(An International Journal for the Americas)
Volume 72, No. 1 March, 1989
TABLE OF CONTENTS
Announcement 72nd Annual Meeting ........................................................ i
SYMPOSIUM: INSECT BEHAVIORAL ECOLOGY-'88
FRANK, J. H., AND E. D. McCoY-Introduction. Behavioral Ecology: From
Fabulous Past to Chaotic Future ...................................................... 1
LEWIS W. J., AND H. R. GROSS-Comparative Studies on Field Performance
of Heliothis Larval Parasitoids Microplitis croceipes and Cardiochiles nig-
riceps at Varying Densities and Under Selected Host Plant Conditions ... 6
ANTOLIN, M. F.-Genetic Considerations in the Study of Attack Behavior of
Parasitoids with Reference to Muscidifurax raptor (Hymenoptera:
Pteromalidae) .......................................... 15
STRAND, M. R.-Clutch Size Sex Ratio and Mating by the Polyembryonic Encyr-
tid Copidosoma floridanum (Hymenptera: Encyrtidae) .......................... 32
WOJCIK, D. P.-Behavioral Interactions Between Ants and Their Parasites. .... 43
ALLEN, J. C.-Patch-Efficient Parasitoids, Chaos and Natural Selection ....... 52
IVIE, M. A., AND S. A. SLIPINSKI--The Pycnomerini (Coleptera: Colydiidae) of
the W est Indies ............................................................................... 64
LLOYD, J. E., S. R. WING, AND T. HONGTRAKUL_-Flash Behavior and Ecology
of Thai Luciola Fireflies (Coleoptera: Lampyridae) .............................. 80
CHAN, K. L., AND J. R. LINLEY-Laboratory Studies on the Immature Stages
of Atrichopogon wirthi (Diptera: Ceratopogonidae) ............................... 85
HABECK, D. H., J. NAGEL, AND J. E. PENA-Colopterus posticus (Erichson), a
Nitidulid Beetle New to the United States ........................................ 89
DEYRUP, M., C. JOHNSON, G. C. WHEELER, AND J. WHEELER-A Preliminary
List of the A nts of F lorida ............................................................... 91
HEPPNER, J. B.-New Argyrotaenia and Choristoneura Moths from Florida
(Lepidoptera: Tortricidae) ......................................... ....................... 101
VINSON, S. B.,AND T. A. SCARABOROUGH-Impact of the Imported Fire Ant on
Laboratory Populations of Cotton Aphid (Aphis gossypii) Predators ....... 107
KLINE, D. L.-Seasonal and Spatial Abundance of Culicoides spp. Larvae in
Roadside Salt Marsh Areas at Yankeetown, Florida .......................... 111
TARRANT, C. A., AND C. W. McCoY-Effect of Temperature and Relative
Humidity on the Egg and Larval Stages of Some Citrus Root Weevils .... 117
GIESEL, J. T., C. A. LANCIANI, AND J. F. ANDERSON-Larval Photoperiod and
Metabolic Rate in Drosophila melanogaster ........................................ 123
ANTOLIN, M. F., AND R. L. WILLIAMS-Host Feeding and Egg Production in
Muscidifurax zaraptor (Hymenoptera: Pteromalidae) .......................... 129
PARKMAN, P., AND R. L. PIENKOWSKI-Response of Three Populations of
Liriomyza trifolii (Diptera: Agromyzidae) to Topical Applications of Per-
m ethrin and Bifenthrin ..................................... ............................ 135
Continued on Back Cover
Published by The Florida Entomological Society
FLORIDA ENTOMOLOGICAL SOCIETY
OFFICERS FOR 1988-89
President ............................. ... ....... ............. .... R. S. Patterson
President-Elect ...................................................... ... J. E. Eger
Vice-President ............................................................ J. F. Price
Secretary ....... .................................... ....... ...................... J. A Coffelt
Treasurer ................................... .. .. ................... A. C. Knapp
Other Members of the Executive Committee .................
J. L. Taylor
C. O. Calkins
J. E. Pefia
M. F. Antolin
J. R. McLaughlin
Editor ......................................... ................... J. R. McLaughlin
John B. Heppner
William W. Wirth
Carl S. Barfield
Michael D. Hubbard
Omelio Sosa, Jr.
Ronald H. Cherry
Lance S. Osborne
Howard V. Weems, Jr.
Business 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.
Membership in the Florida Entomological Society, including subscription to Florida
Entomologist, is $25 per year for regular membership and $10 per year for students.
Inquires regarding membership, subscriptions, and page charges should be addres-
sed to the Business 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 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. Submit the original
manuscript, original figures and tables, and 3 copies of the entire paper. Include an
abstract in Spanish, if possible. Upon receipt, a manuscript is acknowledged by the
Editor and assigned to an Associate Editor who sends it out for review by at least 3
knowledgeable peers. Reviewers are sought with regard only for their expertise; Soci-
ety membership plays no role in their selection. Page charges are assessed for printed
Manuscripts and other editorial matter should be sent to the Editor, JOHN R.
MCLAUGHLIN, 4628 NW 40th Street, Gainesville, FL 32606.
This issue mailed March 31, 1989
THE 72nd ANNUAL MEETING OF THE
FLORIDA ENTOMOLOGICAL SOCIETY
The 72nd annual meeting of the Florida Entomological Society will be held August
7-10, 1989 at the Daytona Beach Hilton, 2637 So. Atlantic Ave., Daytona Beach, FL
32018; telephone (904)-767-7350. Registration forms and information will be mailed to
members and will appear in the Newsletter and the March, 1989 Florida Entomologist.
Notice of Change of Deadline for Submission of Papers
The deadline for submission of papers and posters for the 72nd annual meeting
of the Florida Entomological Society will be May 1, 1989. The meeting format will
be much the same as in the past with eight minutes allotted for presentation of oral
papers (with 2 minutes for discussion) and separate sessions for members who elect to
present a Project (or Poster) Exhibit. The three oral student papers and the three
student Project Exhibits judged to be the best on content and delivery will be awarded
monetary prizes during the meeting. Student participants in the judged sessions must
be Florida Entomological Society Members and must be registered for the meeting.
James R. Price, Chairman
Program Committee, FES
University of Florida, IFAS
Gulf Coast Research & Education Center
5007 60th Street East
Bradenton, FL 34203
*, ci -
k T~f 0
~-4 a. A- -
o ~ -
U c i;b
- .c 0 '
Insect Behavioral Ecology-'88 Frank & McCoy
ATTACK AND DEFENSE: BEHAVIORAL ECOLOGY OF
PARASITES AND PARASITOIDS AND THEIR HOSTS.
BEHAVIORAL ECOLOGY: FROM FABULOUS PAST
TO CHAOTIC FUTURE
J. H. FRANK
Entomology and Nematology Department
3103 McCarty Hall, University of Florida
Gainesville, FL 32611
E. D. McCoY
Department of Biology,
University of South Florida
Tampa, FL 33620
Genesis of this symposium on behavioral ecology owes something to Sir Francis
Drake. His observations in 1579 in the Moluccas were recorded by Hakluyt (ca. 1598)
in the following terms. "Amongst these trees night by night, through the whole land,
did shew themselves an infinite swarme of fiery wormes flying in the ayre, whose bodies
being no bigger then our common English flies, make such a shew and light, as if every
twigge or tree had bene a burning candle."
The spectacular nature of flashing by fireflies in countries of the southwestern Pacific
basin attracted the curiosity of James Lloyd (Lloyd 1973). Lloyd's discoveries on fireflies
were observational and interpretive and, when he discovered the existence of behavioral
ecology as a discipline, he knew he had found his niche (Lloyd 1980). The first seven
symposia held by this society and organized by James Lloyd explored the panoply of
behavioral ecology and fascinated the audiences. The eighth symposium, organized by
Earl McCoy and Don Strong, introduced a theme to the event, and the theme then was
of their choosing: The Behavioral Ecology of Colonization.
This ninth symposium also has a theme: Attack and Defense The Behavioral Ecol-
ogy of Parasites and Parasitoids and their Hosts. The theme was chosen on the premise
that biological control is a rising area of research, and that much that is novel and
interesting in behavioral ecology is being done under the auspices of biological control.
Evolution of words in the field of ecology and its subdiscipline biological control prompts
us here to attempt an etymological clarification.
The noun parasite has been used in English for many hundreds of years, and the
noun parasitism at least since the early 17th century (OED 1971). It was unnecessary
therefore to derive the noun parasitization (as a synonym of parasitism) from the verb
to parasitize. The word parasitization is also longer and clumsier and smacks of the
excesses of Orwell's newspeak (Orwell 1949).
The nouns parasite and predator were used until the 20th century to describe ani-
mals which feed on other animals, but then two inadequacies were raised. One inade-
quacy was in the definition of the behavior of parasites. If we take the definition of
parasite to include only those animals which do not kill their hosts in the sense of having
achieved an evolutionary balance between the immediate and long-term nutritional
needs of the parasite, as exemplified in Aesop's fable The Goose with the Golden Eggs
(Fig. 1), then another word is required for parasite-like animals which do kill their
Florida Entomologist 72(1)
Fig. 1. "The goos and her lord" from Caxton's (1483) translation of Aesop's (ca. 550
BC) fable known widely in English as The Goose with the Golden Eggs. One version of
the fable tells that the goose lays a golden egg every day, but the dissatisfied owner
kills her in expectation of finding a golden treasury inside her. Instead, she is empty
of gold and also is dead.
hosts. This word, parasitoid, was supplied by Reuter (1913), and has been adopted into
English (OED 1971). Yet, in much entomological writing, the word parasite is used in
the older sense to include parasitoid as well as parasite (in the strict sense). Perhaps
this failure to use the word parasitoid wherever it applies is because of uncertainty of
the status of the corresponding noun parasitoidism, verb to parasitoidize, adverb
parasitoidally, and adjective parasitoidal. However, logic directs that these words be
used wherever they are needed.
With acceptance of the noun parasitoid, the three definitions are as follows. Para-
site: An organism that lives in or on the body of its host without killing the host, but
usually debilitating the host to some extent. Parasitoid: An organism that, during its
development, lives in or on the body of a single host individual, eventually killing that
individual. Predator: An organism that, during its development, consumes more than
one prey individual.
The other inadequacy was in the concept of the sort of host or prey consumed. A
popular English name for aphid is plant-louse, implying that aphids, just as lice, are
parasites. The definition above of parasite specifies the body of its host, and body is
defined by Webster (1986) as the organized physical substance of an animal or plant
either living or dead. Therefore, the definitions above do allow inclusion of plants as
hosts of parasites and parasitoids and as prey of predators. To make certain of this
inclusion in the literature of ecology, Price (1975) wrote of aphids, weevils, etc. as
parasites of plants, while Janzen (1971, 1975) wrote of beetles as predators and
parasitoids of seeds and coined the expression seed-predator. Grazing by cattle on
plants could be considered as parasitism if we disregard the relative sizes of parasite
Insect Behavioral Ecology-'88 Frank & McCoy
and host. However, grazing by nudibranchs on bryozoan colonies was dubbed partial
predation by Harvell (1984), so concepts are fluid because authors do differentiate
according to relative sizes of the consumer and the organism consumed.
Other words are now entering into English. Two useful ones are koinobiont and
idiobiont [from Greek koinos (in common), idios (individual), and bios (life)] (Askew
& Shaw 1986, modified from Haeselbarth 1979). A koinobiont is a parasitoid developing
in a parasitoidized host which continues to be mobile and able to defend itself; larval
hosts often are not killed until they have prepared cryptic pupation retreats. The host
may not live very long after parasitoidism, but the koinobiont benefits from the con-
tinued life of its host. An idiobiont is a parasitoid which consumes the host in the location
and stage it is in when attacked.
