted during the first 4 hours
of the 10-hour scotophase (dark) of the light cycle.
All A. vanilla used in laboratory feeding experiments were reared from field-col-
lected larvae or from eggs laid by adults netted in the field. Larvae were held within
screened cages (30 x 30 x 30 cm) in a greenhouse and supplied with Passiflora incar-
nata L. foliage until pupation. Adults were removed, separated by sex and then kept
in separate screen cages (30 x 30 x 30 cm) in the greenhouse. They were supplied with
a 50-ml cup of sucrose-honey solution on cotton after emergence as well as an inverted
water bottle on paper towels at the top of the cage. Sucrose-honey solutions were
removed 48 hours prior to feeding trials and all butterflies used were 3 to 7 days old.
For some greenhouse and field cage tests, A. vanilla were captured as adults, either
in the field by netting or in flight traps (Walker & Whitesell 1993). These individuals
were kept in the laboratory overnight in aluminum screened cages and provided with
both sucrose-honey solution and water.
A feeding platform was constructed to present captive insects with test solutions
and to measure consumption. This device was built from a 35 x 9 x 2 cm plywood base
with five 15 x 0.8 cm wooden dowels inserted at intervals of 7.5 cm across the length
of the board. Glass capillary tubes (200 pl micropipets, 1.4 mm i.d.) were bent to a
J-shape, graduated (cm, later converted to pi consumed) and attached to each dowel
with putty. Manostats were attached at the top of each microcapillary tube for manual
control in maintaining the fluid levels. Smaller dowels (2 x 0.5 cm) were fitted 6 cm in
front of the feeding tubes and clips were attached with a 1-cm length of wire through
a hole drilled in the dowel. Experimental insects were secured by holding the wings
with the clips and were positioned with mouthparts aligned to the shorter end of the
J-shaped capillary tube. Except for one "free-feeding" experiment with A. vanilla,
proboscids were inserted into the capillary and insects were positioned so withdrawal
of mouthparts was difficult. If the insect removed the proboscis, it was reinserted, and
in general remained in contact with the solution throughout the feeding trial.
Using this feeding platform, four feeding comparisons were made. Briefly, these
were: 1) a comparison of feeding rates of 20% sucrose over time (T. ni, H. virescens),
2) effects of concentrations of monocrotaline in 20% sucrose on consumption amounts
(T. ni, H. virescens, and A. vanillae, 3) comparison of consumption of 20% sucrose,
water, or monocrotaline in 20% sucrose when the insects were able to withdraw the
mouthparts (A. vanilla only), and 4) the same comparison of A. vanilla consumption
of 20% sucrose, water, or monocrotaline in 20% sucrose, but conducted in the field,
using netted butterflies.
The first experiment was conducted to establish a baseline of consumption of sucrose
solution over time, by both T. ni (N = 20 females) and H. virescens (N = 5 females).
These were exposed to 20% sucrose solution for 30 min, using the feeding platform, and
amounts consumed (1l) were measured at 2-min intervals. The results of this test were
used to establish time protocols for additional feeding tests.
The second experiment tested for effects of a range of concentrations of mono-
crotaline on consumption of sucrose solution. A series of monocrotaline concentrations
(0, 1, 10, 100, 1000 ig/pl) in aqueous 20% sucrose solutions were presented to T. ni for
30 min (10 replicates, N = 50 females), H. virescens (10 replicates, N = 50 females)
and A. vanilla (5 replicates each, N = 25 females, 25 males) for 30 min with amounts
consumed measured at 2-min intervals. For each species, data were analyzed for re-
lationships between monocrotaline concentration and consumption, using linear regres-
sion (Steel & Torrie 1960).
The third feeding experiment was set up to determine if A. vanilla consumption
of a monocrotaline-saturated 20% sucrose solution is reduced if individuals have the
ability to withdraw mouthparts from the feeding tubes (3 replicates, N = 5 males, 4
Landolt & Lenczewski: Toxic Nectar Hypothesis
females). Feeding tubes were filled with 1 pig/pl monocrotaline in 20% sucrose, 20%
sucrose, or deionized water. Proboscids were initially inserted into the capillary tube
and only reinserted once more if removed. Butterflies were positioned so that extraction
of mouthparts could be accomplished easily at will. The trials were terminated after 30
min. and total amounts consumed (Pl) were recorded. Mean consumption amounts were
analyzed for differences among treatments by Student's t-test.
The fourth experiment used the protocol from the third test, but was conducted in
the field using freshly captured A. vanilla to determine if field collected butterflies
(possibly with greater hunger and thirst) would give consumption results differing from
butterflies held in the laboratory that were provided water and sugar. Water, 20%
sucrose solution or 1 pig/pl monocrotaline in 20% sucrose were fed to butterflies (5
replicates, N = 13 females, 2 males) for 10 min with the total amount consumed (Pl)
recorded at the end of that time. Mean consumption amounts were analyzed for differ-
ences among treatments by Student's t-test.
Since Masters (1991) found that A. vanilla would alter their flower color preference
to avoid monocrotaline in artificial nectar, we constructed another feeding platform to
present butterflies with a choice of artificial flowers and test solutions. This was similar
to that used by Masters (1991) and consisted of a 60 x 60 cm plywood base with 49
evenly spaced dowels 15 cm high and 0.8 cm wide. Each dowel was fitted with a 200 p.1
microcapillary tube closed at the bottom with aquarium sealant and protruding 3 cm
above the top of the dowel. Artificial flowers (3 cm diam) were cut from plastic snap
cap vials and spray-painted with either Arc Yellow # 28-52-016-0, Day-Glo Color Corp.,
Cleveland, Ohio (= orange flowers) or Sun Yellow # 20003, Color Creations, Wal-Mart
Stores Inc., Bentonville, Ark., 72716 (= yellow flowers). These flowers (25 orange, 24
yellow) were positioned flush with the top of the capillary tubes and arranged on the
board in alternating colors.
An experiment was then conducted, using this second feeding platform, placed in a
large flight cage (1.8 x 1.1 x 3 m) in a greenhouse. The purpose of the experiment was
to determine if A. vanilla adults could be induced to shift from a preferred flower color
to a non-preferred color by monocrotaline in the sucrose solution provided with the
artificial flowers, as indicated by Masters (1991). Butterflies were placed in the flight
cage immediately prior to the experiment. The first feeding test using this platform
determined color preference. It involved providing 20% sucrose (with no monocrotaline)
in the capillary tubes of all the flowers, both yellow and orange. Released butterflies
remained in the cage to discover artificial flowers and feed for 4 hours. Total amounts
consumed (.l) from each capillary tube were measured at the end of the four hours.
Color preference (orange versus yellow) was determined from these data. All tests were
run between 1000-1500 hours. Temperature and light intensity were noted at the begin-
ning and end of each set.
On the following day the same butterflies were offered 1 ig/pl monocrotaline in 20%
sucrose in flowers of the preferred color (orange), and 20% sucrose with no monoc-
rotaline in flowers of the nonpreferred color (yellow). These adults were allowed to feed
in the same way for 4 hours and total amounts consumed from the capillary tubes of
each flower were measured at the end of that period. The experiment was conducted 3
times, using 17, 36 and 32 butterflies. All had access to sugar water up to introduction
in the test cages. Mean amounts consumed from the different colored flowers were
analyzed by Student's t-test for significant differences between treatments.
The first experiment determined the patterns of consumption of sugar water by T.
ni and H. virescens over time. The amounts consumed by T. ni at 2-min intervals for
the 30-min test period are shown in Fig. 1. Feeding was strongest within the first 2
Florida Entomologist 76(4)
TBW I CL
2 6 10 14 18 22 26 30
Fig. 1. Mean ( SE) amounts consumed in 30 minutes of 20% sucrose solutions with
different concentrations of monocrotaline by tobacco budworm (TBW) and cabbage
looper (CL) moths and by gulf fritillary butterflies (GF).
min with 9.6 l consumed per min. After only 6 min, feeding was reduced to half that
rate and declined steadily to less than 2 pl consumed per min after 30 min. H. virescens
consumed less 20% sucrose solution than cabbage looper moths (Fig. 1). Most of the
feeding still occurred in the first 6 min and then rapidly declined to nearly no feeding
at 30 min. A 30-min time period was selected for subsequent feeding trials.
In comparisons of consumption by T. ni of 20% sucrose with varying concentrations
of monocrotaline (Fig. 2), there were no significant effects of monocrotaline on the
amount of sucrose solution consumed, at any of the monocrotaline concentrations tested
(r2=0.05, p=0.87). Adult females consumed on average between 72-84 pl of whichever
treatment they were offered within the 30 minute test period.
Heliothis virescens consumption of variable monocrotaline concentrations in 20%
sucrose solution (Fig. 2) was much less than that for the cabbage looper, but as with
the cabbage looper, there were no significant differences in consumption with increased
monocrotaline concentration, with adults consuming between 8-50 pl of solution in any
case (r2=0.62, p=0.54).
For A. vanilla offered concentrations of monocrotaline in 20% sucrose solutions,
there was also no significant regression of amounts consumed versus concentration of
monocrotaline (rW=0.09, p=0.83) (Fig. 2). Adult butterflies consumed approximately
115 (85 to 135) t1 of any of the test solutions in 30 min.
All insects tested for consumption of monocrotaline concentrations in 20% sucrose
solutions were held for 24 h after testing for evidence of toxic effects. No evidence of
Landolt & Lenczewski: Toxic Nectar Hypothesis
0 1 10
MONOCROTALINE (NG PER UL)
Fig. 2. Mean ( SE) consumption rates of a 20% sucrose solution by adult female
cabbage looper (CL) and tobacco budworm (TBW) moths for each minute over a 30
toxicity was observed and no moths or butterflies died during the 24-h period following
consumption of monocrotaline solutions.
In all of the feeding tests conducted up to this point, insect proboscids were rein-
serted when removed and voluntary withdrawal of mouthparts from the capillary tubes
by the insects appeared to be difficult. In a test of A. vanilla consumption of water,
sucrose and monocrotaline-sucrose solutions, with the proboscis free (Fig. 3), the but-
terflies consumed significantly less water than sucrose solution (t = 4.3, df= 4, p = 0.01),
but the average amounts of monocrotaline-sucrose solution imbibed were not signifi-
cantly different from the amounts of water consumed (t= 1.6, df=4, p=0.2) or sucrose
solution (t=1.3, df=4, p=0.3) consumed. When gulf fritillaries were tested in this
manner immediately after being captured in the field (Fig. 3), they drank less water
than either the sucrose (t=3.7, df=8, p=0.005) or monocrotaline-sucrose (t=3.2, df=8,
p=0.01) solutions. However, there was no significant difference between the amounts
of sucrose or monocrotaline-sucrose solutions consumed (t= 0.4, df=8, p= 0.7).
The final experiment allowed butterflies freedom in choice of flower color and amount
of sucrose or monocrotaline consumed. The first trial was run with 20% sucrose in both
yellow and orange flowers and the results are shown in Fig. 4. The amounts of sucrose
solution consumed at orange flowers were greater than the amounts of sucrose solution
consumed at yellow flowers (t=4.0, df=69, ps0.001). When the capillary tubes at
Florida Entomologist 76(4)
W S S&M
W S S&M
Fig. 3. TOP: Mean ( SE) amounts consumed by gulf fritillary butterflies of water
(W), a 20% sucrose solution (S), and a 20% sucrose solution with monocrotaline at 1
pg/Ril (S & M), when provided freedom to remove the proboscis. BOTTOM: Mean (
SE) amounts consumed by gulf fritillary butterflies of water (W), a 20% sucrose solution,
and a 20% sucrose solution with monocrotaline at 1 Jg/pl (S & M), in the field and using
field collected insects.
Landolt & Lenczewski: Toxic Nectar Hypothesis
Fig. 4. Mean ( SE) amounts of 20% sucrose solutions consumed by gulf fritillary
butterflies at orange or yellow artificial flowers, when both colors contained sucrose
with no monocrotaline (S) or when capillary tubes at orange (preferred) flowers con-
tained monocrotaline (1 Rg/1l) in the sucrose solution.
orange artificial flowers were filled with 20% sucrose and monocrotaline at 1 Rg/pl, the
butterflies still selected the orange flowers over yellow and consumed significantly more
of the monocrotaline-sucrose solution at orange flowers (t=4.1, df=44, p 0.001) (Fig.
4). Under these test conditions, there was no indication that the butterflies switched to
yellow (non-preferred color) from orange (preferred color) flowers.
Baker & Baker (1975) proposed that alkaloids in floral nectar are maintained to
discourage Lepidoptera visitors. All of the plants found by Baker & Baker (1975) to
contain alkaloids are usually bee pollinated whereas they found no alkaloids in any of
the 46 species tested which are pollinated by Lepidoptera. Masters (1991) reported
inhibition of feeding on sugar solutions with the alkaloid monocrotaline by the gulf
fritillary butterfly, providing evidence supporting this toxic nectar hypothesis of Baker
& Baker (1975). The alkaloids reported by Baker & Baker (1975) however, were not
quantified or identified and it is not known if monocrotaline occurs in nectars of flowers
visited by Lepidoptera.
Florida Entomologist 76(4)
Our results with the cabbage looper and tobacco budworm moths, and the gulf
fritillary butterfly, indicate that monocrotaline in nectar did not deter these lepidopter-
an flower visitors from feeding. We found no evidence of reduced consumption of
sucrose solutions with the addition of monocrotaline, even at a near-saturation concen-
tration, in either of the two moths tested, or in the butterfly. Since field-collected
butterflies were not deterred from feeding at monocrotaline-treated sucrose solutions,
we assume it was not an artifact of our insect handling procedures in the laboratory.
Individuals collected in the field were not less hungry than individuals in the laboratory,
and were not more discriminating. We expected these Lepidoptera species to be deter-
red from feeding on sugar solution containing this alkaloid, or to become intoxicated by
the monocrotaline dosage consumed in our artificial nectar. This is based on the findings
of alkaloids in nectar by Baker & Baker (1975), on their proposed toxic nectar
hypothesis, and on the reported deterrence of nectar feeding by the butterfly A. vanil-
lae by monocrotaline (Masters 1991). We found no evidence of any overt behavioral
deterrence, and we observed no signs of intoxication or acute toxicity as a result of
There are several possible explanations for the lack of expected deterrence observed
in our tests. The cabbage looper and tobacco budworm moths may be preadapted to
tolerate secondary plant chemicals such as monocrotaline, since larvae of these two
species can feed on plants containing alkaloids, such as Nicotiana tabaccum (tobacco)
(Sutherland & Greene 1984). The inhibitory effects of monocrotaline on feeding seen by
Masters (1991) in the gulf fritillary butterfly may be subtle and limited to longer term
learning based on experience and repeated exposure and choices. Perhaps there are
non-lethal toxic effects permitting them to avoid subsequent exposure through a process
of food aversion learning, as demonstrated for Lepidoptera larvae by Dethier (1980).
Despite the lack of evidence in our studies for any deterrence or inhibition of sugar
feeding in these three species of Lepidoptera by monocrotaline, certainly some plants
do provide nectars with toxic compounds that negatively affect non-adapted flower
visitors (e.g., Catalpa, Stephenson 1981). Such a mechanism could involve other types
of plant compounds, may require blends of secondary compounds acting together, or,
as discussed for A. vanilla, may be a result of a learning process by insects. It is
unfortunate that we do not know what alkaloids or phenolic compounds occur in the
plant nectars analyzed by Baker & Baker (1975) or in what concentrations.
It is also likely that multiple factors are involved in a plant's ability to encourage
and/or discourage insects from visiting its flowers and transporting its pollen. Flower
color, patterns of light reflectance and scent are important plant characteristics for
discriminating flower visitors. Morphology of the flower parts (corolla length, size and
shape of landing platform, defensive structures surrounding nectaries) are also physical
barriers to undesirable guests. Diurnal and seasonal patterns in flowering phenology
may be timed to pollinator activity patterns (e.g., T. ni, Heath et al. 1992). Also, there
is considerable evidence that secondary plant compounds function to attract specialized
pollinators, much as they do specialized herbivores. Faegri & Van der Pijl (1979) found
that floral constancy has been enhanced in orchids by secondary compounds that attract
specialist bees. It has also been demonstrated (Pliske 1975, BopprB 1986) that pyr-
rolizidine alkaloids attract certain ithomiine and danaiine butterflies. Also, secondary
chemicals in nectar may serve to attract or stimulate specially adapted pollinators rather
than functioning to deter generalist flower visitors.
Flower constancy (the limitation of a given flower visiting insect to a single or a
small number of plant species) is likely promoted or enforced by a combination of floral
attraction and recognition factors, which are important in eliciting positive responses
by potential pollinators, and additional physical and chemical factors that limit or deter
nectar robbing by other insect species. In the three species of Lepidoptera studied here,
Landolt & Lenczewski: Toxic Nectar Hypothesis
there was no evidence of feeding deterrence by near saturation concentrations of the
alkaloid monocrotaline in sugar water. These results indicate that it is unlikely that this
compound, if found in flower nectar, could play a major role in limiting nectar consump-
tion and flower visitation by generalist nectivores.
We thank K. Davis-Hernandez, R. Schneider, A. Solares and B. Dueben, USDA/
ARS, Gainesville, FL for technical assistance. Helpful suggestions to improve the man-
uscript were made by D. Robacker, S. Mayer, and H. Oberlander. This material is
based partly on work supported by the Cooperative State Research Service, U.S. De-
partment of Agriculture, under agreement No. 90-37250-5356.
BAKER, H. G., AND BAKER, I. 1975. Studies of nectar-constitution and pollinator-
plant coevolution, pp. 100-140 in L. E. Gilbert and P. H. Raven, [eds.] Coevolu-
tion of Plants and Animals. Univ. of Texas Press, Austin, Texas.
BAKER, H. G. 1978. Chemical aspects of the pollination of woody plants in the tropics,
pp. 57-82 in P. B. Tomlinson and M. Zimmerman, [eds.] Tropical trees as living
systems. Cambridge Univ. Press, New York.
BOPPRe, M. 1986. Insects pharmacophagously utilizing defensive plant chemicals (pyr-
rolizidine alkaloids). Naturwissenschaften 73: 17-26.
CANTELO, W. W., AND M. JACOBSON. 1979. Phenylacetaldehyde attracts moths to
bladder flower and to blacklight traps. Environ. Entomol. 8: 444-447.
DEINZER, M. L., P. A. THOMSON, D. M. BURGETT AND D. L. ISAACSON. 1977.
Pyrrollizidine alkaloids: their occurrence in honey from tansy ragwort (Senecio
jacobaea L.). Science 195: 497-499.
DETHIER, V. G. 1980. Food-aversion learning in two polyphagous caterpillars, Diac-
risia virginica and Estigmene congrua. Physiol. Entomol. 5: 321-325.
FAEGRI, F., AND I. VAN DER PIJL. 1979. The Principles of Pollination Ecology. 3rd
ed. Pergamon Press, New York.
GRANT, G. G. 1970. Activity of adult cabbage loopers on flowers with strong olfactory
stimuli. J. Econ. Entomol. 64: 315-316.
GUY, R. H., N. C. LEPPLA, J. R. RYE, C. W. GREEN, S. L. BARRETTE, AND K. A.
HOLLIEN. 1985. Trichoplusia ni, pp. 487-494 in P. Singh and R. F. Moore,
[eds.] Handbook of Insect Rearing, vol. 2. Elsevier, Amsterdam.
HEATH, R. R., P. J. LANDOLT, B. DUEBEN, AND B. LENCZEWSKI. 1992. Identifica-
tion of floral compounds of night-blooming jessamine attractive to cabbage looper
moths. Environ. Entomol. 21: 854-859.
JANZEN, D. H. 1977. Why don't ants visit flowers. Biotropica 9: 252.
LUKEFAHR, M. J. 1960. Effects of nectariless cottons on populations of three lepidop-
terous insects. J. Econ. Entomol. 53: 242-244.
LUKEFAHR, M. J., AND D. F. MARTIN. 1964. The effects of various larval and adult
diets on the fecundity and longevity of the bollworm, tobacco budworm, and
cotton leafworm. J. Econ. Entomol. 57: 233-235.
MASTERS, A. 1991. Dual role of pyrrolizidine alkaloids in nectar. J. Chem. Ecol. 17:
MAY, P. G. 1986. Foraging selectivity in adult butterflies: Morphological, ecological,
and physiological factors affecting flower choice. Ph.D. dissertation, Univ.
Florida, Gainesville, Florida.
NORDLUND, D. A. 1981. Semiochemicals: A review of the terminology, pp. 3-12 in D.
A. Nordlund, R. L. Jones, and W. J. Lewis, Semiochemicals, their role in pest
control. Wiley and Sons, Inc., New York.
PLISKE, T. E. 1975. Attraction of Lepidoptera to plants containing pyrolizidine al-
kaloids. Environ. Entomol. 4: 455-473.
Florida Entomologist 76(4)
RHOADES, D. F., AND J. C. BERGDAHL. 1981. Adaptive significance of toxic nectar.
American Naturalist 117: 798-803.
SHOREY, H. H., L. A. ANDRES, AND R. L. HALE JR. 1962. The biology of Trichop-
lusia ni (Lepidoptera: Noctuidae). I. Life history and behavior. Ann. Entomol.
Soc. Am. 55: 591-597.
STEEL, R. G. D., AND J. H. TORRIE. 1960. Principles and procedures of statistics.
McGraw-Hill, New York.
STEPHENSON, A. G. 1981. Toxic nectar deters thieves of Catalpa speciosa. American
Mid. Nat. 105: 381-383.
SUTHERLAND, D. W. S., AND G. L. GREENE. 1984. Cultivated and wild host plants,
pp. 1-13 in P. D. Lindgren and G. L. Greene, [eds.] Suppression and Management
of Cabbage Looper Populations. U.S.D.A. Tech. Bull. 1684.
WALKER, T. J., AND J. J. WHITESELL. 1993. A superior trap for migrating butter-
flies. J. Lepidoptera Soc. 47: 140-149.
TOUMEYELLA LIGNUMVITAE, A NEW SPECIES
OF SCALE INSECT FROM THE FLORIDA KEYS
MICHAEL L. WILLIAMS
Department of Entomology and
Alabama Agricultural Experiment Station
Auburn University, AL 36849-5413, U.S.A.
Toumeyella lignumvitae, a new species of soft scale insect known only from lignum-
vitae, Guaiacum sanctum Linnaeus, trees in the Florida Keys, is described and illus-
trated. Lignumvitae scale feeding is seriously affecting growth of lignum-vitae trees on
Lignumvitae Key. Three species of insects are often found associated with lignumvitae
scale. Two species of ants, the Florida carpenter ant, Camponotus abdominalisfloridanus
(Buckley) and the little fire ant, Wasmannia auropunctata (Roger), tend the scale
insects for the honeydew they produce, and a predaceous pyralid caterpillar, Laetilia
coccidivora (Comstock), is occasionally found feeding on colonies of lignumvitae scale.
Se describe y se ilustra Toumeyella lignumvitae, una nueva especie de c6ccido
conocido solamente de arboles de guayacAn blanco, Guaiacum sanctum Linnaeus, en
los cayos de la Florida. El c6ccido de guayacAn blanco esta afectando severamente el
crecimiento de guayacAn blanco en el Cayo Lignumvitae. Se encuentran frequentamente
tres species de insects asociadas al c6ccido de guayacan blanco. Dos species de hor-
migas, incluyendo la hormiga carpenter de la Florida, Camponotus abdominalis
floridanus (Buckley) y la hormiga brava pequefia, Wasmannia auropunctata (Roger),
atienden los c6ccidos para el melado lo cual produce, y se encuentra ocasionalmente
una larva de un piralido, Laetilia coccidivora (Comstock), la cual se alimenta de las
colonies del c6ccido de guayacan blanco.
Williams: Lignumvitae Scale
The soft scale insect genus Toumeyella Cockerell consists of 11 described species,
eight of which occur in North America (Sheffer & Williams, 1990). Five species are
reported in Florida (Hamon & Williams 1984, Williams & Kosztarab 1972).
Most species of Toumeyella are of little economic importance but four are known to
cause extensive damage to their host plants. The tuliptree scale, T. liriodendri (Gmelin),
is reported by Burns & Donley (1970) to be a severe pest of yellow poplar, Li-iodendron
tulipifera L., throughout its range, and Merrill & Chaffin (1923) report it as a serious
pest of landscape plantings of banana shrub, Michelia figo (Lour.) K. Spreng., and
several species of Magnolia. The other three pest species feed on trees in the genus
Pinus. Two of these, the pine tortoise scale, T. parvicornis (Cockerell), and the striped
pine scale, T. pini (King), have become important pests in loblolly pine seed orchards
in the southern United States. Outbreaks of these pests have been associated with the
aerial application of pyrethroid insecticides for seed and cone insect control (DeBarr et
al. 1982). In California, the irregular pine scale, T. pinicola Ferris, is considered the
most serious soft scale pest of pines (Gill 1988).