This symposium is not on the subject of behavioral ecology of predators and their
prey, but our etymological voyage is incomplete without the following. The adjectives
describing the distinguishing behavior of predators are predatory and predacious, also,
erroneously, predaceous, cf. audacious, voracious, ferocious (OED 1971). Adoption of
the spelling predaceous by Webster (1986) presumably is an acknowledgment rather
than an endorsement of current erroneous spelling in North America. Perhaps some
dictionaries of the last century acknowledged the spelling cocoa-nut as the name of the
seed of Cocos nucifera L. (that spelling certainly was used), whereas today the
etymologically more appropriate spelling coconut is accepted universally.
Before progressing to the chaotic future of behavioral ecology it is entertaining and
perhaps illuminating to explore other writings from the past. These contain speculations
which are not based on observation such as that by Drake. The principal character of
a novel (Godwin 1638) of the mid-17th century, Domingo Gonsales, visited the moon,
drawn there in an "engine" pulled by "gansas" [? geese] he had trained, during their
annual migration. On the moon he found a fauna and flora different from that of earth,
but including swallows and other migratory birds from earth. Intended only as a novel,
it may have stimulated speculation in the mind of Charles Morton, a physician and
philosopher (Harrison 1954).
Morton's philosophical work (Morton ca. 1694) speculated on the disappearance of
swallows and other birds each autumn and their return each spring: ". . their cheare-
fulness seems to intimate . that they have some noble design in hand . namely,
to get above the atmosphere, hie and fly away to the other world ... till some other
more fit place can be assigned, do go into and remain in some one of the celestial bodies;
and that must be the moon, which is most likely, because nearest . if the moon be
not allowed, some other place must be found out for them."
Morton thus recognized the disappearance of swallows as a migration, which was an
advance on popular belief of that time. Popular belief was still being voiced some decades
later, by no less an authority than Samuel Johnson (1768), in the following terms:
"Swallows certainly sleep all the winter. A number of them conglobulate together, by
flying round and round, and then all in a heap throw themselves under water, and lye
in the bed of a river."
Unfortunately, both worthies (Morton and Johnson) overlooked or ignored the ear-
lier publication of a serious treatise on ornithology by Francis Willughby (1676), which
included speculation on the migratory habits of swallows: "Quo abeant vel ubi latitent
hirundines hyberno tempore, nec inter rei naturalis scriptores convenit, nec nobis sane
compertum est. Verisimilius tamen videtur eas in regions calidores Aegyptum puta
aut Aethiopiam avolare ..." Willughby had speculated that swallows migrate to Egypt
or Ethiopia in the winter. The later views on the subject by Morton (ca. 1694) and
Johnson (1768) were in vain.
The past of behavioral ecology was fabulous. The beginnings of behavioral ecology's
explosive growth phase coincided with publication of Wilson's Sociobiology (Krebs
4 Florida Entomologist 72(1) March, 1989
1985). Studies of insects played key roles in both the early formation of behavioral
ecology as a discrete subject and its subsequent rapid development (Burk 1988, Krebs
Krebs (1985) [see also Burk 1988 ] tabulated what he and a dozen colleagues thought
were the most important developments in behavioral ecology during the decade 1975-
1985. The areas of investigation they emphasized can be lumped into five categories
relevant to insects: (1) game theory and alternative strategies, (2) optimization, (3) kin
selection and kin recognition, (4) parental investment, and (5) communication and as-
sessment. Krebs (1985), from his poll of colleagues, also endeavored to predict the
immediate future courses of behavioral ecological research. Five paths were predicted:
(1) life history and population dynamics in relation to behavioral ecology, (2) mating
systems, (3) parasites and sexual selection, (4) learning, and (5) the genetic basis of
behavior. Burk (1988), too, predicted five pathways for productive research, in be-
havioral ecology of insects: (1) sexual selection, (2) resource competition among females,
(3) learning, (4) orientation and movement, and (5) communication.
One might infer, from comparing the two sets of predictions, that insect behavioral
ecology lags in but a single area of research: communication. In other areas it seems to
fit in with behavioral ecology of the rest of the animal kingdom. How do our symposium
contributions fit in with predictions by Krebs (1985) and Burk (1988)?
Mike Strand's contribution details the reproductive behavior of a parasitoid, and
examines its potential connection to competition for mates. Investigating mating sys-
tems is recognized by both Krebs (1985) and Burk (1988) as an important future course
of behavioral ecological research. Two of the contributions (by Joe Lewis [with H.R.
Gross, Jr.] and Jim Cronin [text not published here]) deal with the importance of
environmental structure, especially vegetational types, on searching behavior of
parasitoids. Burk (1988) emphasizes the important future role of studies of oriented
movement in insect behavioral ecology. Dan Wojcik examines the myriad interactions
between ants and their parasites and parasitoids, and questions previous generalizations
about these interactions. His contribution raises an important point: future paths of
behavioral ecology must lead from a solid foundation, lest we find ourselves swept up
in a bacchanal of hollow theorizing (see Vehrencamp & Bradbury 1984).
Two of the contributions (by Jon Allen, and Mike Antolin) concern genetic bases of
behavioral interactions between parasitoids and hosts. Investigating the genetic basis
of behavior is included explicitly in Krebs' (1985) tabulation, and is woven implicitly
throughout Burk's (1988) discussion. Jon Allen's assertion that chaotic population fluctu-
ations can result from density dependence is heartening to those of us who sometimes
have tried in vain to see regularity in density dependent systems. Theory and practice
come together here, in chaos.
We thank the following (all of the University of Florida) for help rendered: Dale
Habeck (Entomology & Nematology Department) and Smith Kirkpatrick (English De-
partment) for critical comments on the manuscript of this introduction; Audrey Frank
(formerly of the Reference Department, Main Library) for bibliographic help; Melvyn
New (English Department) for the page reference to Samuel Johnson's statement on
swallows; and Carlos Mantilla, Isa Montenegro, Helena Puche, Alfredo Rueda, and
Victor Salguero (Entomology & Nematology Department) for providing Spanish re-
sumenes of the five contributed papers. Eight reviewers kindly read and helped improve
the five manuscripts of submitted papers: Yolanda Cruz (Oberlin College), Don Feener
(University of California, Los Angeles), Pat Greany (USDA, Gainesville), Will Hudson
(University of Georgia), Mike Strand (University of Wisconsin), James Trager (Univer-
Insect Behavioral Ecology-'88 Frank & McCoy
sity of Missouri), Tom Unruh (USDA, Washington), and Sandy Walde (Dalhousie Uni-
versity). This is University of Florida, Institute of Food & Agricultural Sciences, jour-
nal series no. 9547.
ASKEW, R. R. AND M. R. SHAW. 1986. Parasitoid communities: their size, structure
and development p. 225-64, in J. K. Waage and D. J. Greathead [eds.], Insect
Parasitoids. Academic Press, London, xvii + 389 p.
BURK, T. 1988. Insect behavioral ecology: Some future paths. Annl. Rev. Ent. 33:
CAXTON, W. 1483. Of the goos and of her lord. f. cxviii, in The subtyl historyes and
Fables of Esope which were translated out of Frensshe in to Englysshe by
Wylliam Caxton at westmynstre In the yere of oure Lorde, MDCCCClxxxiii.
Caxton; Westminster, cxlii ff. [reprinted 1972, Theatrum Orbis Terrarum;
GODWIN, F. 1638. The man in the moone; or, a discourse of a voyage thither by
Domingo Gonsales, the speedy messenger. J. Kirton & T. Warren; London, 126
p. [reprinted 1972, Theatrum Orbis Terrarum; Amsterdam].
HAESELBARTH, E. 1979. Zur Parasitierung der Puppen von Forleule (Panolis flam-
mea [Schiff.] ), Kiefernspanner (Bupalus piniarius [L.] ) und Heidelbeerspanner
(Boarmia bistortana [Goeze]) in bayerischen Kiefernwildern. Ztschr. Angew.
Ent. 87: 186-202, 311-22.
HAKLUYT, R. ca. 1598. The Famous Voyage of Sir Francis Drake into the South sea,
and therehence about the whole Globe of the earth, begun in the yeere of our
Lord, 1577 [reprinted, 1926 in] A Selection of the Principal Voyages, Traffiques
and Discoveries of the English nation by Richard Hakluyt 1552-1616; set out with
many embellishments and a preface by Laurence Irving. Alfred A. Knopf; New
York, xxiv + 294 p.
HARRISON, T. P. 1954. Birds in the moon. Isis 45: 323-30.
HARVELL, D. 1984. Why nudibranchs are partial predators: Intracolonial variation
in bryozoan palatability. Ecology 65: 716-24.
JANZEN, D. H. 1971. Seed predation by animals. Annl. Rev. Ecol. Syst. 2: 465-92.
1975. Interactions of seeds and their insect predators/parasitoids in a tropical
deciduous forest p. 154-86, in P. W. Price (ed.), Evolutionary Strategies of
Parasitic Insects and Mites. Plenum; New York, xi + 224 p.
JOHNSON, S. 1768. Natural philosophy, Spring 1768, in Boswell's Life of Johnson.
Oxford Standard Authors Edition, 1953 [reprinted 1965] Oxford University
Press; London, xxiv + 1,491 p.
KREBS, J. R. 1985. Sociobiology ten years on. New Sci. 108(1): 40-43
LLOYD, J. E. 1973. Fireflies of Melanesia: Bioluminescence, mating behavior, and
synchronous flashing (Coleoptera: Lampyridae). Environ. Ent. 2: 991-1008.
1980. Insect behavioral ecology: Coming of age in bionomics or Compleat
biologists have revolutions too. Florida Ent. 63: 1-4.
MORTON, C. ca. 1694. An Enquiry into the Physical and Literal Sense of that Scripture
(Jer. 8.7) . The Stork in the Heavens knoweth her appointed times; and the
turtle, and the Crane, and the Swallow observe the time of their coming ..
Written by an eminent Professor for the use of scholars, and now published at
the earnest desire of some of them. London [reprinted 1810 in The Harleian
Miscellany (London) 5: 498-511].
OED. 1971. The compact edition of the Oxford English Dictionary. Oxford University
Press; Glasgow, [ca. 16,460 p. reproduced micrographically in] 4,116 p.
ORWELL, G. 1949. Nineteen eighty-four, a novel. Harcourt, Brace; New York, 314 p.
PRICE, P. W. 1975. Introduction: The parasitic way of life and its consequences.
p. 1-13, in P. W. Price [ed.], Evolutionary Strategies of Parasitic Insects and
Mites. Plenum; New York, xi + 224 p.
Florida Entomologist 72(1)
REUTER, O. M. 1913. Lebensgewohnheiten und Instinkte der Insekten bis.zum Er-
wachen der sozialen Instinkte. Friedlander; Berlin, xvi + 448 p.
VEHRENCAMP, S. L. and J. W. BRADBURY. 1984. Mating systems and ecology p.
251-78, in J. R. Krebs and N. B. Davies [eds.], Behavioural Ecology: An
Evolutionary Approach. Blackwell; Oxford, 2nd edn., xi + 493 p.
WEBSTER. 1986. Webster's Ninth New Collegiate Dictionary. Merriam-Webster;
Springfield, MA, 1,563 p.
WILLUGHBY, F. 1676. Francisci Willughbeii Ornithologiae libri tres in quibus aves
omnes hactenus cognitae in methodum naturis suis convenientem redactae accu-
rate describuntur, descriptions iconibus elegantissimis & vivarum avium simil-
limis, aeri incisis illustrantur. Royal Society; London [12 +] 307 [ + 5] p. + 77 pl.