The new species described here threatens the only remaining North American stand
of the subtropical lignum-vitae tree, Guaiacum sanctum L. (Zygophyllaceae). There
has been concern for the future of these trees on Florida's Lignumvitae Key Botanical
Site since a report by Dr. Tom Eisner of Cornell University in 1972 indicated the stand
was declining (personal communication). Several researchers associated with the Uni-
versity of Florida's Tropical Research and Education Center in Homestead, Florida,
studied the biology of this scale insect and monitored the effects of the infestation on
trees at the site since its discovery. Schaffer & Mason (1990) found that shading and
herbivory by this scale insect each resulted in decreased growth of lignum-vitae and
that herbivory by the scale resulted in a greater decrease in tree growth for sun-grown
than for shade-grown trees. Examination of this soft scale pest revealed that it is an
undescribed species of Toumeyella. The description of this new species, designated as
Toumeyella lignumvitae Williams, is presented below. Measurements are expressed in
microns and presented as the range seen in a series of 10 specimens. The key to species
of Toumeyella in Florida by Hamon and Williams (1984) is modified to accommodate
the new species.
Toumeyella lignumvitae Williams, New Species
(Fig. 1, 2)
Mature females convex to hemispherical, oval or irregular in outline when crowded
on twigs. Color varies from tan to grey, mottled with black markings (Fig. 2), oviposit-
ing females often exhibit a pinkish cast due to the light red eggs in their body.
Body of slide-mounted female (Fig. 1A) subcircular or oval in outline, 1584-3641
long, 1401-3185 wide; anal cleft 211-882 long, usually about 1/5 the body length. Dorsal
derm only lightly sclerotized.
Dorsum and Margin: Marginal setae (Fig. 1B) slender, pointed, curved or straight,
24-34 long, arranged in an irregular single row around margin, distributed as follows:
18-24 around head between anterior spiracular setae; 6-10 between anterior and posterior
spiracular setae on each side of body, and 18-24 on each side of abdomen posterior to
spiracular setae. Body setae (Fig. 1C) straight, short and stout, tapering to a point,
6-12 long, scattered over dorsum, larger and more numerous in anal area and along
Florida Entomologist 76(4)
Fig. 1. Toumeyella lignumvitae Williams, adult female.
middorsal line. Spiracular setae (Fig. 1D) in groups of 3, median seta 2-3 times longer
than laterals; median seta slender, curved, tapering to a point, 41-65 long, 7-9 wide;
lateral setae short and stout, may be pointed or irregular in shape, 14-30 long, 8-12
wide. Dorsal pores of 3 types: bilocular pores with a long, slender filament (Fig. 1E)
3.4-3.9 in diameter, and simple disc pores (Fig. 1F) 2.0-3.9 in diameter, scattered over
dorsum; flat to slightly convex discoidal pores (Fig. 1G) 8-10 in diameter, concentrated
anterior to anal plates in a middorsal band running up to area between posterior spira-
cles; clustered in area anterior to anal plates.
Williams: Lignumvitae Scale 569
Anal Plates (Fig. 1H1, 1H2): Each triangular in shape, 184-209 long, 91-128 wide;
cephalo-lateral margin nearly straight, 109-168 long, caudo-lateral margin convex, 138-
148 long. Each plate with 4 apical and 5 or 6 subapical setae. Anal fold with a pair of
large fringe setae and 7-9 hypopygial setae. Anal ring (Fig. 1I) with 8-10 setae, usually
10, and an irregular pattern of translucent pores.
Venter: Ventral submarginal setae slender, straight, pointed, 9.9-13.8 long, arranged
in a single row on submargin, numbering 20-36 around head anterior to spiracular setae,
8-12 on each side between spiracular setae, and 23-30 on each side of abdomen posterior
to spiracular setae. Body setae (Fig. 1J), similar to ventral submarginal setae, 8.9-11.8
long, numerous, scattered over venter. With 3 pair of prevulvar setae, and 2 pair of
interantennal setae. Antennae normally 6-segmented (Fig. 1K) although segmentation
often obscure, occasionally third and fourth segments fused, resulting in a 5-segmented
antenna, 73-108 long, scape 34-53 wide. Legs greatly reduced, 5-segmented but segmen-
tation indistinct, total leg lengths as follows: prothoracic leg 59-73 long, 39-55 wide at
base; mesothoracic leg 63-79 long, 39-59 wide; metathoracic leg 59-102 long, 45-73 wide.
Tarsus and claw each with a pair of slender knobbed digitules, claw simple. Spiracles
87-148 long, atrium 59-99 wide. Spiracular pore band 2-4 pores wide, pores mostly
quinquelocular, but range from 3-6 locules (Fig. 1L); anterior pore band with 39-53
pores, posterior pore band with 48-80 pores. With 3-6 locular pores in spiracular fur-
rows, anal area, and in a band running from anal area up to posterior spiracle. Tubular
ducts with slender filament (Fig. 1N) confined to anal area, duct 14-26 long, 3-5 wide.
Microducts (Fig. 1M) with openings 2-4 in diameter, scattered over venter, but more
numerous in marginal 1/3 of venter.
MATERIAL STUDIED. All collected in Florida on Guaiacum sanctum L. as follows:
Key Largo 15 slides (15 specimems), 13 March 1988, M. Williams and L. Mason;
Lignumvitae Key 8(12), 18 July 1971, T. Eisner; 1(1), July 1971, R. Baranowski; 8(8),
4 August 1971, R. Baranowski; 36(36), 15 August 1987, M. Williams; 15(15), 10 March
1988, M. Williams and L. Mason. Plantation Key 2(2), 6 September 1974, S.
Goldweber. Totten Key 1(1), 21 November 1986, C. Lippencott.
TYPE DESIGNATION AND DEPOSITORY. Adult female holotype (1 specimen on
slide). Right label "AL-045-87g/FL: Lignumvitae Key/15 August 1987/M. L. Williams
Coll./Balsam". Left label "AL-045-87g/Toumeyella lignumvitae Williams/HOLOTYPE/
e: Guaiacum sanctum". Deposited in USNM Coccoidea Collection, Beltsville, Mary-
land, USA. In addition, there are 89 paratypes listed in the "Material Studied" section.
Representative paratype specimens deposited in USNM Coccoidea Collection, Auburn
University Coccoidea Collection, Auburn, Alabama and Florida State Collection of Ar-
thropods, Gainesville, Florida.
DISTRIBUTION. Florida Key Largo, Lignumvitae Key, Plantation Key, and Totten
HOST PLANT. The lignum-vitae tree, Guaiacum sanctum L., is the only known host.
ETYMOLOGY. The specific epithet, lignumvitae, refers to the common name of its
only known host, the lignum-vitae tree, and its type locality, Lignumvitae Key, Florida.
Suggested common name: lignumvitae scale.
Hamon and Williams (1984) provided a key to the Toumeyella of Florida which
included 5 species. That key may be modified as follows to accommodate the new species
3. Anal plates each with 4 dorsal setae; discoidal pores extending anteriorly
to above rostrum; on non-coniferous hosts ........................................... 3a
570 Florida Entomologist 76(4) December, 1993
3'. Anal plates each with 4-25 dorsal setae; discoidal pores confined to post-
erior half of body; on coniferous hosts .................................... .......... 4
3a. Spiracular setae subequal in length; marginal setae stout, straight, taper-
ing to a point ....................................................................... liriodendri
3a'. Spiracular setae with median seta 2x longer than lateral setae; marginal
setae slender, straight or curved, pointed .............................. lignumvitae
4. Dorsal discoidal pores conical in cross section; anal plates each with 4 dorsal
setae; marginal setae slender, curved ............................................. pini
4'. Dorsal discoidal pores flat or only slightly convex in cross section; anal plates
each with 15-25 dorsal setae; marginal seta spinelike ................ virginiania
Toumeyella lignumvitae most closely resembles the tuliptree scale, T. liriodendri.
Both species are found on non-coniferous hosts with tuliptree scale most commonly
occurring on yellow popular and various Magnolia spp., while lignumvitae scale is
currently known only from the lignum-vitae tree. The species are easily distinguished
morphologically utilizing the characters presented in the key.
Three insect species are often found associated with lignumvitae scale: Two species
of ants, the Florida carpenter ant, Camponotus abdominalis floridanus (Buckley) (Fig.
2) and the little fire ant, Wasmannia auropunctata (Roger), tend the scale insects for
the honeydew they produce. One or the other of these two species was always found in
association with infestations of lignumvitae scale, and their activity on infested trees
Fig. 2. Adult female lignumvitae scale, Toumeyella lignumvitae Williams, tended
by the Florida carpenter ant, Camponotus abdominalis floridanus (Buckley).
Williams: Lignumvitae Scale 571
Fig. 3. Predaceous larva of a Pyralid moth, Laetilia coccidivora (Comstock), feed-
ing on lignumvitae scale, Toumeyella lignumvitae Williams.
makes the job of locating scale populations much easier. The third insect is a predaceous
pyralid caterpillar, Laetilia coccidivora (Comstock), which is occasionally found feeding
on colonies of the scale insect (Fig. 3). This predaceous caterpillar has a chemical defense
which enables it to repel the tending ants which normally drive off natural enemies
attempting to attack the scale insect (Eisner et al. 1972). This pyralid larva constructs
a shelter of silk webbing over itself and the scale insects which protects it as it feeds.
In feeding, the pyralid larva consumes all but the more sclerotized dorsal derm of its
prey, leaving only the hollow exoskeleton (shell) of the scale insect behind or incorpo-
rated in its web-like shelter.
The lignumvitae scale causes chlorosis, loss of foliage, dieback and sometimes death
of individual trees if left unchecked. The threat of the scale to this native stand of
lignumvitae trees is critical. Unless a means of control or management is found soon,
Lignumvitae Key Botanical Site may lose most of the trees for which it was named.
I would like to thank Thomas Eisner for first calling my attention to the species andl
for the use of photographs shown in Figures 2 and 3; Mark Deyrup for verifying my
ant identifications; James Stevenson for permitting me to collect in the Lignumvitae
State Botanical Site, and Richard Baranowski, Linda Mason and Jeanie Parks for ar-
ranging collecting trips to Lignumvitae Key and providing biological and ecological
information on the species during the course of this study. This manuscript was ap-
proved for publication as Alabama Agricultural Experiment Station Journal Series No.
572 Florida Entomologist 76(4) December, 1993
BURNS, D. P., AND D. E. DONLEY. 1970. Biology of the tuliptree scale, Toumeyella
liriodendri (Homoptera: Coccidae). Ann. Entomol. Soc. Amer. 63: 228-235.
DEBARR, G. L., L. R. BARBER, M. S. McCLURE, ANDJ. C. NORD. 1982. An assess-
ment of the problem of scale and mealybug outbreaks associated with the use of
insecticides in southern pine seed orchards. Unpublished report on file at the
USDA Forest Service, Forestry Sciences Laboratory, Athens, GA. 57 pp.
EISNER, T., P. JUTRO, D. J. ANESHANSLEY, AND R. NIEDHAUK. 1972. Defense
against ants in a caterpillar that feeds on ant-guarded scale insects. Ann. En-
tomol. Soc. Amer. 65: 987-988.
GILL, R. 1988. The scale insects of California. Part 1: The soft scales (Homoptera:
Coccoidea: Coccidae). California Dept. Food and Agr. Tech. Ser. in Agr. Biosys-
tematics and Plant Path. No. 1. 132 pp.
HAMON, A. B., AND M. L. WILLIAMS. 1984. The soft scale insects of Florida (Homop-
tera: Coccoidea: Coccidae). Arthropods of Florida and Neighboring Land Areas.
Florida Dept. Agric. and Consumer Serv. Div. Plant Ind., Vol. 11, 194 pp.
MERRILL, G. B., ANDJ. CHAFFIN. 1923. Scale insects of Florida. Quart. Bull. Florida
State Plant Bd. 7: 177-298.
SCHAFFER, B., AND L. J. MASON. 1990. Effects of scale insect herbivory and shading
on net gas exchange and growth of a subtropical tree species (Guaiacum sanctum
L.). Oecologia 84: 468-473.
SHEFFER, B. J., AND M. L. WILLIAMS. 1990. Descriptions, distribution, and host-
plant records of eight first instars in the genus Toumeyella (Homoptera: Coc-
cidae). Proc. Entomol. Soc. Washington 92: 44-57.
WILLIAMS, M. L., AND M. KOSZTARAB. 1972. Morphology and systematics of the
Coccidae of Virginia, with notes on their biology (Homoptera: Coccoidea). Vir-
ginia Polytech. Inst. and State Univ. Res. Div. Bull. 52. 215 pp.
SUITABILITY OF SIX SPECIES OF COMMELINACEAE
AS LARVAL HOSTS OF MOURALIA TINCTOIDES
(LEPIDOPTERA: NOCTUIDAE, PLUSIINAE)
PETER J. LANDOLT
United States Department of Agriculture
Agriculture Research Service
P. 0. Box 14565, Gainesville, FL 32604
Survival rates to pupal and adult stages, development rates, and pupal weights
were determined for larvae of Mouralia tinctoides (Guen&e) placed on six species of
Commelinaceae. Larvae developed successfully on the dayflower, Commelina diffusa
Burmeister, on two wandering jew species, Tradescantia zebrina G. G. Bosse and
Tradescantia fluminensis Velloso, and on the spiderwort, Tradescantia ohiensis Raf-
finesque. Poor survival and lower pupal weights occurred with larvae placed on foliage
of purple queen Tradescantia pallida (Rose) D. Hunt and on Callisia repens L. Larvae
of M. tinctoides did not survive on two artificial diets used as rearing media for many
pest Lepidoptera or on foliage of cotton, Gossypium hirsutum L.
Key Words: Insecta, host plant, spiderwort, rearing, looper, Mouralia, Tradescantia.
Landolt: Host Plants for Mouralia 573
Se determinaron las tasas de supervivencia de las etapas pupales y imaginales, tasas
de desarrollo, y pesos pupales para larvas de Mouralia tinctoides (Guenee) colocadas
sobre seis species de Commelinaceae. Las larvas se lograron de desarrollar con 6xito
sobre canutillo, Commelina diffusa Burmeister, y sobre Tradescantia zebrina G. C.
Bosse y T. fluminensis Velloso y sobre T. ohiensis Raffinesque. Supervivencia deficiente
y peso pupal mas bajo ocurri6 en las larvas colocadas sobre follaje de T. pallida (Rose)
D. Hunt y sobre canutillo rastrero, Callisia repens L. Las larvas de M. tinctoides no
sobrevivieron ni sobre d6s dietas artificiales usadas como medios de criar para muchos
lepid6pteros plagas, ni sobre el follaje de algod6n, Gosypium hirsutum L.
Mouralia tinctoides (GuenBe) is a noctuid moth (subfamily Plusiinae) and is related
to several important pest insects, most notably in the United States, the cabbage looper,
Trichoplusia ni (Htibner), and the soybean looper, Pseudoplusia includes (Walker).
It has not been reported as a plant pest and is not known to occur on cultivated crop
plants. It is widely distributed in the Neotropics, and in the United States it occurs in
Florida, Texas and California (Eichlin & Cunningham 1978).
As part of a comparative study of sex pheromone chemistry of moths in this subfam-
ily, we attempted to rear larvae of M. tinctoides in the laboratory to provide suitable
numbers of insects for experimental purposes. Initially, attempts were made to use
published rearing methods based on a pinto bean diet that is widely used for a number
of pest species of Noctuidae (Shorey & Hale 1965, Guy et al. 1985). Because we were
unsuccessful in rearing M. tinctoides larvae on these artificial media, subsequent rearing
was accomplished using foliage of host plant species.
The only published records of host plants for M. tinctoides are two species in the
Commelinaceae. These are Tradescantia fluminensis Velloso (Comstock 1938, Crumb
1956, Eichlin & Cunningham 1978), and Tradescantia zebrina G. G. Bosse (=Zebrina
pendula of Eichlin & Cunningham 1978). Comstock (1938) found eggs and larvae on T.
fluminensis plants and documented complete development of M. tinctoides on the
This paper reports the results of comparative assessments of different species of
Commelinaceae as larval hosts of M. tinctoides and reports the rearing of multiple
generations of this moth in the laboratory on host plants.
MATERIALS AND METHODS
A colony of M. tinctoides was started with eggs obtained from mated females col-
lected in a walk-in UV light trap in Gainesville, Florida. Moths collected were placed
in a 18 x 13 x 10 cm clear plastic box with foliage of wandering jew, T. zebrina. Larvae
were maintained on foliage in large plastic storage boxes (45 x 30 x 20 cm) with
screened lids. A 50:50 mixture of ground corn cob and pine shavings was placed in the
bottom of the boxes to absorb excess moisture from chewed plant foliage and to serve
as a pupation medium. Fresh foliage of T. fluminensis and T. zebrina were supplied as
needed until larvae pupated. Larvae and foliage were transferred to new boxes when
frass covered the pine shavings and corn cob. Pupating larvae made cocoons of loose
shavings and leaves. Pupae were placed in a large screened cage (60 x 60 x 60 cm)
containing a potted plant (about 30 cm diam) of T. zebrina for mating and oviposition
by emerging moths. Females laid eggs on plant foliage which were then transferred with
a small paint brush either for experiments or for continued rearing. During this study,
5 generations of M. tinctoides were reared in the laboratory from eggs initially obtained
Florida Entomologist 76(4)
from mated females collected in the light trap. Rearing and experiments were conducted
in a room with fluorescent lighting on a 12:12 (L:D) light cycle, at 23 C, and 50 5% RH.
Rearing Attempts Using Synthetic Diet
Two attempts were made to rear larvae on synthetic diets. First, 20 first instar
larvae were placed in each of five 250 ml waxed carton cups containing 100 grams of a
pinto bean based diet used for the cabbage looper (Guy et al. 1985). In a second attempt,
20 first-instar larvae were placed individually in 40 ml clear plastic cups, each containing
10 ml of a commercial diet used for species of Heliothis (Heliothis premix, by Stonefly
Industries Inc., Bryan, Texas, U.S.A.). In both cases, larvae were checked daily to
assess survival and development.
Comparison of Plant Species as Hosts for M. tinctoides Larvae
Six species of Commelinaceae and cotton, Gossypium hirsutum L. were evaluated
for their suitability as hosts. Plants of T. zebrina were obtained commercially, from
local nurseries. Patches of T. fluminensis and the dayflower Commelina diffusa Bur-
meister were found growing wild in Gainesville and were collected as needed. Both are
apparently introduced species that have become established in this area (Clewell 1985).
The spiderwort, Tradescantia ohiensis Raffinesque, is native to Florida and is abundant
in sunny disturbed areas. It also was collected as needed. Tradescantia pallida (Rose)
D. Hunt was initially found in an outside ornamental planting in Gainesville and was
cultivated for this study in soil on the floor of a greenhouse. Patches of Callisia repens
L. of known origin were found inside and outside of a greenhouse at the USDA, ARS,
Insect Attractants Laboratory, Gainesville, Florida. Cotton (variety Germaine 510) was
grown from seed in pots of sterilized potting soil in a glass greenhouse. Cotton plants
used were about 40 cm tall and had not set floral buds.
First-instar larvae of M. tinctoides were placed in one liter plastic cannisters with
cut foliage of a plant species. Ten larvae were placed in each cannister and were moni-
tored daily until adult emergence. Foliage was added as required, and about 2 cm of a
mixture of ground corn cob and pine shavings was kept in the bottom of cannisters to
absorb excess moisture and to serve as a pupation medium. Larvae were transferred
to new cannisters when the pine shavings became soaked or mold appeared. Four such
cannisters were set up for each Commelinaceae plant species and three cannisters were
set up with cotton foliage. Records were kept of survival rates, as well as pupation and
emergence dates. Pupae were sexed and weighed.
Data comparing development times and pupal weights for M. tinctoides on different
plant species were subjected to analysis of variance (ANOVA), following procedures
for a completely randomized design, and Duncan's Multiple Range Test to determine
significant differences among means (Steel and Torrie 1960). Pupal weights for males
and females were compared using Student's t-test. Voucher specimens of M. tinctoides
were placed in the Florida State Collection of Arthropods and in the collection of the
Of the plants tested for suitability as larval hosts, highest survival rates to the adult
stage were obtained with C. diffusa (at 72.5%), and low survival rates occurred with
T. pallida (7.5%) and C. repens (12.5%) (Table 1). Rates of survival to the adult stage
for M. tinctoides reared on T. fluminensis, T. ohiensis, and T. zebrina were similar
(52.5 to 55%). Rates for survival to pupation were similar to those for adult emergence
Landolt: Host Plants for Mouralia 575
TABLE 1. MEANS (-+ STANDARD ERROR) FOR DEVELOPMENT TIMES IN DAYS FROM
EGG HATCH TO ADULT EMERGENCE, PERCENTAGES OF INDIVIDUALS
SURVIVING FROM EGG HATCH TO ADULT, AND MEAN PUPAL WEIGHTS
FOR M. TINCTOIDES REARED ON 6 SPECIES OF COMMELINACEAE AND
Development % Survival Pupal Weight
Plant Species Time (d) Pupa Adult (mg)
T.fluminensis 33.6 0.3a 60.0% 52.5% 428.4 18.2bc
T. ohiensis 33.7 0.3a 55.0% 55.0% 464.8 12.2c
T. pallida 36.0 1.5c 17.5% 7.5% 345.1 49.6a
T. zebrina 36.1 0.6c 55.0% 52.5% 426.9 12.7bc
C. diffusa 35.5 0.3bc 77.5% 72.5% 412.5 13.5b
C. repens 34.8 + 0.3b 27.5% 12.5% 308.1 39.4a
G. hirsutum 0.0% 0.0% -
Means in a column followed by the same letter are not significantly different by Duncan's (1955) Multiple Range
Test at p s 0.05.
for the four plant species yielding the highest Mouralia survival rates (Table 1), showing
little pupal mortality.
There were significant differences among development times of larvae on different
plant species, as indicated by the ANOVA results (F=7.0, p<0.0001). Mean develop-
ment time from larval hatching to adult emergence ranged from 33.6 days for larvae on
T. fluminensis to 36.0 days for larvae on T. pallida (Table 1). There were also signifi-
cant differences among pupal weights for M. tinctoides on different plants as indicated
by ANOVA (F=6.2, p=0.0005). Highest mean pupal weights were for larvae reared
on T. ohiensis (464.8 mg) and T. fluminensis (428.4 mg), with lowest pupal weights for
C. repens (308.1 mg) and T. pallida (345.1 mg) (Table 1). No larvae survived longer
than two days on cotton foliage.
For pupae obtained from the colony of M. tinctoides maintained on T. fluminensis
and T. zebrina, mean weights of male pupae (n= 111) were significantly greater than
mean weights of female pupae (n= 105) (436.2 7.5 for males, 403.7 7.7 for females,
t=3.03, p=0.02, df=214). The sex ratio (determined for pupae) for larvae reared on
the four suitable species of Commelinaceae was near 1 to 1 (51:49, female:male, n = 96).
No larvae of M. tinctoides placed on either the pinto bean diet or the Heliothis
premix diet survived past the first instar. All died without apparent feeding on the diet.
Previous host plant records for M. tinctoides are limited to reports of two species
of Commelinaceae: the wandering jew species T. zebrina (Zebrina pendula) and T.
fluminensis (Comstock 1938, Eichlin & Cunningham 1978, Crumb 1956). The findings
reported here substantiate that these are good hosts for M. tinctoides. Both supported
rapid development and high rates of survival to the adult stage. This study also docu-
ments the equal suitability of two additional species of Commelinaceae as larval hosts;
T. ohiensis and C. diffusa. The remaining plant species evaluated, T. pallida and C.
repens, were poor hosts for M. tinctoides, as evidenced by poor larval survival and reduced
pupal weights. However, some larvae were able to complete development on these two
host species. First-instar larvae placed on cotton foliage apparently did not feed.
There is little information on what plant species M. tinctoides utilizes as hosts in
nature. Comstock (1938) found M. tinctoides on foliage of T. zebrina and demonstrated
complete development from egg to adult on this plant. Also, six M. tinctoides larvae
576 Florida Entomologist 76(4) December, 1993
collected on escaped plants of T. zebrina in Fresno, California were reared to maturity
by the author. There are no records of collections of larvae on the other four species of
Commelinaceae evaluated. No larvae were found on wild host plants in this study,
although the moth appears to be widely distributed in the state of Florida and can be
found throughout the year (although infrequently) in Gainesville.
Problems in continuous maintenance of a colony of M. tinctoides included the failure
of larvae to consume artificial diet, poor oviposition (and presumably mating) in the
laboratory, and high mortality of mature larvae. Adequate oviposition was obtained by
placing adult moths in cages with potted host plants, on which all eggs were laid. It
was considered that host plant kairomones may be involved in mating behavior and in
the stimulation of oviposition. A high mortality of mature larvae may have been due to
a lack of suitable material to use for the formation of pupation cells and cocoons, and to
excess moisture from chewed foliage. This was alleviated largely by adding the mixture
of ground corncob and pine shavings to boxes housing larvae and plant foliage.
These results and findings together provide information needed to rear significant
numbers of M. tinctoides in the laboratory and add information on the suitability of
several species of Commelinaceae as host plants of this noctuid.
The technical assistance of K. Davis-Hernandez is gratefully acknowledged. Initial
attempts to rear M. tinctoides on diet were made by P. Ponce. C. E. Curtis (USDA,
ARS, Fresno, CA) collected M. tinctoides larvae on T. zebrina in Fresno, CA. Nancy
Coile, Florida State Department of Agriculture, Division of Plant Industry, Gainesville,
Florida, kindly provided determinations on plants used. This work was supported in
part by the Cooperative State Research Service, USDA, Agreement No. 90-37250-5356.
CLEWELL, A. F. 1985. Guide to the vascular plants of the Florida panhandle. Univ.
Press of Florida, Tallahassee. 605 pp.
COMSTOCK, J. A. 1938. A new record and a life history (Lepidoptera). Bull. South.
Calif. Acad. Sci. 32: 169-174.
CRUMB, S. E. 1956. The larvae of the Phalaenidae. U.S.D.A. Technical Bulletin No.
DUNCAN, D. B. 1955. Multiple range and multiple F tests. Biometrics 11: 1-41.
EICHLIN, T. D., AND H. B. CUNNINGHAM. 1978. The Plusiinae (Lepidoptera: Noc-
tuidae) of America north of Mexico, emphasizing genitalic and larval morphology.
U.S.D.A. Technical Bulletin No. 1567: 1-122.
GUY, R. H., N. C. LEPPLA, J. R. RYE, C. W. GREENE, S. L. BARRETTE, AND K.
A. HOLLIEN. 1985. Trichoplusia ni, pp. 487-494 in P. Singh and R. F. Moore,
[eds.] Handbook of insect rearing, vol. 3. Elsevier, Amsterdam.
SHOREY, H. H., AND R. L. HALE. 1965. Mass rearing of the larvae of nine noctuid
species on a simple artificial medium. J. Econ. Entomol. 58: 522-524.
STEEL, R. G. D., AND J. H. TORRIE. 1960. Principles and procedures of statistics.
McGraw Hill Book Co., N.Y. 481 pp.
Portillo et al.: Soybean Looper Pyrethroid Resistance
PYRETHROID RESISTANCE LEVELS IN SOYBEAN LOOPER
(LEPIDOPTERA: NOCTUIDAE) IN MISSISSIPPI
H. E. PORTILLO, C. M. FELLAND, H. N. PITRE, AND R. P. PORTER
Mississippi State University
Mississippi State, MS 39762
Resistance to permethrin was monitored in soybean looper, Pseudoplusia includes
(Walker), strains established with larvae collected from soybean and cotton during the
growing season in the central Delta area in Mississippi in 1989 and 1991. Dose responses
were tested topically using larvae. Populations (= strains) of first generation soybean
looper collected on soybean in August in both years had LDs's [95% confidence limits
(CL), expressed as gg/g larval weight] of 1.30 (0.80-2.30) and 0.77 (0.01-2.19) in 1989
and 1991, respectively, which were not significantly different from the LDo5 (CL) of the
susceptible strain [0.63 (0.47-0.87) in 1989 and 0.18 (0.13-0.23) in 1991]. Soybean looper
strains (second generation on soybean) collected in September showed 6.8 to 10.8-fold
increase in LDso (CL) [4.30 (3.00-6.80) in 1989 and 1.94 (0.31-4.04) in 1991], compared
with the susceptible strain. The two strains collected in September showed 3.31 and
2.52-fold increases in resistance levels, respectively, when compared with the strains
collected in August 1989 and 1991. Soybean looper strains collected on cotton during
September had levels of resistance similar to those collected on soybean in both years.
However, a soybean looper strain collected on cotton in October (third generation) in
1989 showed a significantly higher level of resistance than any other strain. Levels of
soybean looper resistance to pyrethroid insecticides were low early in the season, but
increased as the season progressed.
Key Words: Insecticide resistance, loopers, cotton, soybean.
La resistencia a permetrina en razas del gusano medidor de la soya, Pseudoplusia
includes (Walker), establecidas con larvas colectadas en soya y algod6n durante la
6poca de cultivo en el Area central del Delta en Mississippi durante 1989 y 1991 fue
evaluada. Respuestas a d6sis aplicadas topicalmente en larvas fueron evaluadas. Pobla-
ciones (= razas) de P. includes colectadas en soya en agosto de ambos afos tuvieron
un DLso [lfmites fiduciales al 95% (LF), expresado en ig/g peso larval] de 1.30 (0.80-
2.30) y 0.77 (0.01-2.19) en 1989 y 1991, respectivamente, los cuales no fueron sig-
nificativamente diferentes a el de la raza susceptible [0.63 (0.47-0.87) en 1989 y 0.18
(0.13-0.23) en 1991]. Las razas colectadas en septiembre, segunda generaci6n en soya,
SB89-2 y SB91-2, mostraron incrementos significativos en el DLs (LF) a raz6n de 6.8X
a 10.8X [4.30 (3.00-6.80) en 1989 y 1.94 (0.31-4.04) en 1991], si se comparan con la raza
susceptible de cada aflo. Las razas SB89-2 y SB91-2 mostraron aumentos de 3.31X y
2.52X en el nivel de resistencia al compararse con las razas SB89-1 y SB91-1, respec-
tivamente. Razas colectadas en algod6n durante septiembre tuvieron niveles de resis-
tencia similares a los de las colectadas en soya en ambos afios. Sin embargo, una raza
colectada en algod6n a principios de octubre de 1989, (tercera generaci6n), mostr6
niveles de resistencia significativamente was altos que cualguiera de las otras razas.
Estos datos indican que los niveles de resistencia a insecticides piretroides en P. inclu-
dens fueron bajos al comienzo de la temporada, pero se aumentaron significativamente
a media que la temporada transcurri6.
578 Florida Entomologist 76(4) December, 1993
Resistance of the soybean looper, Pseudoplusia includes (Walker), to pyrethroid
insecticides has been documented in areas of the southern United States where soybean,
Glycine max (L.) Merrill, and cotton, Gossypium hirsutum L., are grown in close
proximity (Felland et al. 1990, Leonard et al. 1990). It has been speculated that selection
for insecticide resistance in the soybean looper can be linked to cotton (Luttrell et al.
1990), which can serve as a host for soybean looper larvae (Canerday & Arant 1966)
and provide nectar for adults (Burleigh 1972, Jensen et al. 1974). Burleigh (1972)
suggested that the soybean looper can develop on a number of other plant species before
they move to soybean plants that are sufficiently developed to be attractive as ovipos-
itional hosts. However, the importance of cotton as a host of the soybean looper was
thought to decline as the crop growing season advanced (Felland et al. 1992). This may
be related to the decline in nectar availability as the cotton plants mature (Stone et al.
Soybean loopers in Puerto Rico were observed to be difficult to control with some
insecticides used on soybean in the United States (Boethel 1990). If pyrethroid insect-
icides are used extensively for control of soybean looper on crops grown outside the
continental United States, the appearance of migratory pyrethroid-resistant loopers in
the southern United States might be explained.
In this study, soybean looper larvae were collected from soybean and cotton in the
Delta of Mississippi and monitored for resistance to permethrin insecticide at selected
times during the crop growing season.
MATERIALS AND METHODS
Soybean looper strains, in 1989 and 1991, were established by collecting larvae from
natural populations on soybean and cotton. A laboratory (LAB-MS) strain, maintained
at the Southern Field Crop Insect Management Laboratory at Stoneville, Mississippi,
was the source of the susceptible insects. This colony was started with moths from a
South Carolina laboratory colony. Moths from larvae collected on untreated soybean in
the Delta in Mississippi were added into the colony in 1981 and 1986. The colony has
not been exposed to insecticides in the laboratory.
Field collections of larvae were made based on larval size. Larvae from generations
one through three were identified and separated according to instars within sample
collections. The first collections were made on 17 August 1989 and 14 August 1991 from
soybean (SB89-1, n>200 and SB91-1, n=250) in a field adjacent to cotton in Holmes
County (fields were separated by about five miles). Larvae representing the second
generation (or possibly new immigrants) were collected from the same soybean fields
and from an adjacent cotton field on 2 September 1989 and 1991 (SB89-2, n>200 and
SB91-2, n=320) and from cotton on 2 September 1989 and 16 September 1991 (C89-2,
n>200 and C91-2, n =219). A third collection of larvae, representing the third genera-
tion, was taken from cotton on 2 October 1989 (C89-3, n>200). The soybean fields had
not been sprayed with pyrethroid insecticide before the insect collections, whereas the
cotton fields had received seven pyrethroid insecticide applications at approximately
10-day intervals from 14 July to 9 September 1989, and five pyrethroid insecticide
applications at approximately 10-day intervals from 17 July to 6 September 1991.
Larvae and soybean or cotton foliage (depending on larval host) were placed in
plastic bags inside a box containing blue ice and transported to a laboratory at the
Department of Entomology, Mississippi State University. Larvae were transferred
individually to plastic cups (29.6 ml) containing nutri-soy flour/wheat germ diet
(Raulston & Shaver 1970). The first laboratory generation for each field strain was
obtained by randomly selecting 25 adult males and 25 adult females from each field
collection. These were maintained in two 4-liter glass jars [no more than 30 adults
Portillo et al.: Soybean Looper Pyrethroid Resistance
(males + females) per jar] and fed a 30% honey water solution. Muslin was hung inside
as an oviposition substrate and was also used to cover the jars. Further laboratory
generations from each field strain were obtained by rearing at least 300 neonates using
the same procedure.
The standard assay procedure for measuring insecticide resistance in Heliothis spp.
was followed (Anonymous 1970). Neonates in the first and second laboratory genera-
tions were randomly selected from each field strain. Early third instars (pooled mean
weight 12.66 2.93 mg, mean SD) were topically treated on the thoracic dorsum
with 1-pl aliquot of acetone (check) or an acetone solution of one of five serial concentra-
tions of technical grade permethrin (FMC Corporation, Philadelphia, PA 19103).
Treated larvae were held in the cups with diet and mortality was recorded 48 h after
treatment. Total mortality was recorded as the number of dead and moribund larvae.
Dose-mortality regressions and the 95% confidence limits (CL) for the LDso (based
on pig of permethrin insecticide per g larval body weight) were obtained by probit
analysis using the microcomputer software Polo (Robertson et al. 1980). Differences
among strains were considered significant based on the failure of 95% CL to overlap.
Resistance ratios (RR) were calculated by dividing the LDo of each field strain by that
of the LAB-MS susceptible strain.
RESULTS AND DISCUSSION
Resistance levels of soybean looper larvae representing the first field generation
were 2.1 and 4.3 times greater than that of the susceptible LAB-MS in 1989 and 1991,
respectively (Table 1). However, the LDsos (95% CL) were not significantly different
from that of the LAB-MS strain. The results suggest that resistance selection pressure
for pyrethroid insecticides at source sites of migratory soybean looper moths arriving
in this area of Mississippi may be relatively low. This assumption needs documentation.
TABLE 1. TOXICITY OF PERMETHRIN TO SEVERAL STRAINS OF SOYBEAN LOOPER
LARVAE COLLECTED IN THE MISSISSIPPI DELTA (TOPICAL BIOASSAYS).
Strain Generation' Generation2 n3 Slope SE LDo (95% CL)4 RR6
Soybean 2 1 146 1.45 0.27 13.31 (8.92-23.6) 21.2
Lab-MS 1 356 2.27 0.24 0.63 (0.47-0.87) -
Soybean 1 2 71 1.61 0.33 1.30 (0.80-2.30) 2.1
Soybean 2 2 107 1.63 0.28 4.30 (3.00-6.80) 6.8
Cotton 2 2 130 1.99 0.27 4.60 (2.20-9.50) 7.3
Cotton 3 1 80 2.59 0.51 34.4 (23.4-51.1) 54.6
Lab-MS 1 500 1.55 0.15 0.18 (0.13-0.23) -
Soybean 1 1 120 1.10 0.36 0.77 (0.01-2.19) 4.3
Soybean 2 1 476 1.24 0.15 1.94 (0.31-4.04) 10.8
Cotton 2 1 375 2.06 0.43 1.79 (0.44-2.90) 9.9
1Represents generations of larvae collected as they appear in the field through time.
2Represents generations of larvae after transferring to the laboratory.
sNumber of subjects excluding controls.
'Doses reported in pg of insecticide/pjg larval weight.
'Resistance ratios (RR) = LDo of the field strain/LDo, Lab-MS.
'From Felland et al. 1990.
580 Florida Entomologist 76(4) December, 1993
The earliest confirmed presence of soybean looper adults in pheromone traps occurred
on 12 July in both 1989 (Porter 1990) and 1991 (Southern Regional Soybean Looper
Resistance Project, 1991). This indicated that in both years some immigrant moths were
exposed to at least one pyrethroid application on cotton. Using the developmental times
for soybean looper larvae, pupae and adults on cotton and the pre-oviposition and ovi-
position periods for females (Mitchell 1967), the number of pyrethroid applications
applied after the first immigrant moths arrived can be calculated (using a conservative
approach of non-overlapping ages of eggs, larvae and adults). Eggs of the first genera-
tion on cotton were exposed to one pyrethroid insecticide application, first generation
larvae to two applications, and first generation adults, possibly, to one insecticide appli-
cation in 1989. Thus, soybean loopers were exposed to 4 and possibly 5 insecticide
applications (including the first on immigrant moths) prior to the collection of the SB89-1
strain. In 1991, eggs and larvae of the first generation on cotton were exposed to one
pyrethroid insecticide application and the adults of the first generation to another appli-
cation. This accounted for 3 to 4 applications prior to the collection of the SB91-1 strain.
If adults resulting from the first generation of larvae on cotton are responsible for eggs
laid on soybean, giving rise to the first generation on soybean, these data would support
the idea of low levels of pyrethroid resistance in migratory soybean looper moths arriving
in Mississippi. Low levels of resistance were observed in the SB89-1 (RR =2.1) and SB91-1
(RR=4.3) strains even after the populations had been exposed to 3 to 5 pyrethroid
insecticide applications. Because of the relatively low population density in the first
generation of soybean looper and the possibility that susceptible migratory moths were
still arriving in the study area, the observed low levels of insecticide resistance may
have been caused by influx of susceptible genes from migratory moths. It is also plausible
that moths emerging from other ecosystems mated with those from the first larval gener-
ation on cotton, and diluted the resistance allele frequency. Both effects can be neglig-
ible in following generations due to explosion in soybean looper population density.
There were significantly higher levels of pyrethroid resistance in the second field
generation of soybean looper larvae collected on soybean (SB-2) compared with the first
field generation on soybean (SB-1) and the LAB-MS strain in both 1989 and 1991.
SB89-2 and SB91-2 strains showed significantly higher levels of resistance (6.8 and
10.8-fold, respectively) as compared with the LAB-MS strains (Table 1). Although
SB89-2 and SB91-2 strains showed higher levels of resistance than that observed in
SB89-1 and SB91-1 strains (3.31 and 2.52-fold, respectively), the increase was signifi-
cantly higher only in the SB89-2 strain. The second field generation of soybean looper
larvae collected on cotton in 1989 and 1991 also showed significantly higher LDes as
compared with the susceptible strain, with 7.3 and 9.9-fold levels of resistance, respec-
tively (Table 1). LD.s observed in C89-2 and C91-2 strains were not significantly
different from levels observed in SB89-2 and SB91-2 strains when comparisons within
years were made. Soybean fields had not been treated with pyrethroids in either years.
These results support the hypothesis that selection for pyrethroid resistance in soybean
looper occurs on cotton. The data also support observations that soybean looper control
failures with pyrethroid insecticides on soybean planted in the vicinity of cotton are
more frequent than insecticide failures on soybean planted away from cotton (Felland
et al. 1990, Leonard et al. 1990).
Soybean loopers collected from soybean (SB-2 strain) and cotton (C-2 strain) during
early to mid-September in 1989 were exposed to 6 to 7 pyrethroid applications. Those
collected from soybean (SB-2 strain) and cotton (C-2 strain) about the same time in 1991
were exposed to 4 to 5 and 5 to 6 applications, respectively. The increase in exposure
of soybean loopers through the second generation to pyrethroid insecticide represents
an increase in selection pressure for pyrethroid resistance. These results indicate that
selection pressure for resistance to pyrethroid insecticide on migratory and first gener-
Portillo et al.: Soybean Looper Pyrethroid Resistance 581
ation soybean looper may be key to the rapid increase in levels of resistance observed
in the second field generation of this insect pest. This increase in resistance in the
soybean looper population may be attributed, in part, to the presence of low to moderate
levels of resistance in migratory moths. These levels of resistance may not have been
detected by our techniques.
The third field generation of soybean looper larvae collected from cotton in 1989 (no
third generation collected on cotton in 1991) showed a significantly higher resistance
level (LDo) to permethrin as compared with the C89-2 (7.5-fold increase) and the sus-
ceptible LAB-MS strain (54.6-fold increase) (Table 1).
Because no additional pyrethoid insecticide applications were made after the collec-
tion of the C89-2 strain, this and the C89-3 strain were exposed to the same number of
pyrethroid applications before resistance testing. The dramatic increase in levels of
pyrethroid resistance observed in the C89-3 strain may be related in great part to
inbreeding of resistant moths in the second generation, migration of moths exposed to
permethrin insecticide in other fields, and to an exponential increase in resistance after
a certain level of resistant allele frequencies is attained (Tabashnik & Croft 1982).
Increased levels of insecticide resistance in H. virescens during the growing season has
been reported (Graves et al. 1991). Resistance was linked with selection pressure on
the insect population exposed to increased number of insecticide applications.
Resistance levels of 1989 strains were generally higher, although not all of them
were significantly different, than those of 1991 strains (Table 1). Felland et al. (1990)
reported resistance in a strain of soybean looper (second generation larvae) collected
from soybean in 1987 at the same location as insects used in the present study. The
dose-mortality lines of the three strains were compared. There was a trend for a higher
level of pyrethroid resistance in the 1987 strain compared with the 1989 and 1991 strains
(Fig. 1). The 1987 strain had a significantly higher LD, than the 1989 or 1991 strains
S 0.8 -
H 4 f /
H 0.4 /
m 0.4 .... ."""0** SB89-2
S/ m O'"0n SB91-2
S ql- im m LAB-MS91
-2 -1 0 1 2 3
LOG DOSE (ug/g Larval weight)
Fig. 1. Dose-mortality lines for permethrin on laboratory and second generation
field strains of soybean loopers collected on soybean in the Mississippi Delta in 1987,
1988, and 1991.
582 Florida Entomologist 76(4) December, 1993
(Table 1). The lower number of pyrethroid applications in 1991 than in 1989 may have
contributed to the lower levels of insecticide resistance in 1991 compared with 1987 and
The number of pyrethroid insecticide applications on cotton were closely related to
the levels of H. virescens infestations on cotton, with 1989 having higher infestations
than 1991 (Head 1990, 1992). Although the number of pyrethroid applications on cotton
in the study fields in 1987 is unknown, the H. virescens infestation levels on cotton in
1987 were higher than in 1989 or 1991. It is reasonable to consider that there were more
pyrethroid insecticide applications on cotton in 1987 than in 1989 or 1991 (King et al.
1988, Head 1990, 1992); thus, the selection pressure for pyrethroid resistance in soybean
looper was greater in 1987 than in the other two years.
Results of this study indicate increased resistance levels to pyrethroid insecticides
in soybean looper as the growing season progressed in 1989. We suggest that the
frequent use of pyrethroid insecticides on cotton early in the growing season signifi-
cantly increases the selection pressure for resistance in soybean looper on cotton. Low
levels of pyrethroid resistance were observed early in the 1989 and 1991 crop growing
seasons with significant increases in levels of resistance in the populations in 1989 and
a 2.52-fold increase in resistance between SB91-1 and SB91-2 strains as the season
progressed. A significant increase in the levels of resistance also was observed between
the C89-2 and C89-3 strains.
Pyrethroid insecticide resistance management programs for soybean looper on soy-
bean should be viewed in the broad scope of integrated pest management in the crop
ecosystem. In areas where the soybean looper is a constraint to soybean production, it
can contribute to high levels of defoliation of cotton. Here it is exposed to large amounts
of insecticide and initiates insecticide resistance build up. The Heliothis pyrethroid
resistance management program in the southern United States advises the use of insec-
ticides other than pyrethroids on the earliest infestations of this pest on the cotton crop,
thereby delaying selection for H. virescens resistance (Anonymous 1986). Soybean
looper control on soybean should benefit, as well, from the reduced number of pyrethroid
insecticide applications on cotton early in the season, especially when soybean is planted
in the proximity of cotton. Additionally, methods for detection of low levels of resistance
in migratory soybean looper moths are needed for the effective implementation of insec-
ticide resistance management programs. A glass vial bioassay technique to test for
permethrin resistance in adult soybean looper (Mink et al. 1993) can be used to document
insecticide resistance in this pest. Identifying low levels of insecticide resistance early
in the year could indicate the potential of soybean looper outbreak. Additional work is
needed to associate topical LD, responses with estimates of gene frequencies and
anticipated levels of control in the field.
We thank Drs. J. Schneider, J. Funderburk, R. Luttrell, and D. Boethel for reviews
and comments on the manuscript. Mississippi Agricultural and Forestry Experiment
Station publication number J-8269.
ANONYMOUS. 1970. Standard test method for determining resistance to insecticides
in Heliothis, pp. 147-149 in J. R. Brazzel [ed.], Second Conference on Test
Methods for Resistance on Insects of Agricultural Importance. College Park,
Md., 1-4 April 1969. Bull. Entomol. Soc. America 16: 147-153.
Portillo et al.: Soybean Looper Pyrethroid Resistance
ANONYMOUS. 1986. Cotton entomologists seek to delay pyrethroid resistance in in-
sects. Mississippi Agricultural & Forestry Experiment Station Research High-
lights 49: 8.
BOETHEL, D. J. 1990. Factors contributing to infestations of resistant soybean loopers
in soybean, pp. 32-36 in J. Hamer and H. Pitre [eds.], Soybean Looper Resist-
ance Workshop 1990, Presentation Abstracts, Mississippi Cooperative Extension
Service, Mississippi State, MS. 41 pp.
BURLEIGH, J. G. 1972. Population dynamics and biotic control of the soybean looper
in Louisiana. Environ. Entomol. 1: 290-294.
CANERDAY, T. D., AND F. S. ARANT. 1966. The looper complex in Alabama
(Lepidoptera: Plusiinae). J. Econ. Entomol. 59: 742-743.
FELLAND, C. M., H. N. PITRE, R. G. LUTRELL, ANDJ. L. HAMER. 1990. Resistance
to pyrethroid insecticides in soybean looper (Lepidoptera: Noctuidae) in Missis-
sippi. J. Econ. Entomol. 83: 35-40.
FELLAND, C. M., R. P. PORTER, AND H. N. PITRE. 1992. Soybean looper (Lepidop-
tera: Noctuidae) oviposition preference relative to plant development in soybean
and cotton. J. Entomol. Sci. 27: 217-223.
GRAVES, J. B., B. R. LEONARD, S. MICINSKI, AND E. BURRIS. 1991. A three year
study of pyrethroid resistance in tobacco budworm in Louisiana: Resistance man-
agement implications. Southwest. Entomol. 15: 33-41.
HEAD, R. B. 1990. Cotton insect losses 1989. 43rd Annual Conference Report on
Cotton Insect Research and Control, pp. 157-162 in Proceedings Beltwide Cotton
Production Research Conferences, Las Vegas, NV. National Cotton Council of
America, Memphis, TN. 203 pp.