COMPARATIVE STUDIES ON FIELD PERFORMANCE OF
HELIOTHIS LARVAL PARASITOIDS MICROPLITIS
CROCEIPES AND CARDIOCHILES NIGRICEPS AT
VARYING DENSITIES AND UNDER SELECTED
HOST PLANT CONDITIONS
W. J. LEWIS AND H. R. GROSS
Insect Biology and Population Management Research Laboratory, Agri. Res. Serv.,
USDA, Tifton, GA 31793
Field performance of the Heliothis (Lepidoptera: Noctuidae) larval parasitoids
Microplitis croceipes (Cresson) and Cardiochiles nigriceps (Viereck) (Hymenoptera:
Braconidae) was assessed at various parasitoid densities and on different plants. Labo-
ratory-reared M. croceipes were released in field plots of soybeans and peanuts at levels
of ca. 400 and 1,350/ha. Observed and calculated data indicate that over a 4-day period
parasitism levels of 45 and 95%, respectively, would be obtained with M. croceipes at
these levels on soybeans. Performance of M. croceipes on peanuts was less consistent,
but in one study, 71% parasitism was obtained during a 2-day interval after releases of
The searching behavior and efficiency of C. nigriceps females were studied on the
hyacinth bean (Dolichos lablab L.) and Florida beggarweed (Desmodium purpureum
(Mill.)). Observed and calculated data indicate that populations of 2,622 and 5,371
females/ha on the hyacinth bean and beggarweed, respectively, would yield 80%
parasitism, as compared to earlier studies showing that 988-1,142 females/ha yield this
level of parasitism on cotton.
These findings show that the efficiency of the parasitoids varies considerably on
different host plants. The distribution of searching parasitoids within the experimental
plots and the distribution of their eggs among the host larvae did not differ significantly
from a random (Poisson) pattern.
Se evalu6 en el campo el comportamiento de Microplitis croceipes (Cresson) y Car-
diochiles nigriceps (Viereck) (Hymenoptera: Braconidae), parasitoides de la larva de
Heliothis (Lepidoptera: Noctuidae). En la evaluacio6i se utilizaron varias densidades de
parasitoides y diferentes species de plants. M. croceipes fueron criadas en el
Insect Behavioral Ecology-'88 Lewis & Gross
laboratorio y liberadas en parcelas de soya y mani a niveles de aproximadamente 400 y
1,350/ha. Datos observados y calculados indican que durante un period de 4 dias se
obtendrian niveles de parasitismo de 45 y 95%, respectivamente, utilizando M. croceipes
a los mencionados niveles en soya. El comportamiento de M. croceipes en mani fue
menos constant, pero en un studio se obtuvo un 71% de parasitismo durante un
period de 2 dias, posterior a la liberaci6n de 1,344 hembras/ha.
El comportamiento de bisqueda y eficiencia de las hembras de C. nigriceps fue
estudiado en Dolichos lablab L. y Desmodium purpureum (Mill.). Datos observados y
calculados indican que poblaciones de 2,622 y 5,371 hembras/ha en Dolichos y De-
smodium, respectivamente, produce un 80% de parasitismo, comparado con studios
previous que muestran que 988-1,142 hembras/ha produce el mismo nivel de parasitismo
en el algod6n.
Estos datos muestran que la eficiencia de los parasitoides varia c considerablemente
en diferentes plantas-hudsped. La distribuci6n de los parasitoides durante la bfsqueda
dentro de las parcelas experimentales, y la distribuci6n de los huevos entire las larvas
hu6sped, no difiri6 significativamente del patron de distribuci6n al azar (Poisson).
Heliothis zea (Boddie) and H. virescens (L.) are attacked by numerous parasitic
insects. However, the kinds and abundance of attacking parasitoids vary significantly
among the many host plants of these phytophagous pests (Danks 1979, Lewis & Brazzel
1968, Mueller & Philips 1983, Neunzig 1963, Nordlund et al. 1986). The larval parasitoids
Microplitis croceipes (Cresson) and Cardiochiles nigriceps (Viereck) are specific to
Heliothis spp.; M. croceipes attacks H. zea and H. virescens, while C. nigriceps is
limited to H. virescens.
Because they are active on most of the different host plants of Heliothis, both of
these parasitoid species, particularly M. croceipes, have been considered for use in
areawide release programs against Heliothis. The varying efficiencies of these
parasitoids in locating and parasitizing hosts on different plants are major factors to
consider in appraising their use for augmentative releases.
The relative occurrence of the parasitoids on different ages of host insects, plant
species, and time of season have been reported by several authors (Lewis & Brazzel
1968, Mueller 1983, Hopper & King 1984). However, data on the actual number of
parasitoids that produce various parasitism levels and specific information on factors
that influence their performance are very limited. Lewis et al. (1972) evaluated the
efficiency of C. nigriceps attacking H. virescens on cotton. The purpose of this study
was to determine the number of females required to yield various parasitism levels on
different host plants, assess the distribution of ovipositional activity among the hosts,
and to obtain preliminary information about how seasonal changes in plant age and size
MATERIALS AND METHODS
The M. croceipes culture used in these studies was originally collected in Tift County,
Georgia, and was maintained in the laboratory for approximately 3 years prior t) the
study. Parasitoids were reared according to the procedure described by Lewis & Burton
(1970); newly emerged adults were held for 2 to 4 days before being used in field
experiments. H. zea used in all studies were from the Tifton, Georgia laboratory colony
developed and reared in accordance to the procedure described by Young et al. (1976).
Larvae were dissected by the procedure reported by Lewis & Brazzel (1966) to
determine levels of parasitism. The observed distribution of parasitoid eggs among host
larvae, and of parasitoid females searching in field plots, were compared to expected
values generated from the Poisson series and tested for significant departures from
randomness by the chi-square goodness-of-fit test (Wadley 1967).
Female M. croceipes released in the field tests were held with males for 1 to 2 days
at a 1:1 ratio in mating cages, as described by Lewis & Burton (1970), and were provided
8 Florida Entomologist 72(1) March, 1989
a 50:50 water:honey solution as food. The releases were made between 8:00 and 9:00
a.m. Paper cups (0.405 liter) in which H. zea larvae had fed on artificial diet were placed
(top open) at the base of plants at the release sites. Shell vials (1.8 ml) that contained
two to four female parasitoids each were smeared at the brim with a small amount of
frass from the cups to reinforce their host-seeking behavior (Gross et al. 1975). The
vials were then placed gently into the cups, and the parasitoids were left to disperse at
their own rate.
Two separate studies that involved the field release of laboratory reared M.
croceipes were conducted. The first was the release of a single density in soybeans,
while the second was a paired comparison of the release of two density levels of M.
croceipes on plots of soybeans and peanuts. The C. nigriceps studies were conducted
on naturally occurring populations of this parasitoid on Dolichos lablab L. (hyacinth
bean) and Desmodium purpureum (Mill.) (Florida beggarweed).
Microplitis croceipes Experiment 1
This study was conducted during late July in four plots of soybeans eight rows wide
by 61 m long (ca. 446 m2 each) with four rows of fallow area between plots.
Soybean plots were infested with first instar H. zea at a density of five per row
meter at 7 and again 5 days prior to release of parasitoids. In total, 72 M. croceipes
females were released, 18 per plot, at 12 evenly spaced locations (three locations/plot).
Heliothis zea larvae were collected from each of the four plots 0, 1 and 4 days after
release of the parasitoids and dissected to determine the level of parasitism.
Microplitis croceipes Experiment 2
This experiment was a paired study of the performance of M. croceipes within 15.2
X 15.2 m plots of soybeans and peanuts at two release rates. Four plots of each crop
were planted alternately with 15.2 m of fallow area between and on all sides of the plots.
Third and fourth instar H. zea larvae that had been fed soybean or peanut foliage,
respectively, for 2 days, were released in the plots 1 day prior to parasitoid release.
One hundred larvae were released over three rows in each of eight evenly spaced
locations in each plot. Either 12 or 32 M. croceipes females were released the next day
in each plot at four locations (three or eight females per location) per plot. H. zea larvae
were then collected 1 and 2 days after the parasitoid release and were dissected to
determine rates of parasitism.
The study was conducted first in early August and then repeated 3 weeks later using
the same procedure, except that 32 females were released in each plot on the later date.
Consequently, in the early August study there were two plots of each release density
for each host plant, while four plots of each plant were available for the one release
density in late August.
Complete records of the occurrence and number of M. croceipes eggs present in the
early August study were maintained and the distributions of eggs and searching
parasitoids were compared to random distributions calculated using the Poisson series.
Cardiochiles nigriceps Studies
Hyacinth bean was interplanted with corn in a 1.2-ha field in alternate four-row
strips with four-row fallow borders between. Beggarweed was volunteer growth in an
abandoned 2.0-ha watermelon field.
Adult parasitoid counts were made in 3 m X 3 m sections selected at random through
the fields. Two or three people took positions around the location using care not to
disturb the movements of the parasitoids, and instantaneous counts were made of the
number of parasitoids searching the area at a selected moment. The count reflected only
the number of searching females. Males of M. nigriceps do not exhibit hovering search
behavior and are distinguished readily with a little experience. The counts were made
Insect Behavioral Ecology-'88 Lewis & Gross
during fair weather at approximately 2:00 p.m., the time of peak searching'activity
(Lewis et al. 1972, Snow & Burton 1967).
Heliothis spp. larvae were collected at random throughout the field to assess
parasitism levels. Although most of the larvae were H. virescens at the time of these
studies, the identity of collected larvae was confirmed by the presence of microspines
on the setigerous tubercles (Brazzel et al. 1953). The proportion of parasitized larvae
was determined by dissection. Prior to dissection, larvae were held on diet for several
days so that the developing parasitoid could be found more easily.
A subsequent study was conducted in another hyacinth bean field to assess specific
aspects of the searching behavior of the ovipositing females. Sixteen searching females
were maintained under continuous visual observation for up to 10 min (7.3 min average).
Data on the number of stops made, location of the stops, hosts found, and attempted
stings were recorded. The hosts found at the landing sites were collected, identified to
species, and held to determine parasitism.
RESULTS AND DISCUSSION
Microplitis croceipes Experiment 1
Eleven percent of the larvae of H. zea collected from soybean plots were parasitized
in this experiment on day 1, and 47.5% were parasitized by day 4. Only 2 of 36 third-
and fourth-instar H. zea collected from throughout the plots on the morning of release
were parasitized, indicating that parasitism was primarily by the released M. croceipes.
Release of the 72 females over the 1,785 m2 (0.178 ha) plot is equivalent to ca. 404/ha.
Microplitis croceipes Experiment 2
The release density of 12 females per plot of soybeans and peanuts in early August
is equivalent to ca. 520 females per ha. Resulting average rates of parasitism of H. zea
TABLE 1. PARASITISM LEVELS (%) OF H. ZEA LARVAE ON SOYBEANS AND
PEANUTS BY LABORATORY-REARED M. CROCEIPES. LARVAE WERE
COLLECTED 1 AND 2 DAYS AFTER RELEASE OF 12 OR 32 PARASITOIDS
Day 1 Day 2 Day 1 Day 2
after release after release after release after release
12 9/ 32 9/ 12 9/ 32 9/ 12 9/ 32 9/ 12 9/ 32 9/
plot plot plot plot plot plot plot plot
23 37 31 48 0 0 14 0
11 8 18 44 0 0 0 38
17 23 25 46 0 0 7 19
42 33 39 50
58 35 64
33 39 -
50 67 55 00
- 50 42
Florida Entomologist 72(1)
larvae in soybeans were 17% on day 1 and 25% on day 2 (Table 1). This release density
and resulting parasitism are very close to those for experiment 1. The release of 32
females per plot is equivalent to 1,385 females per ha and yielded 46% parasitism by
The performance of the parasitoids released in early August on peanuts was much
weaker than in soybeans, with no parasitism being detected in the peanut plots on day
1, and a low level in only half of the plots on day 2 (Table 1).
Parasitoid performance improved in the late August study, particularly on peanuts.
Parasitism rates of 46 and 41% were obtained by day 1 for the soybean and peanut
plots, respectively. Although recapture of host larvae was limited and not available in
all plots, an increase of 71% parasitism was shown for the peanut plots by day 2. The
reason for the greatly improved performance by the parasitoid on peanuts is not known,
but recent data by Drost (1986) show marked differences in the chemically mediated
attractiveness of H. zea larvae feeding on different ages of the same plant. Also, it is
important to note that some recycling of natural parasitoids and populations from the
earlier release could have increased natural parasitism in late August.
Analysis of Microplitis croceipes Egg Distribution
The distribution of M. croceipes eggs among the dissected hosts of the early August
release on soybeans is presented in Table 2. The data show that the distribution of the
eggs is remarkably similar to that expected from the Poisson distribution. This is an
interesting and somewhat surprising result given the apparently nonrandom searching
behavior of M. croceipes females.