HEAD, R. B. 1992. Cotton insect losses 1991. 45th Annual Conference Report on
Cotton Insect Research and Control, pp. 621-625 in Proceedings Beltwide Cotton
Production Research Conferences, Nashville, TN. National Cotton Council of
America, Memphis, TN. 332 pp.
JENSEN, J. L., L. D. NEWSOM, AND J. GIBBENS. 1974. The soybean looper: Effects
of adult nutrition on oviposition, mating frequency, and longevity. J. Econ. En-
tomol. 67: 467-470.
KING, E. G., J. R. PHILLIPS, AND R. B. HEAD. 1988. Highlights of the 1988 cotton
insect research and control conference. 41st Annual Conference Report on Cotton
Insect Research and Control, pp. 188-202 in Proceedings Beltwide Cotton Pro-
duction Research Conferences, New Orleans, LA. National Cotton Council of
America, Memphis, TN. 191 pp.
LEONARD B. R., D. J. BOETHEL, A. N. SPARKS, M. B. LAYTON, J. S. MINK, A. M.
PAVLOFF, E. BURRIS, AND J. B. GRAVES. 1990. Variation in response of soy-
bean looper (Lepidoptera: Noctuidae) to selected insecticides in Louisiana. J.
Econ. Entomol. 83: 27-34.
LUTTRELL, R. G., C. M. FELLAND, AND J. MALLET. 1990. Levels of insecticide
resistance in soybean looper populations in the southern United States in the
1980's: Indications for the future, pp.19-21 in J. Hamer and H. Pitre [eds.] Soy-
bean Looper Resistance Workshop 1990, Presentation Abstracts, Mississippi
Cooperative Extension Service, Mississippi State, MS. 41 pp.
MINK, J. S., D. J. BOETHEL, AND B. R. LEONARD. 1993. Monitoring permethrin
resistance in soybean looper (Lepidoptera: Noctuidae) adults. J. Entomol. Sci.
MITCHELL, E. R. 1967. Life history of Pseudoplusia includes (Walker) (Lepidop-
tera: Noctuidae). J. Georgia Entomol. Soc. 2: 53-57.
PORTER, P. 1990. Monitoring soybean looper populations, p. 27 in J. Hamer and H.
Pitre [eds.] Soybean Looper Resistance Workshop 1990, Presentation Abstracts,
Mississippi Cooperative Extension Service, Mississippi State, MS. 41 pp.
RAULSTON, J. R. AND T. N. SHAVER. 1970. A low agar caseine wheat germ diet for
rearing tobacco budworms. J. Econ. Entomol. 63: 1743-1744.
ROBERTSON, J. L., R. M. RUSSELL, AND N. E. SAVIN. 1980. POLO: A user's guide to
probit or logit analysis. U.S. Forest Service General Technical Report PSW-38.
Florida Entomologist 76(4)
SOUTHERN REGIONAL SOYBEAN LOOPER RESISTANCE PROJECT. 1991. p. 14, in
Southern Regional Soybean Looper Pheromone Trapping Project, 1991 Report.
Prepared by J. Hamer and H. Pitre. Mississippi Cooperative Extension Service,
Mississippi State, MS. 185 pp.
STONE, T. B., H. N. PITRE, AND A. C. THOMPSON. 1984. Relationships of cotton
phenology, leaf soluble protein, extrafloral nectar carbohydrate and fatty acid
concentrations with populations of five predator species. J. Georgia Entomol.
Soc. 19: 204-212.
TABASHNIK, B. E., AND B. A. CROFT. 1982. Managing pesticide resistance in crop-ar-
thropod complexes: Interactions between biological and operational factors. En-
viron. Entomol. 11: 1137-1144.
EVALUATION OF LOSS OF (+)-DISPARLURE FROM
GYPSY MOTH (LEPIDOPTERA: LYMANTRIIDAE)
PHEROMONE DISPENSER TAPES UNDER
FIELD CONDITIONS IN FLORIDA
JAMES L. NATION,1 JOHN L. FOLTZ,'
WAYNE N. DIXON,2 AND HEATHER J. McAUSLANE1
'Department of Entomology & Nematology,
Univ. of Florida, Gainesville, FL 32611-0620
2Florida Division of Forestry
Currently Division of Plant Industry,
Florida Dept. Agric. and Consumer Serv.
Gainesville, FL 32607
The residual pheromone content of laminated plastic, pheromone-dispensing tapes
impregnated with (+)-disparlure, the sex pheromone of the gypsy moth, Lymantria
diapar (L.), was assessed after exposure to field conditions in Gainesville. As deter-
mined by gas chromatography, lure tapes deployed on 26 August 1991 rapidly lost
pheromone during the first 2 months in the field. Loss of pheromone was considerably
less for the remaining 4-month exposure period (28 October 1991 to 3 March 1992). Lure
tapes at 5 locations differed slightly in their rates of pheromone loss, and traps placed
on the north side of tree trunks retained more pheromone than traps placed on the south
side. These data indicate that a pheromone lure used for monitoring gypsy moth during
spring and early summer in north Florida may lose its effectiveness rapidly and may
have to be replaced more often than is currently recommended for other regions of
Key Words: Lymantria dispar, sex pheromone, pheromone trap, population monitor-
ing, gas chromatography.
Se evaluaron despues de exposici6n a las condiciones de campo en Gainesville,
Florida, U.S.A., el contenido residual de feromona de plistico laminado, cintas de dis-
pensar feromona impregnadas con (+)-disparlure, la feromona de la polilla gitana,
Lymantria dispar (L.). Segun anilisis por cromotografia de gas, cintas atrayentes col-
Nation et al.: Lose of Disparlure from Traps in Florida 585
ocadas 26 de agosto de 1991 perdieron rapidamente la feromona durante los primeros
d6s meses en el campo. La p6rdida fue considerablemente menos durante el period
restante de 4 meses (28 de octubre de 1991 hasta 3 de marzo de 1992). Cintas atrayentes
en 5 localidades difieron ligeramente en la tasa de p6rdida de la feromona, y trampas
colocadas en el lado norte de troncos de arboles retuvieron mas feromona que las tram-
pas colocadas en el lado sur. Estos datos indican que un atrayente de feromona usado
para monitoreo de la polilla gitana durante la primavera y los principios de verano en
el norte de la Florida podria perder su efectividad rapidamente y tener que estar
reemplazado mas frequentamente que se recomienda para otras regions del pais.
The larvae of the gypsy moth, Lymantria dispar (L.), defoliate many forest and
shade trees in the northeastern United States. The gypsy moth is slowly spreading
westward and southward from the northern U.S. Small outbreaks are periodically de-
tected and eradicated on the perimeter of the main population (Schwalbe 1981, McManus
et al. 1989). Detection is commonly made with traps baited with the pheromone, (+)-dis-
parlure (R,S-7,8-epoxy-2-methyloctadecane), which attracts male gypsy moths (Bierl et
al. 1970, Yamada et al. 1976).
Disparlure has potential as an effective lure for population control (Beroza & Knip-
ling 1972, Schwalbe & Mastro 1988), and it is used for monitoring population spread
(Elkinton & Carde 1980, Elkinton & Carde 1981, McManus et al. 1989). Trap catches
are reliable indicators of population density (Kolodny-Hirsch & Schwalbe 1990, Thorpe
et al. 1993), which is an important consideration in planning a control/eradication strategy
when a new infestation is discovered.
The loss of disparlure from various dispensers and its effectiveness after deployment
have been evaluated in a number of studies (Leonhardt & Moreno 1982, Leonhardt et
al. 1990, Leonhardt et al. 1992). None of these studies, however, was done under the
typical high temperature and high humidity conditions that exist in Florida. Although
there is no known established population in Florida, small numbers of moths are cap-
tured in Florida in the spring each year (Foltz & Dixon, unpublished data), and are
believed to be from egg masses laid on vehicles or other articles brought into Florida
by tourists. Thus, there is concern that at some point it may be necessary to deploy
large numbers of disparlure-baited traps for monitoring, control and/or evaluation of an
incipient population. This report is an evaluation of the field loss of disparlure from
commercial Hercon Luretape strips in delta traps.
MATERIALS AND METHODS
Pheromone-impregnated strips were from Hercon Luretape, Lot #D0048 remain-
ing from those used by USDA APHIS PPQ in 1991. We purchased 100 mg neat (+)-dis-
parlure from Hercon to use in evaluating our chromatographic system. Pheromone
tapes were kept in a freezer (-20 C) until used. The tapes were advertised to contain
500 pg (+)-disparlure per tape. We stapled 21 tapes into each of 10 Pherocon III D
traps from Trece, Inc., P.O. Box 6278, Salinas, CA 93915. The traps were not coated
with sticky material. Two traps were placed at each of five locations in Gainesville, FL,
one on the north side and one on the south side of a tree at the five different locations.
Traps were placed approximately 1.25 m above the ground. We removed three tapes
from each trap on 26 August 1991 when the traps were initially prepared for field
deployment, and thereafter at monthly intervals on 26 September, 28 October, 26
November 1991, and on 2 January, 3 February, and 3 March 1992. Each tape was
removed with forceps, placed into a small vial (approximately 2 ml capacity) with screw
cap containing a teflon liner, brought to the laboratory and placed into a freezer (approx-
586 Florida Entomologist 76(4) December, 1993
imately -200 C) until analysis. In preparation for analysis, approximately 1.5 ml of
acetone:pentane (50:50 v/v) was added to each vial to cover the tape, and a measured
amount of methyl hexadecanoate (16 carbon fatty acid methyl ester-C16 FAME) in
acetone:pentane (50:50) was added as an internal standard. The vials were allowed to
remain on the bench top overnight to thoroughly solubilize the disparlure remaining in
the tape. The quantity of C16 FAME was reduced each month to give an internal
standard quantity in each vial that approximately equaled the quantity of disparlure
expected to remain in the tapes. Vials were shaken thoroughly before removing approx-
imately 1 pA for injection into the gas chromatograph.
Samples were chromatographed on a non-polar 25 m x 0.25 mm capillary column
containing a bonded polydimethylsiloxane coating (Alltech RSL 150) in a Shimadzu
G14-A gas chromatograph with capillary injector port and flame ionization detector
(FID). The column was initially set at 1500 C for 3 min, and then temperature was
programmed to increased at a rate of 4 C/min to 2000 C and held for 10 min. The column
temperature was then raised rapidly (300 C/min.) to 2900 C to drive off certain impurities
that showed on chromatograms well after the elution of C16 FAME and disparlure. The
injection port was set at 2700 C and the FID at 3200 C. Samples were injected in the
splitless mode, with a purge flow of carrier after 0.5 min. The carrier gas flow (helium)
was adjusted to a flow rate of 25 cm/min.
In order to evaluate C16 FAME as an internal standard, we co-chromatographed
weighed samples of C16 FAME and (+)-disparlure, and calculated percent recovery of
expected (weighed) (+)-disparlure from its peak area as a function of the peak area for
C16 FAME. Freshly weighed samples were chromatographed initially, and again in
November and December 1991, and in January 1992.
Analysis of variance was performed on residual amounts of pheromone in lure tapes
(SAS Institute 1987). Means were compared using LSD test analysis at the 5% signifi-
RESULTS AND DISCUSSION
The purity of the commercial sample of C16 FAME was 98.9% and that of disparlure
was 96.8%, as determined in our GC system. Disparlure eluted from the capillary col-
umn in about 14.4 min while C16 FAME eluted at about 12.1 min throughout the 6
months the experiment was in process. We found 99.8% 8.0% (mean + SD, n = 28) of
expected (weighed) samples of (+)-disparlure with C16 FAME as the internal standard.
This high rate of recovery indicated that the chromatographic system was functioning well
and that C16 FAME was an adequate internal standard. C16 FAME eluted about 2 min
before (+)-disparlure and also contained oxygen in the molecule, as does (+)-disparlure.
The data from recovery of disparlure remaining in dispenser tapes and associated
loss rates are shown in Table 1. The quantity of disparlure (mean SD) initially de-
tected in tapes as the experiment began (432 39 xg, range 357-521 jg, n = 30 tapes)
was about 14% less than the expected 500 pg stated by the manufacturer. There may
have been some loss of disparlure from the tapes during shipment and handling before
we started the experiment. It may be difficult to recover all the disparlure originally
incorporated into lure tapes. Leonhardt et al. (1990) found pheromone amounts in new
tapes within the range of 400-500 pg.
After the traps had been in the field for one month, the' residual amount of
pheromone decreased to 303.7 13.2 Rg (Table 1). This indicates a release of 128 pLg
for the period, with an average loss rate of 172 ng hr- (Table 1). Residual amounts of
pheromone and rate of loss continued to fall until late November and then remained
nearly constant for the next 3 months.
Nation et al.: Lose of Disparlure from Traps in Florida 587
TABLE 1. GAS CHROMATOGRAPHIC DETERMINATION OF CHANGE IN RESIDUAL
AMOUNT OF (+)-DISPARLURE FROM HERCON LURETAPE EXPOSED IN
THE FIELD IN GAINESVILLE, FLORIDA, FROM AUGUST 1991 TO MARCH
Mean + 95% CI Loss Rate
Sample Date Day g in Tape jig/Day ng/h
26Aug 1991 0 432.1 13.9 -
26 Sep 1991 31 303.7 + 13.2 4.1 172
28 Oct 1991 63 234.1 15.7 2.2 91
26 Nov 1991 92 221.8 + 16.4 0.4 17
2 Jan 1992 129 174.6 + 17.3 1.3 53
3 Feb 1992 161 225.6 + 24.7 -
3 Mar 1992 190 198.0 18.6 1.0 40
Data were pooled for all dates except the initial date when tapes were taken out of
their protective packaging and immediately analyzed. As expected, analysis of variance
indicated that sampling date was a major source of variability in pheromone residual
amounts (F5,168 = 26.55; P = 0.0001). Location was also significant (F4,168 = 5.59; P =
0.0003). The compass direction that a trap faced was not quite significant (F,16s = 3.15;
P = 0.0779); however, north-facing traps retained more pheromone than south-facing
traps on 5 of 6 sample dates.
Leonhardt et al. (1990) indicated that traps caught large numbers of male gypsy
moths only when the lure emitted at least 30 ng h-1, and had a residual content of at
least 100 ig. On average, lure tapes from the present experiment in Florida provided
this emission rate over the 6 months of the study, but it should be emphasized that the
experiment was conducted from late August to early March. The average temperatures
for these months in Gainesville are 25.70 C (September), 22.0 C (October), 17.40 C
(November), 13.8 C (December), 13.0C (January), and 13.70 C (February). The aver-
age temperatures for the last 4 months of the experiment (Table 2) are considerably
cooler than historical temperatures for April (20.10 C), May (23.40 C), June (26.6 C)
and July (27.4 C) when traps are most likely to be needed in north Florida to monitor
gypsy moths. Leonhardt et al. (1990) found that release rates increased 3.5-fold for each
100 C increase in temperature, and the log of the release rate was proportional to the
inverse of temperature (Bierl-Leonhardt et al. 1979). Summer months are also consider-
ably more humid and rainy than fall and winter in north Florida, conditions that are
likely to cause rapid aging of the pheromone lure. Lures deployed during the spring
TABLE 2. SUMMARY OF TEMPERATURE RECORD DURING THE EXPERIMENTAL
Temperatures (0 C)
Time Period Max Min Average
26 Aug 91-26 Sep 91 35.6 16.7 26.3
27 Sep 91-28 Oct 91 32.2 4.4 21.3
29 Oct 91-26 Nov 91 30.0 -5.0 15.4
02 Dec 91-02 Jan 92 30.0 -3.9 15.3
03 Jan 92-03 Feb 92 25.6 -6.7 11.1
04 Feb 92-03 Mar 92 32.2 -2.8 15.4
588 Florida Entomologist 76(4) December, 1993
and summer are likely to experience loss rates as high as those occurring during Sep-
tember (172 pLg hr- 1), which would quickly deplete the lure tape of (+)-disparlure.
Additional research under seasonal monitoring conditions should be conducted in order
to refine recommendations for monitoring of gypsy moth in north Florida.
We thank P. MacLean, Product Development Mgr., Hercon Environmental Co.,
Aberdeen Rd., Emigsville, PA 17318 for suggesting the mixture of acetone:pentane as
a solvent for dissolving disparlure from the tapes, and K. Milne and D. Boyd for tech-
nical assistance in gas chromatography. We are grateful to USDA APHIS PPQ, Winter
Haven, FL for the gift of the disparlure impregnated tapes and to Mr. Bill Lingren,
Trece, Inc., P.O. Box 6278, Salinas CA 93915 for the gift of the Pherocon III D traps.
This is Florida Agriculture Experiment Station Journal Series No. R-03274.
BEROZA, M., AND E. F. KNIPLING. 1972. Gypsy moth control with the sex attractant
pheromone. Science 177: 19-27.
BIERL, B. A., M. BEROZA, AND C. W. COLLIER. 1970. Potent sex attractant of the
gypsy moth: its isolation, identification and synthesis. Science 170: 87-89.
BIERL-LEONHARDT, B. A., E. D. DEVILBISS, AND J. R. PLIMMER. 1979. Rate of
release of disparlure from laminated plastic dispensers. J. Econ. Entomol. 72:
ELKINTON, J. S., AND R. T. CARDE. 1980. Distribution, dispersal, and apparent sur-
vival of male gypsy moths as determined by capture in pheromone-baited traps.
Environ. Entomol. 9: 729-737.
ELKINTON, J. S., AND R. T. CARDE. 1981. The use of pheromone traps to monitor
distribution and population trends of the gypsy moth, pp. 41-55 in E. R. Mitchell
[ed.], Management of Insect Pests with Semiochemicals. Plenum, New York.
KOLODNY-HIRSCH, D., AND C. P. SCHWALBE. 1990. Use of disparlure in the man-
agement of the gypsy moth, pp. 363-385 in R. L. Ridgway, R. M. Silverstein
and M. N. Inscoe [eds.], Behavior-Modifying Chemicals for Insect Management.
Marcel Dekker, New York.
LEONHARDT, B. A., AND D. S. MORENO. 1982. Evaluation of controlled release lami-
nate dispensers for pheromones of several insect species, pp. 159-173 in B. A.
Leonhardt and M. Beroza [eds.], Insect Pheromone Technology: Chemistry and
Applications. American Chemical Society Symposium Series 190. American
Chemical Society, Wash., D.C.
LEONHARDT, B. A., V. C. MASTRO, E. C. PASZEK, C. P. SCHWALBE, AND E. D.
DEVILBISS. 1990. Dependence of gypsy moth (Lepidoptera: Lymantriidae) cap-
ture on pheromone release rate from laminate and other dispensers. J. Econ.
Entomol. 83: 1977-1981.
LEONHARDT, B. A., V. C. MASTRO, AND E. D. DEVILBISS. 1992. Evaluation of
pheromone dispensers for use in gypsy moth detection (Lepidoptera: Lyman-
tridae). J. Entomol. Sci. 27: 280-284.
MCMANUS, M., N. SCHNEEBER, R. REARDON, AND G. MASON. 1989. Gypsy moth.
USDA Forest Service, Forest Insect & Disease Leaflet 162. 13 pp.
SAS INSTITUTE. 1987. Guide for personal computers, version 6 ed. Cary, NC.
SCHWALBE, C. P. 1981. Disparlure-baited traps for survey and detection, pp. 542-548
in C. C. Doane and M. L. McManus [eds.], The Gypsy Moth: Research Toward
Integrated Pest Management, Forest Service, Science and Education Agency,
Animal and Plant Health Inspection Service, Technical Bulletin 1584, U.S. De-
partment of Agriculture, Wash., D.C.
SCHWALBE, C. P., AND V. C. MASTRO. 1988. Gypsy moth mating disruption: dosage
effects. J. Chem. Ecol. 14: 581-588.
Deyrup & Atkinson: Evaniids 589
THORPE, K. W., R. L. RIDGWAY, AND B. A. LEONHARDT. 1993. Relationship be-
tween gypsy moth (Lepidoptera: Lymantriidae) pheromone trap catch and popu-
lation density: Comparison of traps baited with 1 and 500 jg (+)-disparlure
lures. J. Econ. Entomol. 86: 86-92.
YAMADA, M., T. SAITO, K. KATAGIRI, S. IWAKI, AND S. MARUMO. 1976. Electroan-
tennogram and behavioral responses of the gypsy moth to enantiomers of dispar-
lure and its trans analogues. J. Insect Physiol. 22: 755-761.
i --- -c -- L--
SURVEY OF EVANIID WASPS
(HYMENOPTERA: EVANIIDAE) AND THEIR COCKROACH
HOSTS (BLATTODEA) IN A NATURAL FLORIDA HABITAT
MARK DEYRUP' AND THOMAS H. ATKINSON2
'Archbold Biological Station
P.O. Box 2057
Lake Placid, FL 33852
2Department of Entomology
University of California
Riverside, CA 92521
Over a 3-year period 1,359 evaniids, representing four species, were collected in a
mature sand pine habitat at the Archbold Biological Station in south-central Florida.
All species fly for at least six months of the year and show annual fluctuations in
abundance. Seventeen species of cockroaches occur at the ABS; the most probable hosts
for evaniids are members of the genera Parcoblatta, Ischnoptera, and Cariblatta.
Key Words: Parasitoids, ecology, behavior, population density.
Durante un period de 3 afios, 1,359 evanfidos, representando 4 species, se colecta-
ron en un habitat de pino de arena en la Estaci6n Biologica Archbold (EBA) en el centro
de la Florida. Todas las species vuelan durante por lo menos 6 meses del afio y muestran
fluctuaciones annuales en abundancia. Dieciseis species de cucarachas ocurren en la
EBA. Las hospederas probables de los evaniidos son miembros de los generous Parcob-
latta, Ischnoptera y Cariblatta.
Evaniid wasps are a small group of specialized solitary parasitoids, living only in
egg cases of cockroaches. What little is known about host relationships (see Townes
1949, Roth & Willis 1960) suggests that each species of evaniid is specialized to attack
egg cases of a particular size, sometimes those of a genus or closely related genera of
cockroaches. This is not surprising, as different cockroaches deposit their egg cases in
different situations, and the egg cases themselves differ in size and structure. It is
difficult to get direct information on hosts and general ecology of evaniids because they
590 Florida Entomologist 76(4) December, 1993
breed in deliberately concealed hosts, and they are inconspicuous insects that spend
much of their time crawling about in dense vegetation. A survey of cockroaches and
evaniids, combined with a study of seasonal flight patterns of the latter, has provided
new biological information, including indirect information on hosts.
MATERIALS AND METHODS
The study site is on the Archbold Biological Station (Highlands County), located at
the southern end of the Lake Wales Ridge in south-central Florida. The site is in a
transitional zone between warm and subtropical zones. Winters are mild and dry, with
temperatures during some years falling below 0 C for a few hours. Sheltered micro-
habitats are frost-free. Summers are warm and humid, with daytime temperatures over
The vegetation of the study site is a thin canopy of sand pine (Pinus clausa Chap-
man), with a thick 1.5-3.5 m understory of scrub oaks (Quercus spp.), staggerbush
(Lyonia spp.), silk bay (Persea humilis Nash), scrub pawpaw (Asimina obovata Nash),
and scrub hickory (Carya floridana Sargent). The paths through this thick brush ap-
peared to act as flight corridors for insects.
Evaniids were collected in 2 small Townes traps (Townes 1972) that were set up
across 2 east-west paths. The traps were kept in place and continuously monitored from
May 1983 through December 1986. Each trap was annually replaced with an identical
trap to forestall the effects of wear. Specimens were collected 3 times a week. Cock-
roaches were collected by Townes traps, pitfall traps and searching litter, rotten wood,
and under bark. The evaniids and cockroaches were identified by the authors. Speci-
mens of all species are in the collection of the Archbold Biological Station.
RESULTS AND DISCUSSION
In the seasonal flight study, we collected 1,359 evaniid specimens, representing the
species Evaniella semaeoda Bradley, Hyptia floridana Ashmead, H. reticulata (Say),
and H. thoracica (Blanchard). The numerical results are summarized in Fig. 1. From
these results we infer the following:
1. Every species has a flight period extending over at least seven months.
These flight periods are all longer than those reported by Townes (1949), and
imply that hosts are breeding over much of the year.
2. There is no evidence of niche partitioning on the basis of seasonality.
3. The time required for development is not known for any of the four evaniid
species. For the introduced peridomestic evaniid Prosevania punctata
(Brull6) developmental time ranges from 40-127 days (Roth & Willis 1960).