Cardiochiles nigriceps Studies
Counts of the number of searching C. nigriceps females in the hyacinth bean plots
were made on 3 separate days (in 7-10 locations per day). Populations were consistent
for all 3 days with mean numbers of females per sampling area per day of 2.2, 2.0, and
2.1, respectively, with an overall mean of 2.04. Each sampling area (9.3 m2) represented
1/1,076 ha. Therefore, we estimate that there was a population of ca. 2,195 C. nigriceps
females per ha in the hyacinth beans. Of 39 larvae collected on the 2nd and 3rd day of
the count, 29 (74%) were parasitized by C. nigriceps in the hyacinth bean.
Counts of the searching females per unit area (9.3 m2) in the beggarweed were made
on two consecutive dates in September at 15 locations in the field. The number of
females averaged 1.40 and 1.25 per location on the two dates with an overall mean of
1.31 females/unit area. Thus, we estimated a population of 1,410 females/ha.
In total, 104 larvae of H. virescens were collected from the beggarweed on the two
dates. Examination revealed 35 of the larvae (34%) were parasitized by C. nigriceps.
Observed Searching Behavior of C. nigriceps Females
Sixteen females that were observed searching for a total of 117 min made 79 land-
ings. Larval hosts were present at 44 of these landing sites, of which 55% were identified
as H. virescens. Eight apparent stings were made during these 44 examinations. These
eight larvae were collected and held to determine whether they had been parasitized.
Three of these larvae were H. zea and were not parasitized by C. nigriceps. Of the five
H. virescens larvae, two were not parasitized successfully as they pupated. The remain-
ing three larvae produced C. nigriceps. These data are similar to those reported by
Lewis et al. (1972) for this parasitoid on cotton in that the parasitoid is highly efficient
at locating sites occupied by the host larvae, but is able to sting only a small proportion
of the available larvae once the site is found (approximately one to two per hour of
search time in this case).
Insect Behavioral Ecology-'88 Lewis & Gross
TABLE 2. DISTRIBUTION OF M. CROCEIPES EGGS ON H.ZEA LARVAE COLLECTED
FROM SOYBEANS 1 AND 2 DAYS AFTER PARASITOID RELEASE AS INDI-
CATED AND AT INDICATED RELEASE RATE AS COMPARED TO THAT
EXPECTED BY RANDOM DISTRIBUTION. NO X2 VALUE WAS SIGNIFI-
CANT AT P= 0.05.
No. of hosts with indicated no. of
No. eggs Observed Expected Chi-square
Day 1 32 parasitoids per plot 0.40
0 32 32.7
1 10 9.0
2 1 1.3
3+ 0 0.2
Day 2 32 parasitoids per plot 2.91
0 17 17.5
1 13 11.2
2 1 3.3
3+ 2 1
Day 1 -12 parasitoids per plot 2.39
0 36 37.0
1 8 6.6
2+ 0 0.4
Day 2 -12 parasitoids per plot 1.01
0 23 23.0
1 7 5.5
2+ 0 0.6
Frequency of C. nigriceps in Sampling Areas
The distribution of C. nigriceps females among unit sampling areas of the two host
plants is described well by the Poisson series (Table 3). Of course, the distribution of
search females could vary according to a number of factors such as variations in quality
of plants, distribution of hosts, and size of sampling area. However, as in the case of
the distribution of parasitoid eggs among the host larvae discussed above, the appar-
ently random pattern is of interest since the host searching behavior of these parasitoids
is known to be mediated by a number of sophisticated guidance mechanisms (Drost et
al. 1986, Lewis & Tumlinson 1988, Jones et al. 1971, Vinson & Lewis 1965). On the
other hand, if the hosts are distributed at random over the field, a corresponding distri-
bution of the parasitoids might be expected.
Calculation of Comparative Efficiencies
To expedite comparisons of the efficiencies of the two parasitoids on different host
plants, the observed data were used to calculate the number of females that would be
Florida Entomologist 72(1)
TABLE 3. DISTRIBUTION OF FEMALES OF C. NIGRICEPS WITHIN SAMPLING
AREAS COMPARED TO THAT PREDICTED BY A RANDOM DISTRIBUTION
HYPOTHESIS. NO X2 VALUE WAS SIGNIFICANT AT P=0.05.
Area sampling frequency with number females indicated
females Observed Expected Chi-square
Desmodium purpureum 1.14
0 10 9.5
1 12 12.3
2 6 8.0
3+ 7 5.2
Dolichos lablab 1.72
0 2 3.0
1 9 6.1
2 4 6.2
3+ 9 8.7
required to deliver 50 and 80% parasitism for the different situations. Results presented
in Table 2 indicate that M. croceipes ovipositions are distributed randomly among the
available hosts. Indications of such a pattern were reported by Lewis et al. (1972) for
C. nigriceps. Consequently the Poisson series can be used to derive the number of
females required to obtain a given percentage of parasitism under similar situations
Microplitis croceipes on Soybeans. The data for M. croceipes on soybeans indicate
that 404 females per ha yielded 48% parasitism over 4 days in experiment 1 and 1,344
females per ha provided 46% parasitism in 1 day for the late August release of experi-
ment 2. For comparison, we decided to make extrapolations on the basis of 4-day
According to the Poisson series, the number of hosts receiving 0 eggs is equal to ex,
where x is the mean number of eggs deposited per host larva available. Consequently,
in the case of experiment 1, we assumed that the 404 M. croceipes females oviposited
0.65 eggs/host larva to yield the 48% parasitism. Fifty and 80% parasitism, respectively,
requires 0.69 eggs/host and 1.6 eggs/host (note the use of these figures in subsequent
calculation). It follows that 0.69/0.65 (404) or 429 females/ha would yield 50% parasitism,
while 1.6/0.65 (404) or 994 females/ha would give 80% parasitism under the conditions
of experiment 1.
In the case of experiment 2, we first estimated how much parasitism would have
occurred from the 1,385 females during 4 days. According to the Poisson series, ovipos-
ition of 0.62 eggs/host larva was necessary for the 46% parasitism observed on 1 day.
At that density, 0.62 (4) or 2.48 eggs would have been laid per host over 4 days and
yielded 92% parasitism. Thus, 0.69/2.48 (1,385) or 385 females/ha would give 50%
parasitism, and 1.6/2.48 (1,385) or 894 females/ha of M. croceipes would provide 80%
parasitism during 4 days under those conditions with soybeans. Thus, both of these
experiments provided similar results, showing that during a 4-day period on soybeans
ca. 400 M. croceipes females/ha yield 50% parasitism, and that ca. 900-1,000 M. croceipes
females/ha would yield 80% parasitism of a similar population of host insects on soy-
Cardiochiles nigriceps Extrapolations. The data from hyacinth beans show that a
C. nigriceps female population of 2,196/ha yielded 74% parasitism of a host population
requiring oviposition of 1.34 eggs/host larva present in the search area according to the
Poisson series. Therefore, a population of 0.69/1.34 (2,196) or 1,131 females/ha and
Insect Behavioral Ecology-'88 Lewis & Gross 13
1.6/1.34 (2,196) or 2,622 females/ha would be required for 50 and 80% parasitism, respec-
tively under similar conditions on the plant.
We estimate that 1,410 females/ha gave 34% parasitism on the beggarweed, which
would require 0.42 eggs/host. Consequently, a population of 0.69/0.42 (1,410) or 2,316
female C. nigriceps/ha would give 50% parasitism, and 1.6/0.42 (1,410) or 5,371 females/
ha would be required for 80% parasitism.
Lewis et al. (1972) estimated that 400-600 C. nigriceps females/acre (988-1,482/ha)
are required for 80% parasitism on cotton. These findings indicate that C. nigriceps is
considerably less efficient at parasitizing hosts on the hyacinth bean than on cotton and
even less efficient on beggarweed than on hyacinth bean.
The findings of these studies, together with the report of Lewis et al. (1972), indicate
that the efficiencies of M. croceipes and C. nigriceps generally are similar and vary
considerably among host plants. The number of parasitoids per ha required to achieve
80% parasitism is estimated to be ca. 900-1,500 females per ha for M. croceipes on
soybeans and C. nigriceps on cotton during a 4-day period. The number required was
found to increase to ca. 2,500 and 5,000 per ha, respectively, for hyacinth bean and
Moreover, the different levels of parasitism produced by M. croceipes on peanuts
indicate that their performance on a given plant may change significantly at different
times in the season. The differences may result from numerous factors, such as volume
of plant material, the ratio of vegetative to fruiting structures of the plants, varying
chemical make-up of the plants, and from increased numbers of parasitoids resulting
from the seasonal cycling of generations.
These variations in the efficiency of Heliothis parasitoids are important considera-
tions in designing and appraising pest management strategies involving their use. Only
by understanding the factors that govern performance of parasitoids can we maximize
their efficiencies in different situations.
We are grateful to Raydene Johnson and Lisa Hill for assistance in conducting the
experiments, and to M. A. Keller for helpful suggestions in design of the studies and
in preparing the manuscript.
This article reports the result of research only. Mention of a proprietary product
does not constitute endorsement or the recommendation for its use by USDA.
In cooperation with the University of Georgia, College of Agriculture Experiment
Station, Coastal Plains Experiment Station, Tifton, GA 31793.
BRAZZEL, J. R., L. D. NEWSOM, J. S. ROUSSEL, C. LINCOLN, W. J. WIILLIAMS,
AND G. BARNES. 1953. Bollworm and tobacco budworms as cotton pests in
Louisiana and Arkansas. Louisiana Agric. Expt. Stn. Tech. Bull. 482, 47 p.
DANKS, H. V., R. L. RABB, AND P. S. SOUTHERN. 1979. Biology of insect parasites
of Heliothis larvae in North Carolina, USA. J. Georgia Ent. Soc. 14: 36-64.
DROST, Y. C., W. J. LEWIS, P. 0. ZANEN, AND M. A. KELLER. 1986. Beneficial
arthropod behavior mediated by airborne semiochemicals. I. Flight behavior and
influence of preflight handling of Microplitis croceipes (Cresson). J. Chem. Ecol.
GROSS, Jr., H. R., W. J. LEWIS, RICHARD L. JONES, AND DONALD A. NORDLUND.
1975. Kairomones and their use for management of entomophagous insects: III.
Stimulation of Trichogramma achaeae, T. pretiosum, and Microplitis croceipes
Florida Entomologist 72(1)
with host-seeking stimuli at time of release to improve their efficiency..J. Chem.
Ecol. 1: 431-38.
HOPPER, K. R., AND E. G. KING. 1984. Preference of Microplitis croceipes
(Hymenoptera: Braconidae) for instars and species of Heliothis (Lepidoptera:
Noctuidae). Environ. Ent. 13: 1145-50.
JONES, R. L., W. J. LEWIS, M. C. BOWMAN, M. BEROZA, AND M. A. BIERL. 1971.
Host-seeking stimulant for parasite of corn earworm: Isolation, identification,
and synthesis. Science 173: 842-43.
LEWIS, W. J., AND J. R. BRAZZEL. 1966. Biological relationships between Car-
diochiles nigriceps Viereck and the Heliothis complex. J. Econ. Ent. 59: 820-23.
- AND 1968. A three-year study of parasites of the bollworm and the
tobacco budworm in Mississippi. J. Econ. Ent. 61: 673-76.
--AND R. L. BURTON. 1970. Rearing Microplitis in the laboratory with Heliothis
zea as hosts. J. Econ. Ent. 63: 656-58.
- A. N. SPARKS, R. L. JONES, AND D. J. BARRAS. 1972. Efficiency of Car-
diochiles nigriceps as a parasite of Heliothis virescens on cotton. Environ. Ent.
-- AND J. H. T UMLINSON. 1988. Host detection by chemically mediated associa-
tive learning in a parasitic wasp. Nature 331: 257-59.