Considering the warm temperatures in the study site between April and
October, it seems likely that all species would have time for more than one
generation within the flight period, the latter presumably synchronized with
4. There appear to be notable variations in abundance and seasonality from
year to year. This is best seen in H. reticulata and H. thoracica, which were
collected in relatively large numbers.
The survey of cockroaches is presented in Table 1. From the combined surveys of
cockroaches and evaniids we can infer the following.
1. The exotic Evania appendigaster (L.), which attacks Periplaneta spp. in
Florida (Stange 1978), was not collected in our traps; Periplaneta spp. at the
Deyrup & Atkinson: Evaniids 591
Seasonal Flight of Evaniella ~seaeoda Seasonal Flight of Hyptia floridana
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
E 1983 t 10B4 m 1985 t 1 se 1983 ~ ns 19S 4 m 1985 ss 186
Seasonal Flight of Hyptia reticulata Seasonal Flight of _hiLa thoracic
0 n.302 180 .-
0 140 I
Fig. 1. Seasonal flight patterns of four species of evaniid wasps.
Archbold Biological Station seem confined to disturbed areas near buildings.
2. There is no evaniid big enough to be associated with Eurycotis floridana
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Doc Jan Feb Mar Apr May Jun Jul Aug Sep Oct Noe Dec
tZ93 = 1904 t loss F 1986 tB 1 98 7 194 t ltess toes
Fig. 1. Seasonal flight patterns of four species of evaniid wasps.
Archbold Biological Station seem confined to disturbed areas near buildings.
2. There is no evaniid big enough to be associated with Eurycotis floridana
(Walker), unless there is a species that consumes only part of the eggs in an
3. Small cockroaches whose egg cases might be appropriate for Hyptia
floridana are Cariblatta lutea (Saussure & Zehntner), C. minima Hebard,
and possibly the small cockroaches Compsodes cucullatus (Saussure &
Zehntner), Euthlasto blatta gemma Hebard, and Chorisoneura texensis
(Saussure & Zehntner). Our impression is that individuals of Cariblatta spp.
vastly outnumber all other possible hosts. It seems highly likely that C. lutea
is the principle host of H. floridana.
4. For Evaniella semaeoda, Hyptia reticulata, and H. thoracica, plausible
hosts are Parcoblattafulvescens (Saussure & Zehntner), Latiblatella rehni,
Ischnoptera deropeltiformis (Brunner), and just possibly Arenivagafloriden-
sis Caudell; the latter is a burrowing species that probably buries its eggs
deep in the sand. All three of these medium-sized evaniids occur far north
of Florida, in areas where Parcoblatta and Ischnoptera species are the only
possible hosts (Carlson 1979). H. thoracica and H. reticulata are known to
parasitize Parcoblatta pensylvanica (DeGeer) (Roth & Willis 1960). Parcob-
latta fulvescens is abundant at the study site. We reared a specimen of E.
semoeoda from an ootheca of I. deropeltiformis found in deep pine litter in
5. The diversity of cockroach genera, both native and exotic species, is greater
in southern Florida than elsewhere in the Southeast (Atkinson et al. 1991).
Florida Entomologist 76(4)
TABLE 1. SPECIES OF COCKROACHES PRESENT AT STUDY SITE AND DIMENSIONS
Length of Height of Size
Family Species Ootheca Ootheca (n)
Blattidae Eurycotisfloridana 16.9 .28 7.4 .09 251
Periplaneta americana 9.0 .13 5.5 .05 251
P. australasiael 1.5 4.5
Polyphagidae Arenivagafloridensis2 7.9 .14 3.8 +.08
Blattellidae Euthlastoblatta gemma4 3.9 2.3 1
C. minima 3.3.04 2.0.03 201
Chorisoneura texensis6 3.4 2.1 1
Ischnoptera deropeltiformis 7.4 .19 4.0 .06 51
I. bilunata 8.0+.38 3.2.05 23'
Latiblattella rehni7 7.2+.09 3.0.03 3
Parcoblattafulvescens 6.3 +.09 3.4 .03 31
Blaberidae Panchlora niveas
'Based on oothecae produced by adults collected in the field by T. H. Atkinson. May-June 1990, Gainesville, FL.
'Based on laboratory colony from Alachua & Levy counties.
'Found in ant nests.
'Based on single ootheca protruding from abdomen of female with following collection of data Georgia, 5 mi NE
Leesburg, 7-XI-63, W. Glover (USNM).
"Ootheca carried by female.
'Based on single ootheca with following collection data: NC, Nag's Head, NII-55, K. V. Krombein (USNM).
'Florida, Archbold Biological Station, 13-14-V-80, Weems & Hohren (FCSA); Florida, Gainesville, 7-VI-61, H. V.
'Ootheca carried internally until hatched.
In spite of this, there is not a greater number of species of evaniids in
Florida, and no species is confined to Florida. On the other hand, there are
at least six southeastern species of Parcoblatta that do not occur in south
Florida (Atkinson et al. 1990) and only one additional evaniid, Hyptia har-
pyoides Bradley, within the range of these species. In other words, cockroach
diversity in the Southeast is in no way correlated with evaniid diversity. The
distribution of the six native evaniids could be explained by the distribution
of the genera, Parcoblatta, Ischnoptera and Cariblatta.
Nancy Deyrup prepared the graphs of flight activity, and Marcia Moretto typed the
ATKINSON, T. H., P. G. KOEHLER, AND R. S. PATTERSON. 1990. Checklist of the
cockroaches of Florida (Dictyoptera: Blattaria: Blattidae, Polyphagidae, Blattel-
lidae, Blaberidae). Florida Entomol. 73: 303-327.
Lounibos & Machado-Allison: Mosquito Ovipositional Cues 593
ATKINSON, T. H., D. G. KOEHLER, AND R. S. PATTERSON. 1991. Catalog and atlas
of the cockroaches (Dictyoptera) of North America north of Mexico. Entomol.
Soc. Amer. Misc. Publ. 78. 43 pp.
CARLSON, R. W. 1979. Superfamily Evanioidea, pp. 1109-1118 in K. V. Krombein, P.
D. Hard, Jr., D. R. Smith, and B. D. Burks [eds.], Catalog of Hymenoptera in
America North of Mexico. Vol. I. Symphyta and Apocrita (Parasitica). Smithso-
nian Inst. Press, Washington, D.C. 1198 pp.
ROTH, L. M., AND E. R. WILLIS. 1960. The biotic associations of cockroaches.
Smiths.Misc. Coll. 141: 1-470.
STANGE, L. A. 1978. Evania appendigaster (L.), a cockroach egg parasitoid
(Hymenoptera: Evaniidae). Florida Dept. Agric. and Consum. Serv., Div. P1.
Indust. Entomol. Circ. No. 191. 2 pp.
TOWNES, H. 1949. The nearctic species of Evaniidae (Hymenoptera). Proc. U.S. Nat.
Mus. 99: 525-539.
TOWNES, H. 1972. A light-weight Malaise trap. Entomol. News 83: 239-247.
FIELD TEST OF MOSQUITO OVIPOSITIONAL CUES
FROM VENEZUELAN PHYTOTELMATA
L. P. LOUNIBOS' AND C. E. MACHADO-ALLISON2
'University of Florida,
Florida Medical Entomology Laboratory,
200 9th St. SE, Vero Beach, FL 32962
2nstituto de Zoologia Tropical,
Universidad Central de Venezuela,
Apartado 47058, Caracas 1040A, Venezuela
Fluids held by four phytotelmata were compared for oviposition by mosquitoes in
lowland rainforest in eastern Venezuela. Significantly more Wyeomyia ulocoma and
Culex pleuristriatus were recovered in fluid from bracts of Heliconia caribaea, than in
fluids collected from axils of Aechmea bromeliads, the aroid Alocasia macrorrhiza, or
interodes of Bambusa vulgaris. Wyeomyia ulocoma, whose immature stages occur
uniquely in Heliconia bracts, was more specific to H. caribaea fluid than was the
phytotelm generalist C. pleuristriatus. No preferences for oviposition site color were
Key Words: Diptera, fluids, culex, Wyeomyia, Heliconia.
Fluidos retenidos por cuatro fitotelmatas fueron comparados en una selva lluviosa
de tierra baja en el oriented de Venezuela con respect a la frecuencia de oviposici6n por
mosquitos. Significativamente mAs Wyeomyia ulocoma y Culex pleuristriatus fueron
colectados en el fluido de las brdcteas de Heliconia caribaea que en los fluidos de las
axilas de bromelias del g6nero Aechmea, de las axilas de Alocasia macrorrhiza, o de
los internodios del bambil Bambusa vulgaris. La W. ulocoma, cuales estadios
594 Florida Entomologist 76(4) December, 1993
preimaginales ocurren unicamente en las bracteas de Heliconia, fu6 mas especifica al
fluido de H. caribaea que la C. pleuristriatus. No se encontraron preferencias por
diferentes colors en las tazas de oviposici6n.
Mosquitoes use visual, olfactory, and tactile cues to determine appropriate egg lay-
ing sites, the relative importance of these stimuli being highly species-dependent
(Bentley & Day 1989). Both site color and chemistry may affect the female's choice, and
these cues may interact (e.g., Wilton 1968). Natural compounds that influence oviposi-
tion include some of plant origin (Bentley & Day 1989).
Phytotelmata are well suited for experiments on oviposition behavior because of
their small size and replicability. These plant-held pools are particularly important as
habitats for mosquitoes in the tropics (Frank & Lounibos 1983). For the few mosquito
species studied, either phytotelm odor or color have been shown to affect oviposition
site selection. Egg laying by both the fruit-husk specialist Eretmapodites sub-
simplicipes (Edw.) and the pitcher-plant mosquito Wyeomyia smithii (Coq.) is enhanced
by unidentified chemicals derived from the preferred habitats of these species (Lounibos
1978, Istock et al. 1983). Compounds that promote oviposition by the treehole mosquito
Aedes triseriatus (Say) have been identified as phenolic derivatives from trees (Bentley
et al. 1979, 1982). Chemical stimuli from their Tillandsia host plants did not affect
oviposition choices of Wyeomyia mitchelli (Theobald) or Wyeomyia vanduzeei Dyar and
Knab, but these bromeliad inhabitants did prefer yellow-green oviposition sites to other
colors (Frank 1986).
The phytotelm fauna of Panaquire, Venezuela is relatively well known thanks to a
series of studies by the authors and their students and colleagues in the 1980s (Machado-
Allison et al. 1983, 1985, Lounibos et al. 1987). Mosquitoes encountered at Panaquire
include specialists inhabiting only one type of water-holding plant and generalists that
occupy diverse phytotelmata (Machado-Allison et al. 1985). Oviposition behavior has
been examined in only one mosquito resident of Panaquire, Trichoprosopon digitatum
(Rondani), whose egg laying in cacao husks was influenced by fluid aroma and husk
shape (Lounibos & Machado-Allison 1983). Here we report results from a field experi-
ment in which choices of fluids from four phytotelmata were presented in colored con-
tainers to gravid females in nature.
SITE AND METHODS
The experiment was performed on a cacao plantation in lowland tropical rain forest
near Panaquire (10 13'N, 660 14'W), Miranda State, Venezuela. Our field test was
conducted in July 1983 in the middle of the long rainy season (Machado-Allison et al.
The aquatic contents of four classes of phytotelmata growing naturally at the planta-
tion were poured or pipetted with a turkey baster into separate plastic buckets.
Phytotelmata sampled were (1) axils of Alocasia macrorrhiza (L.) (Araceae); (2) bracts
of Heliconia caribaea Lamarck (Heliconiaceae); (3) internodes of Bambusa vulgaris
(Schrad.) (Gramineae); (4) axils of Aechmea aquilega (Salisb.) and Aechmea nudicaulis
(L.) (Bromeliaceae)-collections from these two bromeliads were pooled.
The contents of buckets were passed separately through 80-mesh/cm2 sieves to sepa-
rate detritus and invertebrate fauna from fluid. Disposable plastic cups of 20 ml capacity
were used to expose the four fluids for mosquito oviposition. This capacity is slightly
larger than that of an average H. caribaea bract or aroid or bromeliad axil, but consid-
Lounibos & Machado-Allison: Mosquito Ovipositional Cues 595
erably smaller than the average volume held by a B. vulgaris internode. To simulate
the colors of phytotelmata, cups were spray-painted matte red (H. caribaea), green (A.
macrorrhiza or Aechmea sp.), or off-white or black (controls).
Oviposition choices were arranged in a 4 x 4 factorial design of the colors and fluids.
Cups were half-filled with fluid (10 ml) and suspended by hooks on a crosswire 1.5 m
above ground in shade under the canopy because sites with more light were not availa-
ble. The relative position of each fluid/color combination was determined for each repli-
cate by selection of random permutations of 16 (Cochran & Cox 1957). Twelve replicates
of the 16 treatments were separated by one-meter distances on the crosswire.
Cups were set at dusk on the same day as fluids were collected, and the experiment
terminated after 86-88 hours exposure by preserving the contents of each cup in sepa-
rate bags with 70% ethanol. Larvae were counted and identified in the laboratory with
keys (Lounibos et al. 1987). Eggs were not counted because of the lack of specific keys.
Based on an estimate of 1.5 to two days from oviposition to hatch, our methods recorded
only egg-laying occurring within the first 48 hours after setting of cups.
Numbers of mosquitoes per cup were treated by two-way ANOVA with fluid origin
and cup color as independent variables. An Fmax test revealed no significant differences
among variances (F, = 6.06; F(12,11).05 = 7.48) after numbers were transformed by
square root of (Y + 0.5). Fluid source, but not cup color, significantly influenced mos-
quito oviposition, with no significant interaction between the factors (Table 1).
Four species of mosquito were identified from 601 larvae. Culex pleuristriatus
Theobald and Wyeomyia ulocoma (Theobald) accounted for, respectively, 77.9% and
15.3% of identified specimens. Three larvae of Wyeomyia pertinans (Williston) were
recovered in three separate cups of A. macrorrhiza fluid and 38 larvae of T. digitatum
were found in one cup of A. macrorrhiza fluid.
For the two commonest mosquito species captured, numbers of larvae per cup were
compared among the four fluids. As the numbers of W. ulocoma among fluids were
heteroscedastic even after transformation, the Kruskal-Wallis (KW) statistic (H) was
used in place of ANOVA to measure preference (Lounibos 1981).
Larvae of C. pleuristriatus were recovered in all four fluids, but the mean number
in H. caribaea extracts was more than twice that in alternative plant fluids, yielding
the significant difference in location by the KW test (Fig. 1A). Wyeomyia ulocoma also
preferred to oviposit in H. caribaea fluid and more selectively than C. pleuristriatus,
as reflected by the higher H value from the KW test (Fig. 1B); 95.7% (88/92) of the
identified W. ulocoma were recorded from this plant fluid.
TABLE 1. ANALYSIS OF VARIANCE (TWO-WAY) OF THE EFFECTS OF FLUID SOURCE
AND CUP COLOR ON MOSQUITO OVIPOSITION'.
Source of Variation Df MS F
Fluids 3 14.65 11.54***2
Colors 3 0.78 0.61
Fluids X Colors 9 0.40 0.31
Error 176 1.27
'Number of larvae per cup transformed by square root of (Y + 0.5).
2*** = P<0.001.
Florida Entomologist 76(4)
n = 468
H = 13.63
P = 0.0035
A. B. H. Aechmea
macrorrhiza vulgaris caribaea spp.
n = 92
H = 131.96
P << 0.001
A. B. H. Aechmea
macrorrhiza vulgaris caribaea spp.
Fig. 1. Mean numbers of the two most common mosquito species identified in ex-
perimental cups containing fluids from four phytotelmata. Error bars show standard
error of the mean, n is the total number of larvae and H the Kruskal-Wallis statistic
evaluated for significance against chi-square(8).
Lounibos & Machado-Allison: Mosquito Ovipositional Cues 597
It is conceivable, but unlikely, that differential mortality or, in the case of C. pleuri-
striatus differential egg raft size, caused the observed distributions of larvae (Fig. 1)
instead of differential oviposition. A differential mortality hypothesis would require
that egg or larval survival be significantly higher in the H. caribaea fluid. However,
we observed no evidence of mortality such as decaying corpses or the occurrence of
predators in cups, at the time that our experiment was terminated. A differential egg
raft size hypothesis would require that C. pleuristriatus had laid egg masses in H.
caribaea fluid that were more than twice the size as those in the other phytotelm fluids.
While it is impossible to discount this possibility without counting eggs in rafts, we
know of no precedence for such site-specific variation in mosquito egg raft size. There-
fore, we feel it safe to conclude that the distributions of larvae among phytotelm fluids
reflect ovipositional preferences.
Because no behavioral observations were made of the choice by C. pleuristriatus
and W. ulocoma of H. caribaea fluid, we have avoided use of the terms attractantt' and
'stimulant' which have been defined in relation to specific behaviors elicited (Dethier et
Fluids of our chosen phytotelmata have different origins. Rainfall or throughfall
probably accounts for most fluid held by axils of A. macrorrhiza and Aechmea
bromeliads. In contrast, at least some of the fluid in H. caribaea bracts is of plant origin,
because newly opened bracts already contain liquid. These young bracts are highly
attractive to ovipositing C. pleuristriatus and W. ulocoma, as evidenced by the
superabundance of early instar larvae in them (Machado-Allison et al. 1983). It seems
likely that these two mosquito species are responding to phytochemical cues in their
choices of H. caribaea fluid, although our experiment does not rule out other sources
of chemicals. In this respect H. caribaea resembles the North American pitcher plant
Sarracenia purpurea whose newly opened pitchers are favored oviposition sites of the
inquiline mosquito W. smithii (Fish & Hall 1978) owing to a water-soluble chemical
secreted by the plant (Istock et al. 1983).
Of the two commonest mosquitoes recovered from cups, W. ulocoma showed the
stronger preference for H. caribaea fluid, as indicated by the higher H value (Fig. 1).
The higher selectivity of this species agrees with the habitat valency of the two in
nature, W. ulocoma being a specialist in bracts of H. caribaea, and C. pleuristriatus
occupying diverse phytotelmata, including bamboo internodes and the axils of aroids
and bromeliads (Machado-Allison et al. 1985, Lounibos et al. 1987).
The two species, W. pertinans and T. digitatum, recorded infrequently from ovipos-
ition cups, were both regarded as exploiting diverse phytotelmata by Machado-Allison
et al. (1985). Although our experiment recovered the former only in fluid of A. mac-
rorrhiza, in nature it inhabits both bromeliad and aroid axils (Machado-Allison et al.
1985). The one cup with T. digitatum larvae contained A. macrorrhiza fluid, although
bamboo internodes and cacao husks are the preferred habitats of this species at
Panaquire (Machado-Allison et al. 1985). Curiously, T. digitatum occurs commonly in
Heliconia bracts at Rancho Grande, Venezuela (Seifert 1980).
Oviposition behaviors of the mosquito species recorded in cups fall into two
categories: raft formation by C. pleuristriatus and T. digitatum and aerial oviposition
by Wyeomyia spp. which eject eggs singly on to the water surface. Since the raft-for-
mers 'decide' upon one oviposition site for a complete egg clutch, whereas Wyeomyia
'inspect' the fluid surface prior to ejecting each egg (Frank, unpublished), it could be
argued that the five-fold fewer larvae of W. ulocoma still represent a greater number
of oviposition 'decisions' by that species compared to C. pleuristriatus.
Florida Entomologist 76(4)
While Frank (1986) showed that the bromeliad specialists W. mitchelli and W. van-
duzeei discriminated colors of oviposition choices in cages, Istock et al. (1983) could not
detect color preferences for egg sites of W. smithii. It would be premature to conclude
from our experiment that neither C. pleuristriatus nor W. ulocoma use color as an
oviposition site cue. In nature H. caribaea grows in open sunlight and Aechmea spp.
and A. macrorrhiza occur in semi-shade, but our experiment was set in shade, where
light intensity may have been too low for diurnal Wyeomyia spp. (Frank et al. 1985) to
use visual cues. Also, our green and red spray paints were not calibrated to conform
spectrally with the colors of phytotelmata intended for simulation.
Research was supported by NSF INT-8212581 of the USA and CONICIT S1-1332
of Venezuela. This is University of Florida IFAS Journal Series No. R-02481.
BENTLEY, M. D., AND J. F. DAY. 1989. Chemical ecology and behavioral aspects of
mosquito oviposition. Ann. Rev. Entomol. 34: 401-421.
BENTLEY, M. D., I. N. MCDANIEL, M. YATAGAI, H.-P. LEE, AND R. MAYNARD.
1979. p-Cresol: an oviposition attractant ofAedes triseriatus. Environ. Entomol.
BENTLEY, M. D., I. N. MCDANIEL, AND E. E. DAVIS. 1982. Studies of 4-methyl-cyc-
lohexanol: an Aedes triseriatus (Diptera: Culicidae) oviposition attractant. J.
Med. Entomol. 19: 589-592.
COCHRAN, W. G., AND G. M. Cox. 1957. Experimental Designs. Second edition. John
Wiley and Sons, New York.
DETHIER, V. G., L. BARTON BROWNE, AND C. N. SMITH. 1960. The designation of
chemicals in terms of the responses they elicit from insects. J. Econ. Entomol.
FISH, D., AND D. W. HALL. 1978. Succession and stratification of aquatic insects
inhabiting the leaves of the insectivorous pitcher plant, Sarracenia purpurea.
American Midi. Nat. 99: 172-183.
FRANK, J. H. 1986. Bromeliads as ovipositional sites for Wyeomyia mosquitoes: form
and color influence behavior. Florida Entomol. 69: 728-742.
FRANK, J. H., AND L. P. LOUNIBOS (eds.). 1983. Phytotelmata: Terrestrial Plants as
Hosts for Aquatic Insect Communities. Plexus, Marlton, New Jersey. 293 pp.
FRANK, J. H., H. C. LYNN, AND G. M. GOFF. 1985. Diurnal oviposition by Wyeomyia
mitchelli and W. vanduzeei (Diptera: Culicidae). Florida Entomol. 68: 493-496.
ISTOCK, C. A., K. TANNER, AND H. ZIMMER. 1983. Habitat selection by the pitcher-
plant mosquito, Wyeomyia smithii: behavioral and genetic aspects, pp. 191-204
in J. H. Frank and L. P. Lounibos [eds.], Phytotelmata: Terrestrial Plants as
Hosts for Aquatic Insect Communities. Plexus, Marlton, New Jersey.
LOUNIBOS, L. P. 1978. Mosquito breeding and oviposition stimulant in fruit husks.
Ecol. Entomol. 3: 299-304.
LOUNIBOS, L. P. 1981. Habitat segregation among African treehole mosquitoes. Ecol.
Entomol. 6: 129-154.
LOUNIBOS, L. P., AND C. E. MACHADO-ALLISON. 1983. Oviposition and egg brood-
ing by the mosquito Trichoprosopon digitatum in cacao husks. Ecol. Entomol. 6:
LOUNIBOS, L. P., J. H. FRANK, C. E. MACHADO-ALLISON, P. OCANTO, AND J. C.
NAVARRO. 1987. Survival, development and predatory effects of mosquito lar-
vae in Venezuelan phytotelmata. J. Trop. Ecol. 3: 221-242.
MACHADO-ALLISON, C. E., D. J. RODRIGUEZ, R. BARRERA-R., AND C. GOMEZ
COVA. 1983. The insect community associated with inflorescences of Heliconia
Rodriguez & Loera: Damage by Black Cutworm to Corn 599
caribaea Lamarck in Venezuela, pp. 247-270 in J. H. Frank and L. P. Lounibos
[eds.], Phytotelmata: Terrestrial Plants as Hosts for Aquatic Insect Com-
munities. Plexus, Marlton, New Jersey.
MACHADO-ALLISON, C. E., R. BARRERA-R., J. H. FRANK, L. DELGADO, AND C.
GOMEZ COVA. 1985. Mosquito communities in Venezuelan phytotelmata, pp.
79-93 in L. P. Lounibos, J. R. Rey, and J. H. Frank [eds.], Ecology of Mos-
quitoes: Proceedings of a Workshop. Florida Medical Entomology Laboratory,
Vero Beach, Florida.
SEIFERT, R. P. 1980. Mosquito fauna ofHeliconia aurea. J. Anim. Ecol. 49: 687-697.
WILTON, D. P. 1968. Oviposition site selection by the tree-hole mosquito, Aedes
triseriatus (Say). J. Med. Entomol. 5: 189-194.