MUELLER, T. E., AND J. R. PHILLIPS. 1983. Population dynamics of Heliothis spp.
in spring weed hosts in southeastern Arkansas and stage-specific parasitism.
Environ. Ent. 12: 1846-50.
NEUNZIG, H. H. 1963. Wild host plants of the corn earworm and the tobacco budworm
in eastern North Carolina. Environ. Ent. 56: 135-39.
NORDLUND, D. A., W. J. LEWIS, S. B. VINSON, AND H. R. GROSS, Jr. 1986. Be-
havioral manipulation of entomophagous insects, p. 104-105. in Theory and tactics
of Heliothis population management: I Cultural and biological control. South.
Coop. Ser. Bull. 316.
SNOW, J. W., AND R. L. BURTON. 1967. Seasonal occurrence of the Heliothis complex
on Desmodium perpureum with observations on parasitism by Cardiochiles nig-
riceps. J. Georgia Ent. Soc. 2: 47-52.
VINSON, S. B., AND W. J. LEWIS. 1965. A method of host selection by Cardiochiles
nigriceps. J. Econ. Ent. 58: 869-71.
WADLEY, F. M. 1967. Experimental statistics in entomology. Graduate School Press,
USDA, Washington, D.C., 133 p.
YOUNG, J. R., J. J. HAMM, R. L. JONES, W. D. PERKINS, AND R. L. BURTON. 1976.
Development and maintenance of laboratory colony of the corn earworm. U. S.
Dept. Agric. Tech. Bull. ARS-S-110, 9 p.
Insect Behavioral Ecology-'88 Antolin
GENETIC CONSIDERATIONS IN THE STUDY OF
ATTACK BEHAVIOR OF PARASITOIDS, WITH
REFERENCE TO MUSCIDIFURAX RAPTOR
MICHAEL F. ANTOLIN
Department of Biological Science
Florida State University
Tallahassee, FL 32306-2043
Genetic variation in attack behaviors of parasitoids is taxonomically widespread, and
evidence for genetic variation comes from two sources: from comparisons among strains
and from selection experiments. Most workers on genetics of attack in parasitoids treat
attack behaviors as traits with relatively simple inheritance. Attack behaviors are com-
plex traits that have polygenic inheritance that can be complicated by linkage and
genetic correlation with other traits, and by genotype-environment interaction. Ignor-
ing the genetic complexity of attack behavior may limit the effectiveness of searching
for new strains of parasites and may confound selection for favorable traits. I demon-
strate this point through a literature review, and through my research on the genetics
of Muscidifurax raptor Girault & Sanders, a pteromalid that attacks the pupae of flies
that breed in animal filth. Attack rates, functional response, the numbers of eggs laid
per patch, sex ratio, and development times vary among five geographic strains of M.
raptor. Furthermore, brood size, sex ratio, and development time are positively corre-
lated. This tangle can be unravelled by careful quantitative genetic experiments, where
genetic correlation and genotype-environment interaction can be measured.
La variaci6n gen6tica en el comportamiento del ataque de los parasitoides esta
ampliamente distribuida taxon6micamente y la evidencia de la variaci6n gen6tica pro-
viene de dos fuentes: de las comparaciones entire razas y de experiments de selecci6n.
La mayoria de los que trabajan en la gen6tica del ataque en los parasitoides tratan el
comportamiento del ataque como caracteristicas con una herencia relativamente sen-
cilla. Los comportamientos del ataque son caracteristicas complejas de herencia polig6n-
cia las cuales pueden ser complicadas por el ligamiento y la correlaci6n genetica con
otras caracteristicas, y por la interacci6n genotipo-medio ambiente. Ignorar la com-
plejidad gen6tica del comportamiento del ataque puede limitar la efectividad de buscar
nuevas razas de parasitos y puede confundir la selecci6n do caracteristicas favorables.
Yo demuestro este punto a trav6s de una revision de la literature y atrav6s de la
investigaci6n sobre la gen6tica de Muscidifurax raptor Girault y Sanders, un
pteromalido el cual ataca la pupa de los moscas que crecen en feces de animals. El
porcentaje de ataque, la respuesta funcional, el tamafo de la cria, la raz6n del sexo, y
el tiempo de desarrollo varian entire cinco razas geograficas de M. raptor. Ademas, el
tamafo de la cria, la raz6n del sexo y el tiempo de desarrollo estan positivamente
correlacionados. Este enredo puede ser clarificado con experiments geneticos cuan-
titativos cuidadosamente elaborados, donde la correlaci6n genetica y la interacci6n
genotipo-medio ambiente puedan ser medidos.
Entomologists have cast their nets far and wide in search of natural enemies for
biological control of insect and plant pests (Ehler & Andres 1983). The aims of biological
Florida Entomologist 72(1)
control workers have been to match a pest in a particular environment with the natural
enemy that is the most "adapted" to the pest in that environment (Remington 1968,
Mackauer 1972, 1976, Messenger et al. 1976, Scorza 1983). The natural enemies best
adapted to the pest situation will reduce pest populations and possibly become estab-
lished permanently. The question of adaptednesss" has naturally led to discussion of the
population genetics and genetic variability of natural enemies (Legner & Warkentin
1985, Unruh et al. 1986). The effectiveness of a predator or parasitoid in biological
control may be determined by which genotypes are released, and a genotype's effective-
ness may vary among the environments where it is released.
The exploitation of genetic variability within and among natural populations of insect
predators and parasitoids for improved biological control has been considered in every
phase of the search for natural enemies. When choosing a parasite for a particular pest,
host specificity and tolerance to climatic conditions at the area of release is of first
importance (Ehler & Andres 1983, Lenteren 1986). "Co-adaptation" and "coevolution"
of the predator and prey may need to be considered if pests are able to develop resist-
ance to their natural enemies (Hokkanen & Pimentel 1984, Bouletreau 1986). During the
collecting phase, questions of where in a species' range the insects should be collected
arise. Most workers conclude that collections should be made at the center of the range
because of the expectation that populations at the species' margin are genetically de-
pauperate and therefore unlikely to become established (Remington 1968, Mackauer
1972, 1976, but see Lucas 1969, Myers & Sabath 1980, and Simberloff 1986). Once a
colony is established, maintaining genetic variability in laboratory cultures and transfer-
ring "fit" parasites into the field becomes the challenge (Mackauer 1972, Messenger et
al. 1976, Legner & Warkentin 1985, Unruh et al. 1983, 1986). Finally, the question
arises whether predator and parasite performance can be improved through selection
for particular traits or by cross-breeding among strains (Hoy 1979, Roush 1979, Legner
1972, 1988a,b, Legner & Warkentin 1985, Havron et al. 1987).
These ideas are all predicated, however, on fairly complete knowledge of the popu-
lation genetics of the natural enemies, particularly of the genetics of the parasites'
attack behaviors. Although there is growing evidence of genetic variation for these
traits, clearly more knowledge is needed. Behavioral traits typically are controlled by
many genes and the inheritance of polygenic traits is complicated by dominance relation-
ships among the alleles that control the traits, linkage and genetic correlations among
genes, epistasis (e.g., the expression of a gene depends on its genetic background), and
by differential expression of the traits in different environments (Falconer 1981). In this
paper, I plead the case for further and more detailed studies that take into consideration
that attack behavior is not a single trait, but a suite of traits that will be affected by a
combination of both the underlying genetics and the environment. To illustrate this
point, I report some data from my ongoing studies of the genetics of Muscidifurax
raptor Girault and Sanders (Hymenoptera: Pteromalidae), a parasitoid of pupae of vari-
ous dung-inhabiting flies. Specifically, I demonstrate variation among five geographic
strains in attack rates, brood size per patch of hosts, sex ratio, and development time-
traits that may alter M. raptor's effectiveness in biological control.
BIOLOGY OF MUSCIDIFURAX RAPTOR
Muscidifurax raptor has been useful in biological control of house flies and stable
flies at poultry and beef operations (Patterson & Rutz 1986) and the species has a
cosmopolitan temperate distribution except for Asia (Kogan & Legner 1970). The fly
hosts pupate in dry areas of the manure or in the soil nearby, and M. raptor attacks
pupae found in the top 5 cm. of the habitat (Legner 1977, Morgan et al. 1981). Mus-
czdifurax raptor is solitary, usually one egg is laid and one adult emerges per host.
Insect Behavioral Ecology-'88 Antolin 17
Adult females mate and begin to oviposit immediately if males and hosts are available.
If hosts are provided throughout the adult period, females usually live for about 20 days
(although they can live for as long as 53 days), parasitize about 15 hosts per day for the
first 10-15 days of their lives, and have a total fecundity of more than 150 eggs (Coats
1976, Morgan et al. 1979, Wylie 1967, 1971).
MATERIALS AND METHODS
The five strains of M. raptor originate from different parts of the world and were
obtained from laboratories in the United States: West Germany (WG) and Hungary
(HU) from the USDA Insects Affecting Man and Animals Laboratory (IAMARL) in
Gainesville, FL; Israel (IS) from the University of California, Riverside; Nebraska
(NE) from the USDA at the University of Nebraska, Lincoln; and New York (NY) from
Cornell University, Ithaca, New York. The strains were maintained on houseflies raised
on CSMA larval medium (Ralston Purina). The housefly strain originated from the
IAMARL in Gainesville, FL.
Before testing, female parasitoids were standardized using Wylie's (1979) procedure.
Parasitized pupae were isolated in glass screw-top test tubes before emergence. After
emergence, females were given males and honey for 24 hrs, then were presented with
10 hosts for the next 24 hrs and finally were left in the tubes with honey only for a third
24 hr period. This three-day standardization allows a wasp to mature eggs and gain
ovipositional experience before testing (Wylie 1979, Antolin & Williams 1989).
Of the 164 females that were included in this study, 43 produced all-male broods and
are assumed to have been unmated. Unmated females were most common during a
3-week period when females were not isolated before emergence, but were allowed to
emerge into a large vial with hundreds of other females and males. Apparently, at high
densities the mutual interference among males prevents successful matings of many
females (see Assem et al. (1980) for a careful description of mating behavior in M.
raptor). These females were removed from the vial after the first 24 hrs, and were then
standardized as usual with host feeding in isolation.
Testing and observation were carried out between July 1987 and May 1988. Tests
were made in small arenas that have round (11 cm diameter, 95 cm2 area) plaster-of-
Paris bases and 10 cm high clear plastic sides. The bases of the containers were smeared
with used larval medium, because the parasitoids are attracted to larval cues during
host-searching (Murphy 1982), and were moistened with 20 ml of distilled water im-
mediately before each trial. Five, ten, or twenty uniformly sized 2-3 day-old hosts were
arranged in a regular pattern in the center of the arenas. Lone females were introduced
into each arena in large plastic vials that were inverted over the hosts. The vial was
removed when females settled down and began parasitizing. The position and activity
of each female in the arena was noted at five minute intervals, until the female stopped
searching for hosts and left the arena. Up to twelve females could be observed simul-
taneously this way by scanning over the arenas. No female spent less than the five
minute scanning period attacking a host.
Host from the experiments were placed individually into cells of tissue-culture plates
for rearing in an environmental chamber at 25"C, 70% RH, and 14:10 L:D. During
testing, temperature and relative humidity (RH) were 24.70C + 1.65 (mean a s.d.)
and 70.1% + 3.46, respectively. The emergence date and sex of each offspring was
noted. Pupae that had been attacked but did not yield houseflies or parasitoids were
dissected to determine whether an egg had been deposited or whether the eggs, larvae
or pupae did not complete development. The sizes of the puparia of 303 of the 1014
stung hosts were measured on a dissecting microscope with an ocular micrometer.
Florida Entomologist 72(1)
Data from 121 mated and 43 virgin females were included in the analyses. Analyses
of host-feeding and host size, and comparisons between mated and unmated females
were carried out using pooled data from all five strains. Sample sizes for each strain
for mated femles were WG: 31, HU: 23, IS: 24, NE: 19, and NY: 24, and for unmated
females sample sizes were WG: 9, HU: 3, IS: 10, NE: 16, and NY: 5. Females' perform-
ance in the number of pupae attacked, the number of eggs laid, the proportion attacked,
and the time spent in the arenas were analyzed by analysis of covariance (ANCOVA,
Sokal & Rohlf 1981). Pupal density was used as the covariate in the analyses. In each
case, a data transformation was necessary to make the data normal with equal variances
among treatments. Numbers of pupae attacked and offspring were loge transformed,
proportion attacked data were arcsine square root transformed, and time spent in the
arenas was square root transformed.