INFLUENCE OF CORN PHENOLOGY AND PLANTING DATE
ON DAMAGE BY THE BLACK CUTWORM
L. A. RODRIGUEZ-DEL-BOSQUE' AND J. LOERA-GALLARDO
Campo Experimental Rio Bravo, INIFAP-SARH
Apartado Postal 172, Rio Bravo, Tamaulipas, M6xico 88900
Damage by the black cutworm, Agrotis ipsilon (Hufnagel), in relation to corn growth
stage and planting date, was studied in commercial cornfields near Rio Bravo,
Tamaulipas, Mexico, during the 1990, 1991, and 1992 spring seasons. Fields were sam-
pled during each of the 1- to 7-leaf stages to estimate the percentage of plants damaged
by A. ipsilon. Planting dates ranged from 15 January to 15 March. Regardless of
planting date, 81% of the attacks by A. ipsilon occurred during the first 3 leaf stages.
Fields planted during March sustained the highest damage, in contrast to the January
and February plantings, when damage was generally low.
Key Words: Agrotis ipsilon, Mexico, corn damage, agroecosystems.
Se estudi6 el efecto de la etapa de desarrollo del mafz y la fecha de siembra sobre
el dafno del gusano trozador, Agrotis ipsilon (Hufnagel), en parcelas comerciales de maiz
en Rio Bravo, Tamaulipas, M4xico, durante los ciclos de primavera de 1990, 1991, y
1992. Las parcelas fueron muestreadas en cada una de las etapas de 1 a 7 hojas para
estimar el porcentaje de plants dafadas por A. ipsilon. Las parcelas fueron sembradas
desde el 15 de Enero hasta el 15 de Marzo. El 81% del total de plants dafiadas se
observaron durante las etapas de 1 a 3 hojas, independientemente de la fecha de
siembra. El dafio mAs severe ocurri6 en las siembras de Marzo, en contrast con las
siembras en Enero y Febrero, cuando los niveles de dafio fueron en general bajos.
'Current address: Apartado Postal 133, TecomAn, Colima 28130, Mexico.
600 Florida Entomologist 76(4) December, 1993
Incidence of the black cutworm, Agrotis ipsilon (Hufnagel), is chronic in the corn
agroecosystem of northern Tamaulipas, Mexico (Rodriguez-del-Bosque 1986). Although
in most instances the presence of A. ipsilon does not represent an economic threat,
every year a number of fields need to be treated with insecticides or replanted as a
result of severe damage. Several factors are associated with damage by A. ipsilon,
including the phenologies of the crop and the insect (Showers et al. 1985). Most cutting
activity occurs when 4-6th instars feed on plants between the coleoptile and the 4-leaf
stages (Archer & Musick 1977, Clement & McCartney 1982, Showers et al. 1983, Whit-
ford et al. 1989). Plant regrowth and yield losses depend on the corn leaf stage at the
time of damage and on the position of the plant's growing point relative to the damage
(Showers et al. 1979, Levine et al. 1983, Story et al. 1983, Whitford et al. 1989).
Although some information on damage by A. ipsilon to corn has been reported for
northern Tamaulipas (Rodriguez-del-Bosque 1986), factors affecting the incidence and
dynamics of this insect pest are poorly known. The objective of this study was to
determine the influence of planting date and plant growth stage on damage to field corn.
MATERIALS AND METHODS
The study was conducted in commercial cornfields near Rio Bravo, Tamaulipas,
Mexico, during the 1990, 1991, and 1992 spring seasons. Fields were selected and
grouped according to planting date: (a) 15-31 January; (b) 1-15 February; (c) 16-29
February; and (d) 1-15 March (only in 1990 because SARH, the Mexican Department
of Agriculture, banned March plantings in 1991 and 1992 to reduce risk of contamination
by aflatoxin). Total number of fields studied (all planting dates) was 25, 25, and 33
during 1990, 1991, and 1992, respectively. None of the fields studied was treated with
insecticides. Ten samples of 20 m of corn row each (200 m total) were taken at random
within 1 ha in each field every 4 d to include each of the growth stages from 1 to 7
leaves (Ritchie et al. 1986). The total number of plants and plants recently severed by
A. ipsilon (< 24 h) were counted. These plants were normally damaged the night before
and the severed plant was nearby and still turgid. Plants damaged in previous stages
(> 24 h), showing a typical chlorotic regrowth, were not counted so as to avoid overes-
timating damage in the growth stage being sampled. For each sample, the percentage
of damaged plants was calculated. Data (arcsine transformed) were subjected to two-
way ANOVA with planting date and corn growth stage as factors. ANOVA was run
independently for each year after finding differences among years in the overall
analysis. Means were separated by Tukey's studentized range test (P<0.05) (SAS
RESULTS AND DISCUSSION
Damage differed significantly among corn leaf stages in all years (Table 1). Attack
was more frequent in the early plant stages and decreased progressively as corn de-
veloped (Fig. 1). Overall (3 years), 81% of the total plants damaged were found during
the 1- to 3-leaf stages. Similar results have been reported by Archer & Musick (1977),
Clement & McCartney (1982), Showers et al. (1983), and Whitford et al. (1989).
Damage in the March plantings was significantly higher in 1990 than in the earlier
plantings (Table 1). In all years, damage levels were low in the January and February
plantings. Evidently, the early plantings escaped heavy damage. The late plantings
(March) probably offer optimal climatic conditions (e.g., heat accumulation) for maximal
cutting activity, expected when the phenology of the plant (1- to 3-leaf stages) coincides
with that of the insect (4-6th instars). Similar results were reported in India by Kishore
& Misra (1984), who detected more damage by A. ipsilon and A. segetum (Schiff) to
Rodriguez & Loera: Damage by Black Cutworm to Corn 601
TABLE 1. DAILY DAMAGE BY AGROTIS IPSILON IN RELATION TO CORN GROWTH
STAGE AND PLANTING DATE, RIO BRAVO, TAMAULIPAS, MEXICO, 1990-
Planting % Plants Damaged Daily in Leaf Stage:
Date n 1 2 3 4 5 6 7 Total
-------------------- ---------------- 1990-----------------------------
0.39 0.46 0.61
0.42 0.52 0.28
0.33 0.23 0.16
3.13 2.44 1.42
1.la 0.9a 0.6ab
6 0.15 0.10 0.13 0
11 0.54 0.57 0.13 0.0
8 0.55 0.22 0.11 0
X 0.4a 0.3ab 0.lbc Oc
0 0 0
3 0.06 0 0
0 0 0
Oc Oc Oc
3 0.42 0.28 0.21 0.05
21 0.40 0.25 0.14 0.07
9 0.38 0.36 0.23 0.25
X 0.4a 0.3ab 0.2bc 0.1c
0.05 0.42 0.39
0.06 0.02 0
0.03 0.05 0
Oc 0.2c 0.1c
Row totals or column means followed by the same letter are not significantly different (P < 0.05; Tukey's
studentized range test).
3 4 5 6 7
CORN LEAF STAGE
1990 E 1991
Fig. 1. Damage by Agrotis ipsilon to corn in different growth stages, Rio Bravo,
Tamaulipas, Mexico, 1990-1992.
Florida Entomologist 76(4)
potatoes the later they were planted. They concluded that the use of an early planting
schedule would minimize cutworm damage and avoid the need for using insecticides.
Percentages of damage observed in this study were relatively low (Table 1). How-
ever, it is important to point out that these numbers represent the daily damage rate
(plants severed the night before) in a given corn leaf stage. In order to fully assess the
magnitude of damage by A. ipsilon in a given growth stage, the daily percentages
should be multiplied by 4, the average duration (d) of each of the 1- to 7-leaf stages,
assuming the daily damage rate was constant within a specific leaf stage. For instance,
the total damaged plants during the 1-leaf stage in the 1-15 March 1990 planting date
was estimated at 12.5% (3.13% x 4 d). For the same planting date and year, the
cumulative damaged plants in all stages (1-7 leaves) was calculated at 37.6% (9.4% x 4
d). Cumulative percentages of damaged plants in all stages (1-7 leaves) estimated for
the January and February planting dates were relatively low in all years, with a range
of 1.6 (15-31 January 1991) to 7.2 (15-31 January 1992).
In conclusion, the most frequent damage by A. ipsilon to corn occurred during the
1- to 3-leaf stages (first 12 d after plant emergence). Plants from early plantings may
escape heavy cutting activity of A. ipsilon. Growers in northern Tamaulipas who
planted corn as late as March should consider the increased risk of severe attack by A.
ARCHER, T. L., AND G. J. MUSICK. 1977. Cutting potential of the black cutworm on
field corn. J. Econ. Entomol. 70: 745-747.
CLEMENT, S. L., AND D. A. MCCARTNEY. 1982. Black cutworm (Lepidoptera: Noc-
tuidae): measurement of larval feeding parameters on field corn in the
greenhouse. J. Econ. Entomol. 75: 1005-1008.
KISHORE, R., AND S. S. MISRA. 1984. Impact of different planting dates on the inci-
dence of cutworms on potato crop. Indian J. Entomol. 46: 367-368.
LEVINE, R., S. L. CLEMENT, W. L. RUBINK, AND D. A. MCCARTNEY. 1983. Re-
growth of corn seedlings following injury at different growth stages by black
cutworm larvae. J. Econ. Entomol. 76: 389-391.
RITCHIE, S. W., J. J. HANWAY, AND G. O. BENSON. 1986. How a corn plant de-
velops. Iowa State University of Science and Technology Cooperative Extension
Service Special Report 48, Ames.
RODRIGUEZ-DEL-BOSQUE, L. A. 1986. Dafios del gusano trozador Agrotis ipsilon
(Hufnagel) en el cultivo de maiz en el norte de Tamaulipas. Agric. Tec. Mexico.
SAS INSTITUTE. 1988. SAS/STAT user's guide, release 6.03 edition. SAS Institute,
SHOWERS, W. B., R. E. SECHRIEST, F. T. TURPIN, Z. B. MAYO, AND G. SZATMARI-
GOODMAN. 1979. Simulated black cutworm damage to seedling corn. J. Econ.
Entomol. 72: 432-436.
SHOWERS, W. B., L .V. KASTER, AND P. G. MULDER. 1983. Corn seedling growth
stage and black cutworm (Lepidoptera: Noctuidae) damage. Environ. Entomol.
SHOWERS, W. B., L. V. MASTER, T. W. SAPPINGTON, P. G. MULDER, AND F. WHIT-
FORD. 1985. Development and behavior of black cutworm (Lepidoptera: Noc-
tuidae) populations before and after corn emergence. J. Econ. Entomol. 78: 588-
STORY, R. N., A. J. KEASTER, W. B. SHOWERS, J. T. SHAW, AND V. L. WRIGHT.
1983. Economic threshold dynamics of black and claybacked cutworms
(Lepidoptera: Noctuidae) in field corn. J. Econ. Entomol. 12: 1718-1723.
WHITFORD, F., W. B. SHOWERS, AND L. V. KASTER. 1989. Influence of actual and
manual black cutworm (Lepidoptera: Noctuidae) damage on recovery and grain
yield of field corn. J. Econ. Entomol. 82: 1773-1778.
Scheffrahn & Kfedek: New West Indian Parvitermes 603
PARVITERMES SUBTILIS, A NEW SUBTERRANEAN
TERMITE (ISOPTERA: TERMITIDAE) FROM
CUBA AND THE DOMINICAN REPUBLIC
RUDOLF H. SCHEFFRAHN1 AND JAN KECEEK2
'Ft. Lauderdale Research and Education Center
University of Florida, Institute of Food & Agricultural Sciences
3205 College Avenue, Ft. Lauderdale, FL 33314
2Insect Chemical Ecology Unit
Institute of Organic Chemistry and Biochemistry
Academy of Sciences of the Czech Republic
U Balamounky 41, 158 00 PRAHA 5
The soldier and worker caste of Parvitermes subtilis n. sp., from Cuba and the
Dominican Republic, are described for the first time. A key to soldiers of West Indian
species of Parvitermes sens. str. is provided.
Key Words: Nasutitermitinae, West Indies, soldier, worker.
Se described, por primera vez las castas de soldado y de obrero para Parvitermes
subtilis n. sp., de Cuba y la Repfblica Dominicana. Se present una clave para los
soldados de las species de Parvitermes sens. str. de las Indias Occidentales.
The genus Parvitermes Emerson was erected for an assemblage of Neotropical ter-
mite species characterized by small monomorphic or very slightly dimorphic nasute
soldiers with slightly constricted heads. Emerson's description (in Snyder 1949, pp.
376-377) was based mainly on the soldier caste and was comprised of six species which
were listed by Snyder (1949). Emerson included P. laticephalus (Snyder) from Bolivia,
but gave it only a tentative placement in Parvitermes, concerned that further collections
would reveal a distinct second soldier form. The other five species, P. brooks (Snyder),
P. discolor (Banks), P. flaveolus (Banks), P. pallidiceps (Banks), and P. wolcotti
(Snyder), all from the West Indies, were included outright. Mathews (1977) described
an additional species, P. bacchanalis Mathews, collected from Brazilian cerrado, and
noted some additional generic characters of the worker caste.
The imago caste of Parvitermes has never been diagnosed nor has the imago been
described for any species. Paradoxically, Snyder (1956) included a single, brief couplet
for Parvitermes spp. in his key for winged adults of West Indian termites.
In this paper, Parvitermes subtilis n. sp. is described from Cuba and the Dominican
Republic from soldiers and workers and compared with existing species of Parvitermes
sens. str. from the West Indies.
MATERIALS AND METHODS
Foraging groups (soldiers and workers) of P. subtilis n. sp. were collected in Sancti
Spiritus Province, Cuba, at Las Cuevas near Trinidad City (2233'N, 79053'W) on 14-
604 Florida Entomologist 76(4) December, 1993
XII-1973 by Luis de Armaz; in Guantanamo Prov. at Loma de la Herradura, Guan-
tAnamo City (20009'N, 75012'W) on 8-VIII-1974 by J. Kfe6ek; in Santiago de Cuba Prov.
at Playa Siboney (19057'N, 75040'W) on 2-VII-1966 by I. Hrdy, and on 5-XI-1971 by J.
Kfedek; and at Castillo del Morro near the City of Santiago de Cuba (1958'N, 75052'W)
on 23-V-72 by J. Kfecek. Collections in the Dominican Republic were made at La
Guajaca (19042'N, 7142'W), Monte Cristi Prov., 19-VI-1991; 24 km E. Azua (18"25'N,
7126'W), Azua Prov., 26-II-92; Caracoles (18026'N, 71023'W), Azua Prov., 27-II-92;
and Las Lavas(19034'N, 70049'W), Santiago Prov., 8-VI-92. Foragers were collected by
aspirator and field-preserved in 85% ethanol.
Measurements of specimens, made with a calibrated ocular micrometer on an Olym-
pus SZH microscope, follow those defined by Roonwal (1970). Terms used to describe
soldier morphology and color follow those of Sands (1965). Scanning electron micro-
graphs were made with a Hitachi S-4000 field emission microscope (10kV) of two P.
subtilis n. sp. soldiers dehydrated in absolute ethanol and 1,1,1,3,3,3-hexamethyl-
disilazane (Nation 1983) and then sputter coated with gold. Fifty soldiers, including 36
from the Dominican Republic and 14 from Cuba, and 21 workers, including 15 from the
Dominican Republic and 6 from Cuba, representing the later six localities above, were
used for measurements.
The holotype soldier from Caracoles, Dominican Republic, will be deposited in the
collection of the National Museum of Natural History, Washington, D.C. Paratype
soldiers will be deposited in the Florida State Collection of Arthropods, Fla. Dept.
Agric. Cons. Serv., Division of Plant Industries, Gainesville, Florida, at the institutions
of the first and second authors (in Ft. Lauderdale and Prague, respectively), the In-
stituto de Ecologia y Sistematica, Academia de Ciencias de Cuba, and at the Univer-
sidad de Santo Domingo, Dominican Republic.
PARVITERMES SUBTILIS SCHEFFRAHN AND KRECEK, NEW SPECIES
Soldier (Figs. 1-2). Head capsule pale yellow, nasus deeper yellow; antennae and
pronotum yellow-white; remainder of body whitish.
Head capsule in dorsal view with very slight constriction behind antennae; in lateral
view, vertex raised in front of and behind constriction above line of nasus. Nasus,
vertex, and sides of head densely covered with several hundred fine, parallel, short,
and anterior-leaning (450) setae; setae on head of equal length, shorter on nasus; head
surface anterior to constriction with four, and posterior to constriction with 2-4 more
erect setae twice as long as other setae on head.
Antennae 12-segmented; second, third, and fourth subequal in length; third narrowest.
Mandibles without points. Nasus slender and nearly cylindrical; projecting straight
forward in dorsal plane of head. Tibia thin, much shorter than head length with nasus.
Measurement in mm (n= 50) Range Mean SD Holotype
Head length with nasus 1.06-1.23 1.130.041 1.11
Head length without nasus 0.66-0.74 0.70 0.021 0.70
Head width, maximum 0.60-0.68 0.64 0.023 0.65
Nasus width at base 0.13-0.16 0.14 0.007 0.14
Nasus width at middle 0.063-0.075 0.0670.004 0.063
Head height, maximum 0.40-0.50 0.44 0.026 0.46
Pronotum width 0.33-0.40 0.360.018 0.36
Pronotum length, maximum 0.14-0.19 0.160.013 0.16
Hind tibia length 0.64-0.78 0.69 0.042 0.66
Scheffrahn & Kfe6ek: New West Indian Parvitermes
Figs. 1-2. Scanning electron micrographs of Parvitermes subtilis n. sp. soldier. 1)
Dorsal (Caracoles, Dominican Republic) and 2) lateral (Playa Siboney, Cuba) views of
606 Florida Entomologist 76(4) December, 1993
Key to Soldiers of West Indian Parvitermes Emerson sens. str.
1. Head length including nasus 1.23-1.42 mm; head and nasus pale brown (sometimes
with lighter yellowish pigmentation in posterior third of head) .............. discolor
- Head length including nasus 1.05-1.23 mm; head pale yellow with yellow nasus or
head pale ferruginous orange with concolorous or chestnut brown nasus ........ 2
2. Head pale yellow, nasus more yellow; nasus slender, nearly cylindrical, median
width 0.063-0.075 mm; mandibles without points; fine parallel setae on vertex
forming dense, well-combed mat ............................................. subtilis n. sp.
- Head pale ferruginous orange; nasus concolorous or chestnut brown; nasus
more conical, median width 0.073-0.100 mm; mandibles usually with points;
setae on vertex variable in orientation, length, and density ......................... 3
3. Nasus concolorous with head or only very slightly darker; some setae on vertex
nonparallel, 16-22 setae cresting along a 0.2 mm length of vertex horizon flaveolus
- Nasus chestnut brown; setae on vertex clearly irregular in length when viewed
laterally, 8-14 setae cresting a 0.2 mm margin of vertex ...................... brooks
Worker. Head and antennae yellow-white, remainder of integument unpigmented.
Head, including postclypeus, covered with about 100 fine setae of varied length.
Antennae 13-segmented; second and fourth subequal, third shortest, narrowest, and
least pigmented. Postclypeus twice as wide as long. Tibia generally more slender and
short compared to other Parvitermes workers.
Molar plate of right mandible with 6 ridges; apical tooth similar to first marginal tooth.
Measurement in mm (n = 21) Range Mean SD
Head width, maximum 0.63-0.88 0.790.072
Head length to postclypeus 0.56-0.86 0.730.079
Postclypeus width 0.26-0.35 0.33 0.024
Postclypeus length 0.15-0.19 0.17 0.014
Hind tibia length 0.63-0.75 0.670.037
Hind tibia width, maximum 0.054-0.075 0.064 0.006
Comparisons. All workers of P. subtilis n. sp. examined (n=58) have 13 antennal
segments versus the usual 14 segments for other Parvitermes workers. Mean head and
hind tibia widths are also the smallest among West Indian Parvitermes.
Etymology. The Latin "subtilis" describes the fine, slender, and rather delicate
nature of the headcapsule setae, nasi, tibia, and delicate bodies of the soldiers, and legs
and bodies of workers of this species.
Parvitermes subtilis foragers were collected in and under dried ruminant dung on
soil, under stones, and under thin soil sheeting covering dried grasses. In Cuba, P.
subtilis were collected sympatrically with P. brooks, at Las Cuevas and El Morro, and
with Nasutitermes rippertii (Rambur) at Siboney. In collections from the Dominican
Republic, P. subtilis foragers were aspirated from soil concurrently with two sympatric
species; once with P. flaveolus foragers (24 km E. Azua), and once with Anoplotermes
sp. workers (Las Lavas). Localities where P. subtilis have been collected indicate that
this species is adapted to dry tropical forest and scrub habitats of 1000 mm or less of
Scheffrahn & Kkedek: New West Indian Parvitermes
annual precipitation where dried herbaceous growth is eaten. As with other Par-
vitermes spp., P. subtilis does not build mounds but forages in galleries of irregular
dimension beneath stones or other surface debris. The underground nest structure of
P. subtilis is unknown.
Parvitermes subtilis is a relatively rare small nasute species in both Cuba (5/90
samples of Parvitermes and Obtusitermes) and the Dominican Republic (4/162 samples
of Parvitermes and Velocitermes). Extensive collections in recent years indicate that
P. subtilis is the only small nasute species found on both Cuba and Hispaniola. Snyder's
(1956) listing of P. discolor in Cuba and P. brooks on Bimini Island (Bahamas) are very
Recent collections have shown that P. pallidiceps from Hispaniola (V. Spaeth, un-
publ. thesis; Scheffrahn, unpubl. data) and P. wolcotti from Puerto Rico (Scheffrahn
and S. C. Jones, unpubl. data) have a distinctly dimorphic soldier caste and, therefore,
should not be included in Parvitermes sens. str. The Hispaniolan P. pallidiceps fits well
into Velocitermes, while the generic placement of P. wolcotti from Puerto Rico remains
to be determined.
We are indebted to J. A. Chase, J. de la Rosa G., and J. R. Mangold for contributing
to the collection of P. subtilis n. sp. in the Dominican Republic; D. S. Williams of the
ICBR Electron Microscope Core Facility at the University of Florida, Gainesville, for
technical assistance with electron microscopy; M. S. Collins, Smithsonian Institution,
for loan of identified specimens; and J. A. Chase, M. S. Collins, R. M. Giblin-Davis, J.
R. Mangold, and N.-Y. Su for critically reviewing and improving this contribution no.
R-03138 of the University of Florida Experiment Stations Series.
MATHEWS, A. G. A. 1977. Studies on termites from the Mato Grosso State, Brazil.
267 pp. Academia Bras. de Ciencias, Rio de Janeiro.
NATION, J. A. 1983. A new method using hexamethyldisilazane for the preparation of
soft insect tissue for scanning electron microscopy. Stain Technol. 55: 347-352.
ROONWAL, M. L. 1970. Measurements of termites (Isoptera) for taxonomic purposes.
J. Zool. Soc. India 21: 9-66.
SANDS, W. A. 1965. A revision of the termite subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Bull. British Mus. Nat. Hist., Entomol.
Suppl. 4: 1-172.
SNYDER, T. E. 1949. Catalog of the termites (Isoptera) of the world. Smithson. Misc.
Collect. 112: 1-490.
608 Florida Entomologist 76(4) December, 1993
A NEW SPECIES OF PITYOPHTHORUS EICHHOFF
(COLEOPTERA: SCOLYTIDAE) FROM SOUTHERN FLORIDA
WITH A KEY TO THE FLORIDA SPECIES
T. H. ATKINSON
Entomology & Nematology Department
University of Florida, Gainesville, FL 326111
Pityophthorus pecki, a new species, is described from southern Florida. It is related
to Neotropical species from Mexico and Central America in Bright's juglandis group.
A key to the species of Pityophthorus in Florida is included.
Key Words: Bark beetle, taxonomy.
Pityophthorus pecki, especie nueva, se describe del sur de Florida. Es emparentada
a species neotropicales de Mexico y CentroamBrica del grupo juglandis de Bright. Se
incluye una clave a las species de Pityophthorus en Florida.
While examining Scolytidae collected by Dr. Stewart Peck as part of an ongoing
survey of the Coleoptera of southern Florida (Peck 1989), I discovered a previously
undescribed species of Pityophthorus. This species is described here and an illustrated
key is provided for the Florida Pityophthorus. Previous keys (Bright 1981, Wood 1982)
are difficult to use for Florida because they include a very large number of species from
extensive geographic areas. An additional species, Pityophthorus pinavorus Bright, was
described since the publication of these keys (Bright 1985b).