Frequency data, including the proportion of pupae that were used for host-feeding
and the sex ratio among offspring, were analyzed using log-likelihood statistics (G2) for
log-linear models and logistic regression (Sokal & Rohlf 1981, Bishop et al. 1975). Log-
linear models differ from analysis of variance (ANOVA), where tests are made on the
statistical significance of particular effects of a model. Models are fitted by iteration,
where expected frequencies are calulated by computer so that the marginal frequencies
of expected and observed data agree [see Sokal & Rohlf 1981, p. 748]. In log-linear
analysis, tests are made for the goodness-of-fit of the data to models and models are
rejected if they provide a significantly greater G2 statistic than expected by chance.
Particular effects of a fitted model can be tested by subtraction of two models that differ
only by a single main effect or interaction in a way that is analogous to ANOVA.
RESULTS AND DISCUSSION
HOST-FEEDING AND HOST SIZE
Muscidifurax species host-feed on the fluids that exude from the oviposition hole
through the puparium, and host-feeding is necessary for egg maturation (Legner &
Gerling 1967, Antolin & Williams 1989). Jervis & Kidd (1986) categorized M. raptor
host-feeding as non-concurrent, meaning that different hosts are used for host-feeding
and oviposition. This was not the case in this experiment. Of the 1,014 pupae that were
killed by M. raptor, 23.1% of the hosts were used for host-feeding and 87.6% had eggs
laid on them. Host-feeding and oviposition were completely independent (G2 = 0.19,
d.f., P>0.65); host-feeding was as common on pupae used for oviposition (203/888) as
on those that were destroyed without oviposition (31/126). Comparisons among the five
geographic strains by log-linear models demonstrated no differences in the frequencies
of host-feeding (G2 = 1.38, 4 d.f., P>0.50) or oviposition (G2 = 0.68, 4 d.f., P>0.90)
among strains. Hosts used for feeding were not larger or smaller than other hosts (both
6.06 mm, t = 0.070, 301 d.f., P>0.90).
Some parasitoids allocate the sex of their offspring according to host quality or size
(Waage 1986), but Wylie (1967) found no host size effects in M. raptor's close relative
M. zaraptor Kogan and Legner. In the present study, hosts that were used for feeding
by M. raptor did not yield relatively more or fewer male and female offspring: host-feed-
ing was independent of sex allocation (G2 = 5.36, 2 d.f., P>0.06). The sizes of pupae
that yielded male offspring (6.06 0.27mm), female offspring (6.06 0.24mm), or
inviable offspring (6.11 + 0.15mm) were not different (F2,244 = 0.602, P>0.50).
Insect Behavioral Ecology-'88 Antolin
ATTACK BEHAVIOR OF MATED AND UNMATED FEMALES
Mated females attacked 50% more pupae than did virgin females (Fig. 1), although
neither mated nor unmated M. raptor attacked all pupae in the arenas. When offered
five, ten or twenty hosts, mated females killed 2.8 + 1.7, 5.9 1.7, and 10.4 + 1.6
pupae, respectively, whereas unmated females killed significantly fewer, only 2.3 +
2.0, 3.6 + 2.4, and 7.6 2.5 pupae (F1,161 = 21.7, P<0.001). The slope of the relation-
ship between the number of pupae attacked and pupal density was significantly positive
(F1,161 = 111.7, P<0001), but the slopes of mated (0.082, transformed scale, r2 = 0.506)
and unmated females (0.075, r2 = 0.240) were the same (F1,16o = .183, P>0.65).
Only 87.6% of the pupae that were stung and killed were used for oviposition (Fig.
1). Again, mated females laid approximately 50% more eggs than did unmated females
(F1,161 = 15.3, P<0.001) and both mated and unmated females laid more eggs per patch
as pupal density increased (F1,161 = 89.1, P<0.001). At densities of 5, 10, and 20 pupae,
mated females laid 2.3 1.7, 4.7 1.8, and 8.6 + 1.8 eggs per patch, unmated females
laid 2.0 1.9, 3.3 + 2.5, and 6.3 + 2.8 eggs per patch.
Muscidifurax raptor displays an inversely density-dependent functional response
(Propp & Morgan 1985). In this study, the proportion attacked (the functional response)
declined slightly with pupal density (slope = -0.007, F,161 = 4.01, P<0.05). Mated
females attacked about 55% of the pupae before leaving the arenas, whereas virgin
females attacked only about 40% (Fig. 2); this difference is significant (F1,161 = 128,
P<0.001). The parasitism rates found in this study match those described for M. raptor
in the laboratory by Legner (1967, 1979b) and Morgan et al. (1979). Propp & Morgan
(1985) described a more steeply negative functional response for M. raptor from a field
experiment in which laboratory-reared housefly pupae were exposed to natural
parasitism at three poultry farms. Hunting by M. raptor is inversely density-dependent
even when it is not constrained by a laboratory set-up.
15 I 15
U STUNG U STUNG
0 EGGS 0 E EGGS
5. 10. 20. 5. 10. 20.
PUPAL DENSITY PUPAL DENSITY
Fig. 1. The number of hosts attacked by M. raptor divided into those used for
oviposition (eggs) and those killed (stung) without eggs laid on them. Mated females
killed more hosts than did unmated females.
Florida Entomologist 72(1)
0 5 10
Fig. 2. The functional responses (proportion attacked) of
mated and unmated
The average number of minutes required to successfully parasitize a host varied
(mean = 26.7, s.d. 23.3). Mated females spent on average 123.6 + 2.9, 207.5 + 5.8,
and 285.6 10.8 minutes in the arenas with 5, 10 and 20 pupae, whereas unmated
females spent only 96.6 11.2, 97.8 16.2, and 29.8 + 19.4 minutes in the arenas
(Fig. 3). The difference between mated and unmated females was statistically significant
(F1,161 = 13.41, P<0.001). Oddly, the females did not increase the time spent in the
arenas proportionately to pupal density. Females given 20 pupae did not spend four
times as long as females given 5 pupae, although the increase with pupal density was
statistically significant (Fi.16o = 73.63, P<0.001). The time spent in the arena could be
divided into the time spent walking and searching for pupae (no female ever took flight),
and the time spent on the pupae, or the handling time. Although handling times were
fairly long, the attack was fairly efficient; while in the arenas mated females spent most
of their time (79.3%) actually attacking pupae, unmated females spent 81.2% of their
time attacking (Fig. 3).
Insect Behavioral Ecology-'88 Antolin 21
U SEARCH U SEARCH
0 U HANDLING O U HANDLING
5. 10. 20. 5. 10. 20.
PUPAL DENSITY PUPAL DENSITY
Fig. 3. Time spent in the arenas by unmated and mated females. The total time was
divided into the time spent searching in the arenas and time spent handling the hosts.
THE SEARCH FOR GENETIC VARIATION IN ATTACK BEHAVIOR
Evidence for genetic variation in attack behavior in hymenopterous parasitoids is
taxonomically widespread and comes from two sources: comparisons among geographic
strains and selection experiments. Simply demonstrating genetic variation for attack
behaviors still tells us little about the probable success or failure of any release for
biological control, but knowledge of genetic variation is an important first step. The
remainder of this paper reports variation among the five strains of M. raptor, reviews
other evidence for genetic variation in attack behaviors of parasitoids, spells out the
advantages and pitfalls of careful genetic testing of parasitoids, and points the way to
how more detailed information on the genetic level could help predict the success or
failure of natural enemies in the control of insect pests.
COMPARISONS AMONG STRAINS
Variation among strains in traits that affect the attack behaviors of parasitoids have
been reported numerous times. Strain differences in lifetime fecundity and sex ratio
have been described in M. raptor, M. zaraptor, and M. uniraptor Kogan & Legner
(Legner 1979a, 1988c) and Nasonia vitripennis (Walker) [Pteromalidae] (Parker & Or-
zack 1985, Fried & Pimentel 1986). Parasitism rates in patches of Ephestia kuehniella
Zell. eggs vary among several Trichogramma species and among strains within several
of the species (Pak & Heiningen 1985), and egg-clumping in patches of E. kuehniella
eggs of differs among strains of Trichogramma maidis Pintereau & Voegel6 [Trichog-
rammatidae] (Chassain & Bouletreau 1985). Bouletreau (1986) recently reviewed the
extensive research on variation in rates of failed parasitism among geographic strains
of cynipid endoparasites of Drosophila, and also reviewed the work of Pimentel, Chab-
ora, and colleagues on strains of Nasonia vitripennis attacking several fly species.
Florida Entomologist 72(1)
Because the effectiveness of a parasitoid is influenced by a number of traits, I
studied attack rates, eggs laid per patch, time spent per patch, sex ratio and develop-
ment time of five strains of M. raptor. In this part of the study, comparisons among
strains were possible only for mated females. The five strains attacked different num-
bers of pupae (Table 1, Fig. 4), but the slope of the relationship between number
attacked and pupal density did not differ among the strains. The strain that attacked
the highest number, WG, attacked twice as many pupae as did NY, the strain that
attacked the fewest, with the other strains spaced in between. The proportion attacked
(i.e., the functional response) also differed among strains (Table 1), but the only signif-
icant difference was between the strain with the highest attack rate, WG (63% attacked)
and the lowest, NY (44% attacked). Strain differences in the number of eggs laid per
patch echo those of the numbers of pupae attacked with one exception: the strain that
laid the fewest eggs per patch was HU, which laid significantly fewer eggs than the
strain that laid the most, WG (Table 1).
STRAIN COMPARISONS AMONG MATED FEMALES: NUMBER OF PUPAE
ATTACKED PER PATCH, NUMBER OF EGGS LAID PER PATCH, AND THE
PROPORTION OF PUPAE ATTACKED. ANALYSES ARE OF LOG, (NUMBER
ATTACKED AND NUMBER OF EGGS LAID) AND ARCSINE SQUARE ROOT
(PROPORTION ATTACKED) TRANSFORMED DATA.
ANCOVA, P<0.05, ** P<0.01
Number Number Proportion
Attacked of Eggs Attacked
r2 0.584 0.511 0.134
Source d.f. MS F-value MS F-value MS F-value
Equal 4 0.264 1.091 0.513 1.666 0.068 0.890
Strain 4 1.307 5.376 ** 1.128 3.578** 0.269 3.540**
Pupal 1 31.376 129.088 31.314 99.317** 0.366 4.803*
Error 115 0.243 0.315 0.076
TABLE lB. SLOPES, AND INTERCEPTS ADJUSTED BY THE COMMON SLOPE. INTER-
CEPTS FOLLOWED BY THE SAME LETTER ARE NOT SIGNIFICANTLY
DIFFERENT FROM EACH OTHER AT THE 0.05 LEVEL, DETERMINED BY
THE TUKEY-KRAMER MSD PROCEDURE (SOKAL AND ROHLF 1981, P.
Number Number Proportion
Strain Attacked of Eggs Attacked
slope = 0.082 slope = 0.082 slope =-0.007
N intercept intercept intercept
NY 24 0.376 a 0.198 ab 0.814 a
HU 23 0.414 ab 0.149 a 0.863 ab
IS 24 0.769 bc 0.531 ab 1.023 ab
NE 19 0.815 bc 0.567 ab 1.024 ab
WG 31 0.852 c 0.596 b 1.033 b
Insect Behavioral Ecology-'88 Antolin
Fig. 4. The numbers of hosts attacked by the five geographic strains of M. raptor;
only mated females were included in this analysis. See TABLE 1 for analysis and
No statistically significant differences among strains could be detected in either the
total time spent in the arenas (F4,115 = 1.243, P>0.29), or in the time spent handling
the pupae (F4,115 = 1.115, P>0.35). The data were quite variable: the maximum times
spent in the arenas at each pupal density were about 10 times greater than the minimum
The overall sex ratio was 27.6% males and sex ratios varied greatly both with pupal
density and among strains (Fig. 5). Only the 103 females that had greater than 75%
survival of their offspring were included in the analysis (the sample sizes for each strain
were NY: 22, NE: 15, IS: 20, HU: 20, and WG: 26). The data were analyzed using
log-linear models (Bishop et al. 1975) by considering each egg laid by a female as an
independent trial of a binomial experiment where progeny can be either male or female.
This analysis is necessary on proportional data with many values near 1.0 or 0 (see
Trexler et al. 1988). Many females in this experiment laid only female eggs in a patch,
which would have a sex ratio of 0.0% males. For the five strains of M. raptor, models
Florida Entomologist 72(1)
- - -NY
- - HU
Fig. 5. Sex ratios of the five geographic strains of M.
raptor. See TABLE 2 for
that included strain effects, pupal density, and both strain and density with their in-
teraction provided good fits to the data, indicating that sex ratio in M. raptor differs
among strains and across pupal densities, and that sex ratio for each strain depends on
the density of the patch where a female is ovipositing (Table 2).
Sex allocation within a sequence of eggs laid during an oviposition bout is not random
(Fig. 6). The first egg is usually female, the second, third and fourth eggs have a greater
probability of being male, but after the fourth egg the sex ratio becomes increasingly
skewed toward females. The method of Green et al. (1982) was used to show that the
sex ratio pattern is more precisely determined than would be expected by chance (x2
= 73.34, 101 d.f., P<0.05). With a sequence of oviposition where males are laid early
in the sequence, variation in sex ratio among strains and across pupal densities would
arise if strains lay different numbers of eggs in a patch-parasitoids that lay few eggs
per patch would have a more male-biased sex ratio. Some of the variability in sex ratio
among the strains could arise if each strain has a different sequence of sex allocation
during an oviposition bout, but I was unable to detect statistical differences in oviposi-
tion sequences among the strains.
0 5 10 15 20
Insect Behavioral Ecology-'88 Antolin
TABLE 2. LOG-LINEEAR MODELS AND GOODNESS-OF-FIT STATISTICS FOR SEX
RATIO. NON-VIABLE EGGS OR LARVAE (31 OF 546 EGGS) HAVE BEEN
EXCLUDED FROM THE ANALYSIS. S PARASITOID STRAIN, D = PUPAL
Model G2 d.f. P
S + D + SxD 5.380 9 >0.60
S + D 4.331 5 >0.50
S 3.985 4 >0.40
D 0.192 1 >0.80
Development time of M. raptor is approximately 19 to 22 days at 25C (Legner
1979b, Morgan et al. 1981), males develop more quickly than do females (Fig. 7; F1,455
= 54.671, P<0.001), and there are significant differences among the strains in develop-
ment time (F4,455 = 13.068, P<.001). Further analysis by the Tukey-Kramer MSD
procedure (Sokal & Rohlf 1981, p. 507) showed that the difference arises because NE
and WG developed more quickly than did IS, HU, and NY.
Although the strains tested here are laboratory strains and possibly subject to in-
breeding depression, the differences among the strains are not likely due to this cause.
Muscidifurax raptor is not very susceptible to inbreeding depression: Fabritius (1984
5 10 15
Fig. 6. The sex ratio of eggs laid during the sequence of oviposition (bottom) and
the cumulative sex ratio through the sequence of oviposition (top).
Fig. 7. Development times (at 250C) of the
Error bars are + 1 s.e.
five geographic strains of M. raptor.
found that a stock of M. raptor that was maintained each generation by four mated
pairs of siblings showed little inbreeding depression in longevity, fecundity and sex
ratio over thirty generations. Forty-seven generations were required for the population
to fail completely. In the present study, the WG and IS strains have been in culture
for hundreds of generations, while the HU, NY and NE strains were begun in 1986 or
1987. The "old" strains were not necessarily the poorer ones: HU and NY, two of the
"newer" strains, had the lowest attack rates, smallest brood sizes and slowest develop-
ment times, while the "old" strain WG was just the opposite.
These five geographic strains differ in several traits that may govern their effective-
ness as biological control agents. Variation among geographic strains provides evidence
of genetic variation within a species, but this variation tells us little about the popula-
tions that were the sources of the strains, so that recommendations for which strain
would be the best should be approached with caution. Variation among strains, if it is
indeed genetic and due to genes on the chromosomes, may arise in several ways [see
Legner (1987) and Werren (1987a,b) for a description of the effects of microorganisms
and cytoplasmic genes on trait variation]. First, the observed differences could repre-
Florida Entomologist 72(1)
I I I I I
NE WG IS HU NN
Insect Behavioral Ecology-'88 Antolin
sent true differences in the source populations and represent the genes that are
"adapted" in each environment. However, strain differences could arise under other
circumstances, even if the genetic compositions of the source populations were the
same. The differences could arise by chance via founder effects when the colonies were
established (e.g., different genotypes were sampled from each source population). Dif-
ferences could also arise by chance via genetic drift within the colonies. For example,
even if all the genetic variability were originally sampled, rare genes could be lost
during culturing because of population bottlenecks or because of small effective popula-
tions (Unruh et al. 1983, 1986) Also, culturing procedures may inadvertently select for
differences among the strains, so that even if populations at different part of the species
range originally have the same gene frequencies, after a number of generations the
geographic strains may differ from each other.
SELECTION EXPERIMENTS AND GENETIC CORRELATIONS
Successful artificial selection of a trait is a second line of evidence for genetic vari-
ation in parasitoids. Selection experiments are "cleaner", in that when a trait responds
to selection this is unequivocal evidence that there is genetic variation for the trait
(Falconer 1981). Selection for high and low sex ratio has been reported three times; in
Mastrus carpocapsae (Cushman) [Ichneumonidae] (Simmonds 1947), in Dahlbominus
fuliginosus (Nees) [Eulophidae] (Wilkes (1964), and in Nasonia vitripennis (Parker &
Orzack (1985). Wilkes (1947) also successfully increased the fecundity of D. fuliginosus,
Ram & Sharma (1977) were able to increase lifetime fecundity, but not sex ratio in
Trichogramma fasciatum (Perkins). Selection for temperature tolerance has been car-
ried out in Dahlbominus fuscipennis (Zetterstedt) (Wilkes 1942), Aphytis lingnanensis
Compare [Aphelinidae] (White et al. 1970), and Trichogramma pretiosum Riley (Ashley
et al. 1974). DDT resistance has been improved through selection in Macrocentrus
ancylivorus Rohwer [Braconidae] (Pielou & Glasser 1952). Rosenheim & Hoy (1986)
reported the selection of pesticide resistance in twelve field populations of Aphytis
melinus DeBach, where overall pesticide resistance in the populations was correlated
with the history of pesticide use at the sites where the populations were collected.
Selection for favorable traits may have drawbacks. For example, in M. raptor,
selection of a more female-biased sex ratio could result in too few males emerging to
successfully inseminate all of the females in a strain. This may affect laboratory rearing.
In addition, the effectiveness of a severely female-biased strain in the field may be
limited because the resulting unmated females would have lower attack rates than
would mated females. Not only would the unmated females have lower attack rates,
but they would produce only male offspring in the Fl generation, so that the lowered
effectiveness might cascade through several generations.
Linkage or genetic correlation between the selected trait and other traits may cause
other traits to change as well and may limit the response to selection (Falconer 1981).
Ashley et al. (1974) found that strains and hybrids of Trichogramma pretiosum selected
for high temperature tolerance had lower parasitism of host eggs in field cages. On the
other hand, if traits are positively correlated selection on one trait may result in gains
in other traits as well. In the present study of M. raptor, brood size, sex ratio and
development time were positively correlated. Larger broods were more female-biased:
the correlation between brood size (in broods of five or more offspring) and sex ratio
(arcsine square root transformed) was -0.352 (N = 46, P<0.05). Female M. raptor that
lay more female-biased broods have female offspring that develop faster (r= 0.316, N
= 46, P<0.05) and females that lay larger broods have progeny of both sexes that
develop faster (males: r=-0.304, N = 82, P<0.01; females: r=-0.232, N = 111,
P<0.05). If these traits are correlated because they are affected by the same block of
Florida Entomologist 72(1)
genes, selection for one of the traits may provide additional benefits because other traits
may become improved as well.
Another consideration is whether the expression of traits in a number of strains or
genotypes change in the same direction when an environmental variable changes (such
as temperature or humidity), or whether the response to the environment varies among
strains (Force 1967, Falconer 1981). This phenomenon is generally known as genotype-
environment interaction (also see Bulmer 1985). For example, Tingle & Mitchell (1975)
tested two strains of M. raptor, a laboratory strain and a field strain, in three environ-
mental conditions. They found that the strains had equal parasitism levels at low tem-
peratures but the lab strain performed slightly better at higher temperatures. Develop-
ment times were just the opposite. Both strains had similar development times at high
temperatures, but the lab strain developed much more slowly at lower temperatures
than did the field strain. Although genotype-environment interactions are cited as a
mechanism that promotes genetic variability in natural populations (Mukai 1988), these
sorts of effects might be undesirable in parasitoids released for biological control in an
area with widely variable climatic conditions.
Clearly, ample genetic variation in attack behaviors exists in parasitoids, but how
to exploit this variation is an important question that remains largely unresolved. Most
of the writing about genetic aspects of introduction and release of insect parasitoids
used in biocontrol has been under the assumption that attack behaviors are inherited
independently of the other attributes of the insect, and that simple considerations of
genetic variability will provide valuable improvements in collection methods, rearing
practices, and control of pests in the field. This has not proved to be the case: laboratory
hybridization and selection experiments have not generally produced better control,
and spectacular successes have been recorded even when the founder populations have
been small and presumably genetically depauperate (Lucas 1969, Myers & Sabath 1980,
Simberloff 1986, Legner & Warkentin 1987).
But does this mean that genetics can add nothing new to the field, and that the "trial
and error" methods are still the best solution (Lenteren 1980)? Certainly, it does not.
When importing natural enemies for control of exotic pests or when augmenting native
parasitoid populations to control native hosts, attempts should be made to release the
natural enemy that is best "adapted" to specific conditions, either by releases of suitable
strains or by improving performance via selection. But any genetic work should be
undertaken with the realization that "adaptation" likely is influenced by a number of
traits whose inheritance is polygenic and may be complex. The complexity of traits like
attack behaviors could be sorted out through quantitative genetic experiments (Fal-
coner 1981, Bulmer 1985). Careful quantitative genetic experiments could be used to
evaluate the amount and form of genetic variation in attack behaviors, the genetic
correlations among these traits and other life-history attributes, and the environmental
influences on the expression of genes that control attack behaviors. When more is
known of the genetic control of the behaviors that influence effectiveness of biocontrol
agents, we can learn more about some of the failures of biological control and perhaps
better predict the success of future efforts.
Insect Behavioral Ecology-'88 Antolin
Strains were kindly provided by E. F. Legner, P. B. Morgan, J. J Peterson, and
D. A. Rutz. I especially thank P. B. Morgan for good advice on how to rear the
parasitoids and flies, and for cheerful encouragement. D. Simberloff, P. Stiling, S.
Strauss and J. Travis read the manuscript, and I thank J. Travis for use of his SYSTAT
program for data analysis and for producing the figures. Editorial assistance from H.
Frank and E. McCoy was invaluable. I was supported by funds from the Department
of Biological Science, FSU, a Grant-in-aid from Sigma Xi, and Mini-grants and Scholar-
ships provided by the Florida Entomological Society.
ANTOLIN, M. F., and R. L. WILLIAMS. 1989. Host-feeding and egg production in
Muscidifurax zaraptor (Hymenoptera: Pteromalidae). Florida Ent. 72: 129-134.
ASHLEY, T. R., D. GONZALEZ, AND T. F. LEIGH. 1974. Selection and hybridization
of Trichogramma. Environ. Ent. 3: 43-48.