This genus Pityophthorus Eichhoff contains nearly 400 described species, the bulk
of which are known from North and Central America, with smaller numbers of species
in the Caribbean, South America, Eurasia and Africa (Bright 1981, Wood & Bright
1993). Most species breed in conifers, but many neotropical species breed in broad-
leaved trees, shrubs, and vines. Most species are phloeophagous (breed in phloem, or
inner bark) and harem polygynous (gallery systems initiated by males which are later
joined by several females), although there are an appreciable number with other breed-
The following key includes the twelve species known from Florida. It is loosely
based on the key by Bright (1981), but does not specifically indicate his species groups
because of the small number of taxa included here.
Key to species of Pityophthorus of Florida
1. Asperities on anterior slope of pronotum arranged into 2 or more definite
concentric rings (Fig. 1 A, D). Not in coniferous hosts ........................... 2
- Asperities on anterior slope of pronotum showing no indication of concen-
tric rows (Fig. 3 C, 4 A). In coniferous hosts ....................................... 7
'Current address: Department of Entomology, University of California, Riverside, California 92521.
Atkinson: New Pityophthorus from Southern Florida
Fig. 1. Pityophthorus pecki, male. A. Dorsal view. B. Declivity. C. Frons. P. con-
centralis D. Dorsal view. E. Declivity. F. Frons. White lines represent 0.5 mm in A,
B, D, E, 0.05 mm in C, F.
2 (1). Interstriae 2 on declivity as wide as on disc, distinctly impressed and flat,
punctures on striae 1 and 2 distinct on declivity, usually equal in size to
those on disc (Fig. 1 D, E) ........................................... ............. 3
Interstriae 2 on declivity narrower than discal width, not impressed,
punctures on striae 1 and 2 indistinct on declivity, smaller than those on
disc (Fig. 1 A, C). Southern Florida. 1.3-1.5 mm ............... pecki Atkinson
Florida Entomologist 76(4)
Fig. 2. Pityophthorus lautus. A. Dorsal view. B. Frons. C. Lateral view. D. Decliv-
ity. White lines represent 0.5 mm in A and C, 0.05 mm in B and D.
3 (2). Granules on interstriae 1 smaller than those on interstriae 3, setae on
interstriae 3 shorter than width of interstriae, spaced within rows by a
distance greater than their length (Fig. 1 E, D). Male and female frons
similar, pubescence sparse ................................................................ 4
Granules on both interstriae 1 and 3 large, similar in size, setae on in-
terstriae 3 longer than width of interstriae, spaced within row by a dis-
tance less than their length. Male and female frons sexually dimorphic,
female frons pubescent, male frons with sparse pubescence ................... 6
4 (4). Declivital interstriae 2 with row of setiferous punctures (Fig. 1 D, E).
Southern Florida and Cuba. In Metopium toxiferum (poisonwood, Anacar-
diaceae). 1.2-1.5 mm ............................................. concentralis Eichhoff
- Declivital interstriae 2 impunctate, without setae (Fig. 2 A, D) .............. 5
5 (4). Frons flattened or weakly, transversely concave to upper level of eyes,
divided by weak longitudinal carina. Widespread in eastern North
America (probably a species complex). Many hosts (principal host in penin-
sular Florida: Toxicodendron radicans, poison ivy, Anacardiaceae).
1.3-1.6 mm ................................................................... lautus Eichhoff
Frons convex, rugose, with small elongate callus at upper level of
eyes. Southern Florida. In Borrichia spp. (Compositae). 1.0-1.3 mm
............................................................................... borrichiae W ood
Atkinson: New Pityophthorus from Southern Florida
6 (3). First 2 segments of antennal club occupy more than half of club length, club
1.4 or less times as long as wide; lower half of female frons distinctly punc-
tured. Color reddish brown. Southeastern U.S. In Liquidambar styraciflua
(sweet gum, Hamamelidaceae). 1.3-1.5 mm .......... liquidambarus Blackman
First 2 segments of antennal club about 1/3 of club length, club 1.5 times
as long as wide; lower half of female frons smooth, shining. Color black.
Southeastern U.S. In Toxicodendron radicans (poison ivy, Anacardiaceae).
1.3-1.6 mm ............................................................ crinalis Blackman
7 (1). Elytral apex rounded (Fig. 3 C) ........................................... ........ 8
Elytral apex acuminate (Fig. 4 A, 6 A-D) ........................................... 9
8 (7). Pronotum evenly arched in profile without strongly elevated summit (Fig.
3 A), strial and interstrial punctures on disc confused (Fig. 3 C). Southeastern
U.S., Caribbean. In pines. 1.3-2.0 mm .............. pulicarius (Zimmermann)
Prontum with distinctly elevated summit; striae distinct on disc, inter-
strial punctures uniseriate. Peninsular Florida, Caribbean. 1.2-1.4 mm
......... ..................... ........................................ pinavorus Bright
9 (7). Pronotum with distinct groove on posterolateral margin (Fig. 4 B, D, 5 D) 10
Pronotum without groove on posterolateral margin (Fig. 5 A-C) ............. 11
10 (9). Female frons deeply concave over very broad area, central area deeply punc-
tured, long, incurved setae on periphery not obscuring central area (Fig.
Fig. 3. Pityophthorus pulicarius. A. Lateral view, female. B. Frons, male. C. Dor-
sal view, male. D. Frons, female. White lines represent 0.5 mm.
Florida Entomologist 76(4)
Fig. 4. Pityophthorus confusus. A. Dorsal view, female. B. Lateral view, female.
C. Lateral view of head and prothorax, male. D. Lateral view of head and prothorax,
female. E. Ventral view of head, female. F. Protibia. Posterior view, left; anterior
view, right. White lines represent 0.5 mm in A-E, 0.05 mm in F.
4 D); male frons shallowly concave, setae sparse (Fig. 4 C). Southeastern
U.S. to Central America. In pines. 2.0-2.9 mm .... confusus bellus Blackman
Female frons weakly concave or flat; long incurved setae on periphery dense,
obscuring central area (Fig. 5 D); male frons transversely impressed. South-
eastern U.S. to Central America. In pines. 1.4-1.7 mm .. annectens LeConte
11 (9). Interstriae on disc impunctate and without setae; uniseriate rows of erect
Atkinson: New Pityophthorus from Southern Florida 613
interstrial setae in interstriae 1, 3, 5, 7 on declivity (Fig. 6 B). Central
Canada, eastern North America south to Alabama and Florida. In pines.
1.5-2.0 mm ............................................................ consimilis LeConte
Some interstriae on disc with setae; uniseriate rows of interstrial setae on
all declivital interstriae (Fig. 6 A). Southeastern U.S. from Massachussetts
to Florida. In pines. 2.2-2.6 mm ............................. pullus (Zimmermann)
PITYOPHTHORUS PECKI ATKINSON, NEW SPECIES
(Fig. 1 a-c)
DIAGNOSTIC CHARACTERS. This species belongs to Bright's juglandis group
(Bright 1981) and is the only species from that group known from the southeastern
United States. Members of this species group are characterized by having the pronotal
asperities arranged in concentric rings, a rounded declivital apex, obsolete or reduced
punctures on declivital striae 2, and declivital interstriae 2 which are not impressed.
All other species known from the region with the pronotal asperities arranged in concen-
tric rings and having rounded declivital apices belong to Bright's lautus group in which
the punctures on striae 2 on the declivity are not reduced in size and the second declivital
interstriae are flattened and impressed with respect to interstriae 1 and 3.
Fig. 5. Head and prothorax of Pityophthorus spp. A. P. consimilis, female. B. P.
consimilis, male. C. P. pullus, male. D. P. annectens, female. White lines represent
Florida Entomologist 76(4)
This species would key to P. strictus Wood in published keys (couplet 6 in Bright
1981, couplet 7 in Bright 1985a). Pityophthorus strictus is known from Costa Rica and
breeds in Rheedia sp. (Guttiferae), a host which does not occur in Florida. P. strictus
has minute granules on declivital interstriae 3 which are absent in P. pecki. Interstriae
2 of P. strictus are equally wide on the disc and declivity, while they are clearly nar-
rower on the declivity than on the disc in P. pecki.
MALE. Length: 1.3-1.5 mm, 2.4 times longer than wide. Color dark brown.
Frons concave, rising to highest point slightly above upper level of eyes, with im-
punctate line (not elevated) from this summit to vertex, slight transverse impression
above epistoma. Surface reticulate between deep punctures, distance separating
punctures approximately equal to their diameters. Fine, short setae arising from
punctures. Antennal club oval, approximately twice as long as wide, with 3 slightly
Pronotum parallel-sided, broadly rounded in front, 1.1 times wider than long. As-
perities on anterior portion arranged in 3-5 concentric rings. Clearly marked summit
visible in lateral view. Postero-lateral areas glossy, strongly punctured, punctures sepa-
Fig. 6. Declivities of Pityophthorus spp. A. P. pullus, male. B. P. consimilis,
male. C. P. confusus, female. D. P. confusus, male. White lines represent 0.5 mm.
Atkinson: New Pityophthorus from Southern Florida
rated by distance approximately equal to diameters. Midline impunctate from base to
Elytra 1.6 times as long as wide, apex broadly rounded. Striae clearly marked on
disc by rows of fine punctures, not impressed. Strial punctures shallow, separated
within rows by distance equal to twice the diameters, without associated setae on disc.
Discal interstriae flat, impunctate, about twice as wide as striae, surface glossy.
Declivity convex, occupying posterior third of elytra, evenly curved to base. Striae
1 impressed, punctures scarcely distinguishable. Sutural interstriae slightly elevated,
finely, uniseriately granulate to apex. Interstriae 2 not impressed, narrower than on
disc, impunctate. Punctures of striae 2 reduced. Uniseriate setae on all declivital in-
terstriae except 1, setae hairlike, length approximately 2.5 times width of interstriae.
FEMALE. Not present or not distinguishable in material at hand. Two specimens,
both males, were sexed by dissection.
TYPE MATERIAL. This description is based on 18 specimens. The holotype male
bears the following labels: "Florida: Monroe Co.: Big Pine Key, 4-VI-86, S. & J. Peck,
beating vegetation" and "HOLOTYPE: Pityophthorus pecki Atkinson 1993" and is de-
posited in the U.S. National Museum of Natural History, Washington, D.C. Paratypes,
listed below, bear yellow labels and are deposited in the Florida State Collections of
Arthropods, Gainesville, Canadian National Collection, Canadian Museum of Nature,
Ottawa, Ontario, and in my personal collection. Florida: Dade Co.: S. Miami, Deering
Estate park, 21-11-86 to 1-VI-86, S. &. J. Peck, malaise trap (CMNC, 1); Monroe Co.:
Big Pine Key, 2-VI-86, S. & J. Peck, beating vegetation, (CMNC, 3); 4-VI-86, same
data (THAC, 6); Big Pine Key, 7-VI-86, J. Browne (CNCC, 3); Big Pine Key, Watson's
Hammock, 3-VI-86 to 27-VIII-86, S. & J. Peck, malaise-flight intercept trap (FSCA,
2); Watson's Hammock, 14-XII-86, Klimaszewski & Peck, hammock litter (FSCA, 1);
Watson's Hammock, 20-V-90, R. S. Anderson (CMNC, 1); No Name Key, 5-VI-86, M.
Kaulburs, night beating (USNM, 2). (Collection abbreviations from Arnett & Samuelson
DISTRIBUTION. Known only from Monroe and Dade Counties in southern Florida.
HOSTS. Unknown. All specimens seen were collected in flight traps. Hosts of species
in the juglandis group of Bright (1981, 1985a) are from a variety of plant families
including the Guttiferae, Burseraceae, Apocynaceae, Anacardiaceae, Compositae, and
Juglandaceae. The most likely host of P. pecki is Bursera simaruba (gumbo limbo).
This speculation is based on the known distribution of P. pecki, the potential host plants
in southern Florida from those families with similar distributions (Long & Lakela 1971),
and the observation that most of the species most closely related to it breed in species
ETYMOLOGY: This species is named in honor of Stewart Peck, the collector of most
of the known specimens, for his contributions to the knowledge of the beetle fauna and
biogeography of southern Florida.
Travel to Provo, Utah to visit the S. L. Wood collection was supported by a grant
from the American Philosophical Society. Scanning electron microscopy was done with
the facilities of the Electron Microscope Core Facility, IFAS ICBR, University of
Florida, and the assistance of associated staff. This is Florida Agricultural Experiment
Station Journal Series No. R-03194.
ARNETT, R., AND P. A. SAMUELSON. 1986. The Insect and Spider Collections of the
World. Brill/Flora and Fauna Publications. Gainesville, Florida.
Florida Entomologist 76(4)
BRIGHT, D. E. 1981. Taxonomic monograph of the genus Pityophthorus Eichhoff in
North and Central America (Coleoptera: Scolytidae). Mem. Entomol. Soc.
Canada No. 118.
BRIGHT, D. E. 1985a. New species and new records of North American Pityophthorus
(Coleoptera: Scolytidae), Part V: the juglandis group. Great Basin Nat. 45: 476-
BRIGHT, D. E. 1985b. Studies on West Indian Scolytidae (Coleoptera) 3. Checklist of
the Scolytidae of the West Indies, with descriptions of new species and taxonomic
notes. Entomol. Arb. Mus. Frey 33/34: 169-187.
LONG, R. W., AND O. LAKELA. 1971. A flora of tropical Florida: a manual of the seed
plants and ferns of southern peninsular Florida. Univ. Miami Press, Coral Ga-
PECK, S. B. 1989. A survey of insects of the Florida Keys: post-Pleistocene land-bridge
islands: Introduction. Florida Entomol. 72: 603-612.
WOOD, S. L. 1982. The bark and ambrosia beetles of North and Central America (Col-
eoptera: Scolytidae), a taxonomic monograph. Great Basin Nat. Mem. 6: 1-1356.
WOOD, S. L., AND D. E. BRIGHT. 1993. A catalog of Scolytidae and Platypodidae
(Coleoptera), Part II: Taxonomic index. Great Basin Nat. Mem. 13: 1-1553.
ARTHROPODS ASSOCIATED WITH
IN MORELOS, MEXICO
MIRNA GUTIERREZ OCHOA, MARIO CAMINO LAVIN,
FEDERICO CASTREJON AYALA, AND ALFREDO JIMENEZ PEREZ
Departamento de Entomologia
Centro de Desarrollo de Productos Bi6ticos
Institute Politecnico Nacional
Apartado Postal 24, Yautepec, Morelos, Mexico
A survey was conducted of the arthropod fauna associated with Bromelia hemis-
phaerica Lamarck (Bromeliaceae) at Yautepec, Morelos, Mexico. In this survey, con-
ducted from January to September 1989, 40.3 percent of the species represented in the
collections were predatory, belonging to the orders Araneae, Acarina (Parasitiformes
and Acariformes), Hemiptera, Coleoptera and Hymenoptera. Herbivorous species ac-
counted for 16.4 percent of the total species collected. Among these were Ferrisia
virgata (Cockerell) and Dysmicoccus brevipes (Cockerell), which might become pests
when these plants are cultivated in monoculture.
Key Words: Bromeliads, ecological niche, feeding habits, monoculture.
Se hicieron colectas de artropodos que se alimentan o habitan de la plant Bromelia
hemisphaerica Lamarck (Bromeliaceae) en Yautepec, Morelos, M6xico. Los muestreos
fueron conducidos de Enero a Septiembre de 1989. Los resultados mostraron que el 40.3
Gutigrrez et al.: Arthropods of Bromelia hemisphaerica
por ciento de las species colectadas fueron de habitos depredadores, perteneciendo a
los ordenes Araneae, Acarina (Parasitiformes y Acariformes), Hemiptera, Coleoptera
e Hymenoptera. Las species herbivoras fueron el 16.4 por ciento del total de las es-
pecies colectadas, dentro de ellas Ferrisia virgata (Cockerell) y Dysmicoccus brevipes
(Cockerell) pueden presentarse como plagas si se hace monocultivo.
Bromelia hemisphaerica Lamarck (Bromeliaceae) is native to parts of Mexico and
Central America. It produces hemisphericin, a proteolytic enzyme, and its cultivation
in monoculture is planned at the Centro de Desarrollo de Productos Bi6ticos, Yautepec,
Morelos, Mexico. Nothing has been published about the arthropods associated with it.
Therefore, we undertook a survey of associated arthropods in a study site where it has
been growing undisturbed for 5 years. We thought the survey would provide a useful
background for the control of arthropod pests that might attack it when it is grown in
MATERIALS AND METHODS
The study site was an area of 2,500 m2 at Yautepec, where B. hemisphaerica plants
grow. For the last five years the plants have been growing under neglected conditions;
they were planted and left without any cultivation. They were surrounded by some
other cultivated plant species, including Carica papaya L., Leucopremna mexicana
(A.DC.) Stand. and Cnidoscolus chayamansa McVaugh. The area has a semitropical
climate, with two seasons: dry season occurs from December to mid-June, and wet
season from mid-June to November. Annual rainfall averages 900 mm, and the study
area is at 970 m altitude (Anonymous 1980).
Twenty-six random samples were taken, each of three plants, from January to Sep-
tember. The plants selected were inspected, and arthropods found were placed in vials
with 70% ethanol. The plants were cut and placed into plastic bags and transported to
the laboratory, where they were inspected again. Then, the inflorescences were placed
in water in buckets and left there for 24 h, after which arthropods were picked out of
the water and preserved in 70% ethanol. The remaining non-living plant material left
inside the plastic bags was placed into Berlese funnels to collect additional arthropods.
Arthropods were separated to orders and families and were sent to taxonomists for
identification. Representative arthropods, identified and unidentified, were placed in the
zoology collection of the University of Morelos State in Cuernavaca, Morelos.
RESULTS AND DISCUSSION
Some of the arthropods collected were identified to genus or species level (Table 1).
Fifty-two species were recognized, though some of them were identified only to the
generic level. Some specimens were determined to family, superfamily or order only.
The feeding behavior was classified as far as possible (Table 2).
The Mexican flora contains 364 species of bromeliads in 44 genera (Garcia-Franco
1986). Six of the species are known to have high concentrations of proteolytic enzymes
(Gardufio et al. 1974). The plant of current interest belongs to the genus Bromelia,
whose arthropod fauna was practically unknown.
De Buen (1953) reported mosquito larvae from leaf axils of B. plumeri (E. Morren)
(as B. karatas L.), and mosquitoes are the best-studied arthropod associates of
bromeliads in general (Downs & Pittendrigh 1946, Frank 1990). Bromeliads of many,
though far from all, genera and species are capable of impounding water in leaf axils,
Florida Entomologist 76(4)
TABLE 1. ARTHROPODS COLLECTED FROM BROMELIA HEMISPHAERICA.
Taxonomic position abundance'
Buthidae: Centruroides limpidus (Karsch) P-V.F.
Oxyopidae: Peucetia viridans (Hentz) P-V.F.
Scytodidae: Scytodes sp. P-V.F.
Salticidae: Marchena sp. P-V.F.
Chalcoscirtus sp. P-V.F.
Dictynidae: Argena sp. P-V.F.
Mallos sp. P-V.F.
Caponiidae: Tarsonops sp. P-V.F.
Oonopidae: Scaphiella sp. P-V.F.
Theridiidae: Steatoda sp. P-V.F.
Hersiliidae: Tama sp. P-V.F.
Clubionidae: Strotarchus sp. P-V.F.
Anyphaenidae: Aysha sp. P-V.F.
Sparassidae and Homalonychidae P-V.F.
Macrochelidae: Macrocheles muscaedomesticae (Scopoli) N-V.F.
Acaridae: Tyrophagus butrescentiae (Schrank) P-V.F.
Neanuridae: Brachystomella barrerai
Palacios-Vargas & Najt N-V.Ma.
Entomobryidae: Seira sp. N-V.Ma.
Blattidae: Parcoblatta sp. O-Mo
Pyragridae: Pyragra sp. O-V.F.
Forficulidae: Forficula sp. O-V.F.
Liposcelidae: Liposcelis bostrychophilus Badonnel N-Ma
Belaphotroctes sp. N-Ma
Psoquillidae: Rhyopsocus pescadori Garcia-Alderete N-R
Noctuidae (3 species) N-R
"microlepidoptera" (2 species) N-R
Culicidae, Chironomidae, and Syrphidae N-V.F.
Phlaeothripidae: Apterygothrips n. sp. (Fide R. Johansen) F-V.F.
Hoplandrothrips sp. F-V.F.
Gutigrrez et al.: Arthropods of Bromelia hemisphaerica 619
TABLE 1. (Continued)
Taxonomic position abundance'
Phopasidae: Jadera sp. N-V.F.
Coreidae: Anasa scorbutica (Fabricius) H-V.F.
Pyrrhocoridae: Dysdercus nimulus Hussey H-V.F.
Lygaeidae: Ligyrocoris sp. H-V.F.
Blissus sp. H-V.F.
Xilocoridae: Xilocoris sp. P-V.F.
Exetochiomera sp. H-V.F.
Anthocoridae: Anthocoris sp. P-V.F.
Orius sp. P-V.F.
Reduviidae: Zelus sp. P-V.F.
Cydnidae: Annestus sp. N-V.F.
Atrazonotus sp. N-V.F.
Pseudococcidae: Dysmicoccus brevipes H-V.Ma.
Ferrisia virgata H-V.Ma.
Histeridae: Omalodes sobrinus Erichson N-V.F.
Anthicidae: Anthicus asphaltinus Champion N-V.F.
Nitidulidae: Carpophilus sp. P-V.F.
Calopterus sp. N-V.F.
Coccinellidae: Scymnus sp. P-V.F.
Carabidae: Lebia sp. P-V.F.
Phloeoxena sp. P-V.F.
Colydiidae: Lasconotus sp. H-V.F.
Chrysomelidae: Pachybrachys sp. H-V.F.
Mordellidae: Naucles sp. N-V.F.
Tenebrionidae: Stibia sp. N-V.F.
Cucujidae: Laemophloeus sp. N-V.F.
Nausibius sp. P-V.F.
Neuroptera (2 species) P-R
Eupelmidae, Encyrtidae, Mymaridae,
Braconidae and Chalcididae. Pa-V.F.
Formicidae: Odontomachus clarus Roger P-V.Ma.
Crematogaster brevispinosa Mayr O-V.Ma.
Camponotus sp. O-V.Ma.
Gnamtogenys sp. P-V.Ma.
Amblyopone sp. P-V.Ma.
Aphaenogaster sp. O-V.Ma.
Tetramorium sp. O-V.Ma.
Strumigenys sp. O-V.Ma.
Leptothorax sp. O-V.Ma.
Pheidole sp. O-V.Ma.
Monomorium sp. O-V.Ma.
Brachymyrmex sp. O-V.Ma.
'Ecological niche-abundance key.
H = Herbivore; F = Fungivore; P = Predator; 0 = Omnivorous; Pa. = Parasitoid; N = Not known.
V. Ma. = Very Many; Ma. = Many; Mo. = Moderate; F = Few; R = Rare.
Florida Entomologist 76(4)
TABLE 2. PERCENTAGES OF IDENTIFIED SPECIMENS LIVING OR FEEDING ON
Feeding Behavior Genus/Species Level Family Level
Herbivores 16.4 13.0
Fungivores 6.0 13.0
Detritivores 16.4 8.6
Predators 40.3 40.0
Omnivores 14.9 3.6
Unknown 6.0 -
and the phytotelmata thus formed, with their associated aquatic fauna, have fascinated
biologists for nearly a century (Calvert 1911, Alexander 1912, Laessle 1961, Frank 1983).
The non-aquatic fauna of bromeliads has been the subject of far fewer studies
(Biezanko 1961, Beutelspacher 1969, Benzing 1970, Palacios & Vargas 1981, Murillo &
Palacios 1983). A few agricultural textbooks contain information on pests of pineapples.
A few horticultural textbooks specifically on bromeliads contain a little information on
pests of ornamental bromeliads. There is no comprehensive review of the non-aquatic
arthropod fauna of bromeliads.
In Table 2 we have tried to classify the feeding habits of the arthropods from B.
hemisphaerica at our study site in Morelos. The classification is at two levels: first for
species for which we had identifications at least to the generic level, and second to
include also those species identified only to the family level.
About 16.4% of the genera and species are herbivorous. Two of these species are
known as pests of tropical plants; the scale insects, Ferrisia virgata (Cockerell) and
Dysmicoccus brevipes (Cockerell), may cause pest problems in a monoculture of B.
hemisphaerica. It is also possible that some of the Noctuidae could be pests, but these
have not been identified.
Most of the species collected were predators belonging to Scorpionida, Araneae,
Acarina, Hemiptera, Coleoptera, and Hymenoptera. The preponderance of predators
may be due to the wild conditions of the plants and may change in a monoculture crop.
The results suggest that B. hemisphaerica may be a good refuge for predators.
An undescribed species of Thysanoptera, which may be fungivorous, was found.
Loomis (1969) and Palacios & Vargas (1981) indicate that new species of miriapods and
Collembola, respectively, may be found in bromeliads. Omnivorous species accounted
for 14.9% of the identified genera and species, including ants that may be nesting in the
leaf axils. Additional work on the biology of these ants is needed.
Numerous species and individuals of aquatic Diptera occur in bromeliad leaf axils in
general, and mosquito larvae are among the most abundant of the Diptera (Frank 1983).
Adults of some of these mosquitoes transmit diseases to man (Downs & Pittendrigh
1946, De Buen 1953, Frank 1983, 1990). The experimental area of the present work is
in an area of malaria distribution in Mexico, so it will be important to know the role of
B. hemisphaerica plants in the development of mosquito larvae.
We thank the following taxonomists for identification of the specimens collected in
the present work: Dr. Carlos R. Beutelspacher (Scorpionida), Ignacio Vazquez
(Araneae), Dr. Jos6 Palacios (Oribatulidae, Galumnoideae and Collembola), Maria Luisa
Estebanes (Acarina), Dr. Alfonso Garcia (Psocoptera), Eduardo Saucedo (Dictyoptera),
Gutidrrez et al.: Arthropods of Bromelia hemisphaerica 621
Guillermina Ortega (Dermaptera), Dr. Roberto Johansen (Thysanoptera), Luis Cer-
vantes (Hemiptera), Hector Gonzalez (Homoptera), Dr. Santiago Zaragoza (Coleoptera)
and Luis Quiroz (Hymenoptera). We also thank Dr J. H. Frank (Entomology and
Nematology Department, University of Florida) and Dr. P. Landolt (Insect Attrac-
tants, Behavior and Basic Biology Research Laboratory, USDA, Gainesville, Florida)
for guidance and corrections to a manuscript draft and I.P.N.-COFAA.
ALEXANDER, C. P. 1912. A second bromeliad inhabiting crane fly (Tipulidae: Diptera)
Mongoma bromeliadicola, sp. nov. Entomol. News 26: 109-115.
ANONYMOUS. 1980. Sintesis geogrAfica de Morelos. Secretaria de Programaci6n y
Presupuesto. Coordinaci6n General de los Servicios Nacionales de Estadistica,
Geografia e Informatica. M6xico, D.F., M6xico.
BENZING, D. H. 1970. An investigation of two bromeliad myrmecophytes: Tillandsia
butzii (Mez.), T. caput-medusae E. Morren, and their ants. Bull. Torrey Bot.
Club 97: 109-115.
BEUTELSPACHER B., C. 1969. Una especie nueva de Acrolophus Poey, 1832 en
bromeliiceas (Lepidoptera: Acrolophidae). An. Inst. Biol., Univ. Nac. Auton.
Mexico, Ser. Zool. 40: 43-48.
BIEZANKO, C. M. 1961. Contribuigo ao conhecimento da fisiografia do Rio Grande do
Sul. XIV. Castniidae, Zygaenidae, Dalceridae, Eucleidae da zona missioneira do
Rio Grande do Sul. Arq. Entomol. B.: 45-62.
CALVERT, P. P. 1911. Studies on Costa Rican Odonata. II. The habits of the plant-
dwelling larva of Mecistogaster modestus Selys. Entomol. News 22: 402-411.
DE BUEN, A. 1953. Observaciones ecol6gicas sobre mosquitos de "El Ajenjibre," Pue-
bla, M6xico. An. Inst. Biol., Univ. Nac. Auton. M6xico 24: 177-204.
DOWNS, G. W., AND C. S. PITTENDRIGH. 1946. Bromeliad malaria in Trinidad,
BritishWest Indies. American J. Trop. Med. 26: 46-66.
FRANK, J. H. 1983. Bromeliad phytotelmata and their biota, especially mosquitoes,
pp. 101-128 in J. H. Frank and L. P. Lounibos [eds.], Phytotelmata: Terrestrial
Plants as Hosts for Aquatic Insect Communities. Plexus, Medford, NJ.
FRANK, J. H. 1990. Mosquito production from bromeliads in Florida. Florida Agric.
Exp. Stn. Bull. 877: 1-17.
GARCIA-FRANCO, J. G. 1986. Las bromelias de M6xico: Revisi6n bibliogrAfica y de
herbario. Institute Nacional de Investigaciones sobre Recursos Bi6ticos, Xalapa.
GARDURO, R., M. SORIANO, E. CHAVEZ, M. T. CRUZ, L. M. CASTILLO, AND M.
CASTAREDA AGULLO. 1974. Proteinasas de plants mexicanas II. Puntos
isoelectricos y caracterizaci6n de formas moleculares multiples en enzimas de
bromeliAceas. Revista Latinoamericana de Quimica 5: 243-248.
LAESSLE, M. A. 1961. A micro-limnological study of Jamaican bromeliads. Ecology
LOOMIS, H. F. 1969. Additions to the millipedes of Mexico (Myriapoda: Diplopoda).
An. Inst. Biol., Univ. Nac. Auton. M6xico, Ser. Zool. 40: 49-53.
MURILLO, R. M., AND J. G. PALACIOS V. 1983. Variaci6n estacional de la fauna
asociada a Tillandsia spp. en una zona de transici6n bi6tica. Southwestern En-
tomol. 8: 292-302.
PALACIOS V., J. G., AND F. VARGAS. 1981. Tres nuevas Brachystomella (Collembola:
Neanuridae) de M6xico. Bull. Soc. Hist. Nat. Toulouse 117: 263-271.
Florida Entomologist 76(4)
EFFECT OF PHOTOPERIOD ON WING AREA IN
ANOPHELES QUADRIMACULATUS (DIPTERA: CULICIDAE)
CARMINE A. LANCIANI AND RONALD EDWARDS
Department of Zoology, University of Florida
Gainesville, FL 32611, U.S.A.
The effect of photoperiod on wing area of a mosquito in the species complex
Anopheles quadrimaculatus was tested on adults reared from a field sample. Short-
photoperiod [8:16 (L:D)] individuals of each sex had greater wing areas than did their
long-photoperiod [16:8 (L:D)] siblings of the same wing length or body weight. The
greater wing areas of short-photoperiod mosquitoes, given any wing length, established
that short-photoperiod individuals had broader wings. The greater wing areas of short-
photoperiod mosquitoes, given any body weight, established that short-photoperiod
individuals had a greater wing area per unit body weight.
Key Words: Wing dimensions, wing loading, mosquito, light.
El efecto del fotoperiodo sobre el area de la ala de una especie de mosquito del
complejo de Anopheles quadrimaculatus fu4 ensayado en adults criados de muestreos
de campo. Individuos de ambos sexos expuestos a fotoperiodo corto [8:16 (L:D)] tuvieron
areas de las alas mas grandes que las tuvieron sus hermanos de la misma largura de ala
y del mismo peso del cuerpo los cuales fueron expuestos a fotoperiodos largos [16:8
(L:D)]. El area mAs grande de alas de mosquitoes expuestos a fotoperiodos cortos, dado
cualquier largura de ala, estableci6 que individuos expuestos a fotoperiodo corto
tuvieron alas mas anchas. El area de ala mas grande de las alas de mosquito expuestos
a fotoperiodos cortos, dado cualquer peso, estableci6 que los individuos expuestos a
fotoperiodo corto tenfan un area mAs grande de ala por unidad de peso del cuerpo.
Mosquitoes reared under different photoperiods develop different wing lengths: indi-
viduals exposed to short photoperiods consistently develop disproportionately longer
wings than do those exposed to long photoperiods (Lanciani 1992). The present study
was done to determine whether wing area, too, is affected by photoperiod because of
the importance of wing area to flight (Vogel 1988, Nachtigall 1989). Our particular
objectives were to determine if the relationships between (1) wing area and wing length
and (2) wing area and body weight change under different developmental photoperiods.
If the wing area-wing length relationship changes, then wing shape must vary under
different photoperiods; if the wing area-body weight relationship changes, then body
weight transported per unit wing area must vary under different photoperiods. To
attain these objectives, we reared members of the mosquito species complex Anopheles
quadrimaculatus (probably species A on the basis of locale and habitat; Kaiser et al.
1988) under short and long photoperiods. We then analyzed their body weights, wing
lengths, and wing areas. We have observed blood-fed and gravid females throughout
the year in the area of study (Alachua County, Florida, U.S.A.), so flight during both
long-photoperiod, warm seasons and short-photoperiod, cool seasons is known to occur
in this species.
Lanciani & Edwards: Photoperiod and Wing Area
MATERIALS AND METHODS
Rearing followed the procedure described in Lanciani (1992). Gravid females were
collected on 13-1-1992 from the shore of a river (Styx) near Gainesville, Florida and
were held in separate vials in a constant-temperature chamber set at 28 C and a 12-h
light-dark cycle. Only a single batch of eggs, laid by the first ovipositing female, was
reared to ensure that all analyzed mosquitoes were siblings. Half of the eggs from this
batch were put in a short-photoperiod constant-temperature chamber [8:16 (L:D)] set
at 280 C and the other half in a long-photoperiod constant-temperature chamber [16:8
(L:D)] also set at 280 C. At a latitude of 300 (close to that of Gainesville), the longest
photoperiod (including 56 min of twilight) is 15 h and 1 min, and the shortest photoperiod
(including 52 min of twilight) is 11 h and 4 min (List 1971). The experimental photo-
periods, although different from the natural photoperiods, were selected because our
study could be more readily compared with other studies that use these common experi-
mental photoperiods and because these photoperiods have been shown to produce differ-
ent wing lengths and body weights in this species (Lanciani 1992).
To be sure that a constant temperature of 280 C was maintained in both short and
long-photoperiod chambers, we set temperature controls lower during light periods to
compensate for heat given off by the lights (a pair of 20W fluorescent lights in each
chamber). Before experiments began, these controls were adjusted while temperature
was checked with a thermometer placed in a water-filled 250-ml Erlenmeyer flask lo-
cated next to the larval rearing pans. These thermometer readings agreed with average
thermocouple readings recorded from the top 5 mm of water in different parts of the
rearing pans. Thus, the temperature experienced by the mosquito larvae, which occupy
the upper few mm of water, was probably the same at both photoperiods. Thermometers
were monitored throughout the experiments to verify that temperature remained at 280
Eggs were held in 500 ml of tap water in a white enamel pan, and, on the day after
oviposition, 0.05 g of a 2:1 mixture of baby-fish food and brewer's yeast was added to
each pan. Two days later, groups of approximately 40 larvae of similar size were selected
from each photoperiod group and placed in separate pans containing 500 ml of tap water
and 0.06 g of food. On subsequent days, larvae were transferred to clean pans with 500
ml of fresh tap water and fed successively 0.06, 0.07, and then 0.09 g of food per pan
until pupation. The pans were covered with clear plastic sheets to reduce evaporation.
As pupae appeared, they were held individually in screen-covered vials in the same
constant-temperature chamber in which they developed. Pupation occurred an average
of 7.35 days after oviposition (standard error = 0.05) in short-photoperiod individuals
and an average of 7.05 days after oviposition (standard error = 0.03) in long-photoperiod
individuals. Although the difference in time to pupation was statistically significant (F
test: F = 24.33; df = 1, 156; P 0.0001), other hearings of this species in our laboratory
have shown no consistent differences in time to pupation between photoperiod groups.
Adults were removed from constant-temperature chambers within 8 h of emergence
and were frozen. Later they were dried for 2 days at 600 C and weighed individually to
the nearest 0.002 mg. (Dry weight rather than live weight was used throughout the
analysis because it is a more consistent indicator of weight in small organisms.) One
wing was removed from each specimen, briefly immersed in 70% ethanol to remove
624 Florida Entomologist 76(4) December, 1993
wrinkles, and mounted in a drop of Hoyer's medium. Wing length was measured from
the axillary incision to the apex, excluding scales.
Wing area was measured on the slide-mounted material. Wing images were trans-
mitted to a microcomputer with a microscope attachment and video camera. From 25
to 30 points along the perimeter of each wing image were digitized for analysis by
MicroComp software (Southern Micro Instruments).
We used analysis of covariance to determine how photoperiod affected wing area.
In an analysis of covariance, the effect of photoperiod on wing area can be seen after
the effects of other variables, such as wing length, body weight, and gender, have been
removed (Packard & Boardman 1987). Two analyses were run with wing area as a
dependent variable: one with the independent variables wing length, gender, and photo-
period and the other with the independent variables body weight (dry), gender, and
photoperiod. Wing area, wing length, and body weight were logarithmically trans-
formed before the analysis to improve the linear relationship among these variables.
Statistical analyses were executed using SuperANOVA software (Abacus Concepts).
Average wing lengths, wing areas, and body weights of adult mosquitoes reared
under short and long photoperiods are listed in Table 1. In the analysis of covariance
involving wing length as an independent variable, wing area was significantly affected
by photoperiod (F = 11.69; df = 1, 154; P = 0.0008), gender (F = 95.56; df = 1, 154;
P = 0.0001), and wing length (F = 70.61; df = 1, 154; P = 0.0001). Specifically, wing
areas were greater in short-photoperiod, female, and long-winged individuals. From the
covariance model, we predicted wing areas to be (1) 3.428 mm2 and 3.296 mm2 in short
and long-photoperiod individuals having the same average wing length of 3.565 mm and
(2) 3.090 mm2 and 3.656 mm2 in males and females of that same average wing length.
Thus, given the same wing length and gender, the average short-photoperiod mosquito
had a greater wing area than did the average long-photoperiod one. Also, given the
same wing length and photoperiod, the average female had a greater wing area than
did the average male.
In the analysis of covariance involving body weight as an independent variable, wing
area was again significantly affected in the same way by photoperiod (F = 67.69; df =
1, 152; P = 0.0001) and gender (F =17.77; df = 1, 152; P = 0.0001). In addition, wing
area was significantly affected by body weight (F = 31.92; df = 1, 152; P = 0.0001),
i.e., heavier individuals had greater wing areas; a gender by photoperiod interaction (F
TABLE 1. AVERAGE WING LENGTH (MM), AVERAGE WING AREA (MM2), AND AVER-
AGE DRY BODY WEIGHT (MG) IN ADULT MOSQUITOES OF THE ANOPHELES
QUADRIMACULATUS COMPLEX REARED UNDER SHORT AND LONG
Gender Photoperiod Avg WL SE Avg WA + SE Avg DBW SE N
Male Short 3.486 0.015 3.063 0.021 0.633 0.007 48
Long 3.330 0.014 2.827 0.025 0.613 0.007 47
Female Short 4.012 0.018 4.285 0.042 0.821 0.015 31
Long 3.648 0.013 3.656 0.037 0.631 0.010 32
Lanciani & Edwards: Photoperiod and Wing Area 625
TABLE 2. PARAMETERS OF REGRESSION EQUATIONS PREDICTING LOG WING AREA
(MM2) FROM LOG DRY BODY WEIGHT (MG) IN ADULT MOSQUITOES OF THE
ANOPHELES QUADRIMACULATUS COMPLEX REARED UNDER SHORT AND
Gender Photoperiod Slope' Intercept'
Male Short 0.376 0.560
Long 0.376 0.531
Female Short 0.137 0.643
Long 0.137 0.590
'The equations are of the form Y = b X + a, in which Y is log wing area, b is the slope, X is log dry body weight,
and a is the Y intercept. For example in short-photoperiod males, log wing area = (0.376) log dry body weight + 0.560.
= 5.508; df = 1, 152; P = 0.0202); and a gender by body weight interaction (F = 6.93;
df = 1, 152; P = 0.0094). This model, which accounted for 91.2% of the variation in
wing area, produced 4 regression equations relating wing area to body weight (Table
2). The greater wing area of short-photoperiod mosquitoes is reflected by the larger
intercept of the short-photoperiod equation within each gender.
Photoperiod significantly changed wing dimensions in this species of the An. quad-
rimaculatus complex. Individuals reared under a short photoperiod developed greater
wing areas at all wing lengths and body weights than did siblings reared under a long
photoperiod. The different wing areas of short and long-photoperiod mosquitoes, given
any wing length, established that photoperiod affected wing shape; short-photoperiod
individuals had broader wings. The different wing areas of short and long-photoperiod
mosquitoes, given any body weight, established that photoperiod affected wing area
per unit body weight; short-photoperiod individuals had a greater wing area per unit
body weight. Thus, short-photoperiod mosquitoes in flight would carry less weight per
unit wing area and would therefore have a lower wing loading (body weight divided by
wing area; Lighthill 1977).
Photoperiod thus influences allometric development in the wings of this species.
Other photoperiod-induced effects in insect morphology, physiology, and natural history
have been documented (Giesel et al. 1989, Lanciani et al. 1990, and Lanciani 1992).
Some of these same effects are also induced by temperature in mosquitoes (Hosoi 1954,
van den Heuvel 1963, and Nayar 1968), emphasizing that insects often respond similarly
to temperature and photoperiod.
GIESEL, J. T., C. A. LANCIANI, AND J. F. ANDERSON. 1989. Larval photoperiod and
metabolic rate in Drosophila melanogaster. Florida Entomol. 71: 123-128.
HosoI, T. 1954. Egg production in Culex pipiens pallens Coquillett. IV. Influence of
breeding conditions on wing length, body weight, and follicle production.
Japanese J. Med. Sci. Biol. 7: 129-134.
KAISER, P. E., S. E. MITCHELL, G. C. LANZARO, AND J. A. SEAWRIGHT. 1988.
Hybridization of laboratory strains of sibling species A and B of Anopheles quad-
rimaculatus. J. American Mosq. Control Assoc. 4: 34-38.
LANCIANI, C. A. 1992. Photoperiod and the relationship between wing length and
body weight in Anopheles quadrimaculatus. J. American Mosq. Control Assoc.
626 Florida Entomologist 76(4) December, 1993;
LANCIANI, C. A., J. T. GIESEL, J. F. ANDERSON, AND S. S. EMERSON. 1990. Photo-
period-induced changes in metabolic response to temperature in Drosophila
melanogaster Meigen. Funct. Ecol. 4: 41-45.
LIGHTHILL, J. 1977. Introduction to the scaling of aerial locomotion, pp. 365-404 in T.
J. Pedley [ed.], Scale effects in animal locomotion. Academic Press, London.
LIST, R. J. 1971. Smithsonian meteorological tables. Smithsonian Institution Press,
City of Washington.
NACHTIGALL, W. 1989. Mechanics and aerodynamics of flight, pp. 1-30 in G. J.
Goldsworthy and C. H. Wheeler [eds.], Insect flight. CRC press, Boca Raton,
NAYAR, J. K. 1968. The biology of Culex nigripalpus Theobald (Diptera: Culicidae).
Part 2. Adult characteristics at emergence and adult survival without nourish-
ment. J. Med. Entomol. 5: 203-210.
PACKARD, G. C., AND T. J. BOARDMAN. 1987. The misuse of ratios to scale physiolog-
ical data that vary allometrically with body size, pp. 216-236 in M. E. Feder, A.
F. Bennett, W. Burggren, and R. B. Huey [eds.], New directions in ecological
physiology. Cambridge University Press, Cambridge.
VAN DEN HEUVEL, M. J. 1963. The effect of rearing temperature on the wing length,
thorax length, leg length and ovariole number of the adult mosquito, Aedes
aegypti (L.). Trans. Roy. Entomol. Soc. London. 115: 197-216.
VOGEL, S. 1988. Life's devices. Princeton University Press, Princeton, New Jersey.
EVALUATION OF PROTEIN BAIT FORMULATIONS FOR
THE CARIBBEAN FRUIT FLY (DIPTERA: TEPHRITIDAE)
NANCY D. EPSKY,' ROBERT R. HEATH,' JOHN M. SIVINSKI,'
CARROL O. CALKINS,' RICHARD M. BARANOWSKI,2 AND ANN H. FRITZ'
'U.S. Department of Agriculture
Agricultural Research Service
Insect Attractants, Behavior and Basic Biology
Gainesville, FL 32608
2Tropical Research & Education Center
University of Florida
Homestead, FL 33031
Laboratory and field trials were conducted to determine the preference of the Carib-
bean fruit fly, Anastrepha suspense (Loew), for aqueous formulations of the protein
bait NuLure and standard torula yeast plus sodium borate (HTY-borax) pellets. Addi-
tion of 1-10% borax to 10% NuLure solution increased bait pH, and this increase was
directly correlated with increase in number of female flies trapped in two-choice labora-
tory bioassays and in field trials conducted in three locations in south Florida during
the spring of 1992. Overall, significantly more flies were attracted to volatiles from
HTY-borax solution than to volatiles from any of the NuLure solutions. There was
variation in the response of flies to baits observed among the three test locations. Age
structures of the populations at the different locations were compared by determining
Epsky et al.: Protein Bait Formulations for Caribflies
the percentage of females mated among the trapped females. There were significant
differences in the percentage of mated females at the 3 locations, which ranged from
26-65% mated. In the locations with a high percentage of unmated female flies, the flies
were less discriminating, at least among the protein baits offered in this study. Thus,
age structure of the target population may alter response to bait. There was also vari-
ation in the pH of the baits at different locations due to the pH of the water used to
prepare the solutions. Bait pH may be a significant factor that has been overlooked in
field tests conducted at different locations or at different times as changes in water pH,
as well as bait protein pH, may strongly affect trap efficacy.
Key Words: Anastrepha suspense, Tephritidae, protein baits, trapping, bait pH.
Se llevaron a cabo ensayos de laborat6rio y de campo para determinar la preferencia
de la mosca caribefia de las frutas, Anastrepha suspense (Loew), por formulaciones
acueas del cebo proteinaceo NuLure y levadura torula standard mas pellotillas de
borato de sodio (TY-borax). La adici6n de 1-10% borax a una suluci6n de 10% de NuLure
aumento el pH del cebo, y este aumento fu4 corelacionado directamente con el aumento
en el nmmero de hembras de las moscas entrampadas en bioensayos de laboratorio de
dos surtidos y en ensayos de campo llevado a cabo en 4 localidades in el sur de Florida
durante la primavera de 1992. Por lo general, significativamente mas moscas fueron
atraidas a los volatiles de la solution TY-borax que a los volatiles de las soluciones de
NuLure. Hubo una variaci6n en la respuesta de las moscas a los cebos observados entire
las tres localidades de las pruebas. Las estructuras de las edades de las poblaciones en
las diferentes localidades fueron comparadas para determinear el percentage de las
hembras pareadas entire las hembras entrampadas. Hubieron diferencias significativas
en el percentage de las hembras pareadas en las tres localidades, lo cual vari6 entire 26
hasta 65% pareadas. En las localidades con un alto percentage de moscas hembras no
pareadas, las moscas fueron menos discriminatorias, por lo menos entire los cebos pro-
teinaceos ofrecidos en este studio. Asi, la estructura de las edades de la poblaci6n
objective puede alterar la respuesta al cebo. Hubo tambien variaci6n en el pH de los
cebos en diferentes localidades, debido al pH del agua usada para preparar las soluciones.
El pH del cebo puede ser un factor significativo, de lo cual no se ha hecho caso en
pruebas de campo llevado a cabo en diferentes localidades o durante diferentes periods,
porque cambios en el pH del agua, igual como pH del cebo proteinAceo, puede afectar
fuertamente la eficiencia de la trampa.
The presence of the Caribbean fruit fly (caribfly), Anastrepha suspense (Loew)
(Diptera: Tephritidae), in citrus growing regions of Florida is of considerable economic
importance. Even a small infestation of flies in an area creates a quarantine problem
and renders citrus unsalable to many potential domestic and foreign markets without
costly post-harvest treatments (Greany & Riherd 1993, Simpson 1993). Because of this
threat and the potential for introduction of the caribfly into current fly-free areas, much
emphasis has been placed on detection of this species. The development of improved
lures for monitoring and suppressing populations of the caribfly would safeguard current
fly-free zones and permit expansion of these zones to other citrus growing regions in
Traps baited with proteins have been used historically to detect and monitor popu-
lations of Anastrepha fruit flies. Hydrolyzed proteins, often by-products of commercial
operations, have been found to be superior to non-hydrolyzed proteins, and a number
of hydolyzed and non-hydrolyzed proteins have been tested for attractant activity
(Steiner 1955). The proteins in these food-based lures often provide nutrients critical