ASSEM, J. VAN DEN, M. J. GIJWIJT, AND B. K. NUBEL. 1980. Observations on
courtship and mating strategies in a few species of parasitic wasps (Chalcidoidea).
Netherlands. J. Zool. 30: 208-27.
BISHOP, Y. M. M., S. E. FIENBERG, AND P. W. HOLLAND. 1975. Discrete Mul-
tivariate Analysis. MIT Press, Cambridge, MA.
BOULETREAU, M. 1986. The genetic and coevolutionary interactions between
parasitoids and their hosts. Pp. 169-200. in Insect Parasitoids, Waage, J. K. and
D. J. Greathead [eds.]. Academic Press, New York.
BULMER, M. G. 1985. The Mathematical Theory of Quantitative Genetics. Clarendon,
CHASSAIN, C., AND M. BOULETREAU. 1987. Genetic variability in the egg-laying
behaviour of Trichogramma maidis. Entomophaga 32: 149-57.
COATS, S. A. 1976. Life cycle and behavior of Muscidifurax zaraptor (Hymenoptera:
Pteromalidae). Ann. Ent. Soc. America 69: 772-80.
EHLER, L. E., AND L. A. ANDRES. 1983. Biological control: exotic natural enemies
to control exotic pests. Pp. 395-48. in Exotic Plant Pests and North American
Agriculture, Wilson, C. L., and C. L. Graham [eds.]. Academic Press, New
FABRITIUS, K. 1984. Investigations on the inbreeding of Muscidifurax raptor under
laboratory conditions (Hymenoptera: Pteromalidae). Ent. Gen. 9: 237-41.
FALCONER, D. S. 1981. An Introduction to Quantitative Genetics, 2nd edn.,
Longman, New York.
FORCE, D. C. 1967. Genetics in the colonization of natural enemies for biological
control. Ann. Ent. Soc. America 60: 722-29.
FRIED, C. S., AND D. PIMENTEL. 1986. Host parasitization and progeny production
of two strains of the parasitoid Nasoni vitripennis (Hymenoptera: Pteromalidae)
an the house fly (Diptera: Muscidae). Environ. Ent. 15: 439-42.
GREEN, R. F., G. GORDH, AND B. A. HAWKINS. 1982. Precise sex ratios in highly
inbred parasitic wasps. American Nat. 120: 653-65.
HAVRON, A., D. ROSEN, Y. ROSSLER, AND J. HILLEL. 1987. Selection on the male
hemizygous genotype in arrhenotokous insects and mites. Entomophaga 32: 261-
HOKKANEN, H., AND D. PIMENTEL. 1984. New approach for selecting biological
control agents. Canadian Ent. 116: 1109-121.
HoY, M. A. 1979. The potential for genetic improvement of predators for pest manage-
ment programs. Pp. 106-115. in Genetics in Relation to Insect Management,
Hoy, M. A., and J. J. McKelvey [eds.]. Rockefeller Foundation, New York.
Florida Entomologist 72(1)
JERVIS, M. A., AND N. A. C. KIDD. 1986. Host-feeding strategies in hymenopteran
parasitoids. Biol. Rev. 6: 395-434.
KOGAN, M., AND E. F. LEGNER. 1970. A biosystematic revision of the genus Mus-
cidifurax (Hymenoptera: Pteromalidae) with descriptions of four new species.
Canadian Ent. 102: 1268-90.
LEGNER, E. F. 1967. Behavior changes the reproduction of Spalangia cameroni, S.
endius, Muscidifurax raptor and Nasonia vitripennis (Hymenoptera:
Pteromalidae) at increasing fly densities. Ann. Ent. Soc. America 60: 819-26.
1972. Observations on hybridization and heterosis in parasitoids of synan-
thropic flies. Ann. Ent. Soc. America 65: 254-63.
1977. Temperature, humidity and depth of habitat influencing host destruction
and fecundity of muscoid fly parasites. Entomophaga 22: 199-206.
1979a. Prolonged culture and inbreeding effects on reproductive rates of two
pteromalid parasites of muscoid flies. Ann. Ent. Soc. America 72: 114-18.
S1979b. The relationship between host destruction and parasite reproductive
potential in Muscidifurax raptor, M. zaraptor, and Spalangia endius (Chal-
cidoidea: Pteromalidae). Entomophaga 24: 145-52.
S1987. Further insights into extranuclear influences on behavior elicited by
males in the genus Muscidifurax (Hymenoptera: Pteromalidae). Proc. Papers
55th Annl. Conf. California Mosquito Vector Control Assoc.: 127-30.
1988a. Hybridization in principal parasitoids of synanthropic Diptera: the
genus Muscidifurax (Hymenoptera: Pteromalidae). Hilgardia 56 (4): 1-36.
1988b. Quantation of heterotic behavior in parasitic Hymenoptera. Ann. Ent.
Soc. America 81: 657-81.
1988c. Studies of four thelytokous Puerto Rican isolates of Muscidifurax un-
iraptor (Hymenoptera: Pteromalidae). Entomophaga 33: 269-80.
--, AND D. GERLING. 1967. Host-feeding and oviposition on Musca domestic
by Spalangia cameroni, Nasonia vitripennis and Muscidifurax raptor
(Hymenoptera: Pteromalidae) influences their longevity and fecundity. Ann. Ent.
Soc. America 60: 678-91.
AND R. W. WARKENTON. 1985. Genetic improvement and inbreeding effects
in culture of beneficial arthropods. Proc. Papers 52nd Annl. Conf. California
Mosquito Vector Control Assoc.: 156-61.
LENTEREN, J. C. VAN. 1980. Evaluation of control capabilities of natural enemies:
does art have to become science? Netherlands J. Zool. 30: 369-81.
1986. Parasitoids in the greenhouse: successes with seasonal inoculative re-
lease systems. Pp. 342-74. in Insect Parasitoids, Waage, J. K., and D. J. Great-
head [eds.]. Academic Press, New York.
LUCAS, A. M. 1969. The effect of population structure on the success of insect intro-
ductions. Heredity 24: 15-57.
MACKAUER, M. 1972. Genetic aspects of insect production. Entomophaga 17: 27-48.
S1976. Genetic problems in the production of biological control agents. Annl.
Rev. Ent. 21: 369-85.
MESSENGER, P. S., F. WILSON, AND M. J. WHITTEN. 1976. Variation, fitness, and
adaptability of natural enemies. Pp. 209-31. in Huffaker, C. B., and P. S. Mes-
senger [eds.], Theory and Practive of Biological Control, Academic Press, New
MORGAN, P. B., D. E. WEIDHAAS, AND G. C. LABREQUE. 1979. Host-parasite
relationship of the house fly, Musca domestic, and the microhymenopteran
pupal parasite, Muscidifurax raptor G.& S. (Diptera: Muscidae and Hymenopt-
era: Pteromalidae). J. Kansas Ent. Soc. 52: 276-81.
AND R. S. PATTERSON. 1981. Programmed release of Spalangia
endius and Muscidifurax raptor (Hymenoptera: Pteromalidae) against estimated
populations of Musca domestic (Diptera: Muscidae). J. Med. Ent. 18: 158-66.
MUKAI, T. 1988. Genotype-environment interaction in relation to maintainences of
genetic variability in populations of Drosophila melanogaster. Pp. 31-30, in Weir,
B. S., E. J. Eisen, M. M. Goodman, and G. Namkoong [eds.], Proceedings of the
Second International Conference on Quantitative Genetics. Sinauer Associates,
Insect Behavioral Ecology-'88 Antolin
MURPHY, S. T. 1982. Host-finding behaviour of some hymenopterous parasitoids of
Musca domestic. Ann. Appl. Biol. 101: 148-51.
MYERS, J. H., AND M. D. SABATH. 1980. Genetic and phenotypic variability, genetic
variance, and the success of establishment of insect introductions for the biolog-
ical control of weeds. Pp. 91-100, in DelFosse, E. S. [ed.], Proceedings of the V
International Symposium on the Biological Control of Weeds, CSIRO, Mel-
PAK, G. A., AND T. G. VAN HEININGEN. 1985. Behavioural variations among strains
of Trichogramma spp.: adaptability to field temperature conditions. Ent. Exp.
Appl. 38: 3-3.
PARKER, E. D., AND S. H. ORZACK. 1985. Genetic variation for the sex ratio in-
Nasonia vitripennis. Genetics 110: 93-105.
PATTERSON, R. S., AND D. A. RUTZ (eds). 1986. Biological Control of Muscoid Flies.
Misc. Publ. Ent. Soc. America 61: 1-174.
PIELOU, D. P., AND R. F. GLASSER. 1952. Selection for DDT resistance in a benefi-
cial insect parasite. Science 115: 117-18.
PROPP, G. D., AND P. B. MORGAN. 1985. Effect of host distribution on parasitism of
house-fly (Diptera: Muscidae) pupae by Spalangia spp. and Muscidifurax raptor
(Hymenoptera: Pteromalidae). Canadian Ent. 117: 515-24.
RAM, A., AND A. K. SHARMA. 1977. Selective breeding for improving the fecundity
and sex ratio of Trichogramma fasciatum (Perkins) (Trichogrammatidae:
Hymenoptera), an egg parasite of lepidopterous hosts. Entomon 2: 133-37.
REMINGTON, C. L. 1968. The population genetics of insect introduction. Annl. Rev.
Ent. 13: 415-26.
ROUSH, R. T. 1979. Genetic improvement of parasites. Pp. 97-105, in Hoy, M. A.,
and J. J. McKelvey [eds.], Genetics in Relation to Insect Management, Rockefel-
ler Foundation, New York.
ROSENHEIM, J. A., AND M. A. HOY. 1986. Intraspecific variation in levels ofpestiide
resistance in field populations of a parasitoid, Aphytis melinus (Hymenoptera:
Aphelinidae). J. Econ. Ent. 79: 1161-73.
SCORZA, R. 1983. Ecology and genetics of exotics. Pp. 395-41, in Wilson, C. L., and
C. L. Graham [eds.], Exotic Plant Pests and North American Agriculture,
Academic Press, New York.
SIMBERLOFF, D. 1986. Introduced insects: a biogeographic and systematic perspec-
tive. Pp. 3-26. in Mooney, H. A. and J. A. Drake [eds.], The Ecology of Biological
Invasions of North America and Hawaii, Springer, New York. Verlag.
SIMMONDS, F. J. 1947. Improvement of the sex-ratio of a parasite by selection. Cana-
dian Ent. 79: 41-44.
SOKAL, R. R., ANDJ. F. ROHLF. 1981. Biometry. 2nd edn. Freeman, San Francisco.
TINGLE, F. C., AND E. R. MITCHELL. 1975. Seasonal reproduction of Muscidifurax
raptor and Spalangia endius under simulated Florida conditions. Environ. Ent.
TREXLER, J. C., C. E. MCCULLOCH, AND J. TRAVIS. 1988. How can the functional
response best be determined? Oecologia 76: 206-14.
UNRUH, T. R., W. WHITE, D. GONZALEZ, G. GORDH, AND R. F. LUCK. 1983.
Heterozygosity and effective size in laboratory populations of Aphidius ervi
(Hym.: Aphidiidae). Entomophaga 28: 245-58.
--- AND R. F. LUCK. 1986. Electrophoretic studies of parasitic
Hymenoptera and implications for biological control. Pp. 150-63. in Patterson,
R. S. and D. A. Rutz [eds.], Biological Control of Muscoid Flies, Misc. Publ. Ent.
Soc. America 61: 1-174.
WAAGE, J. K. 1986. Family planning in parasitoids: adaptive patterns of progeny and
sex allocation. Pp. 63-95. in Waage. K. L. and D. Greathead [eds.], Insect
Parasitoids, Academic Press, New York.
WERREN, J. H. 1987a. Labile sex ratios in wasps and bees. Bioscience 37: 498-506.
1987b. The coevolution of autosomal and cytoplasmic sex ratio factors. J.
Theor. Biol. 124: 317-34.
Florida Entomologist 72(1)
WHITE, E. D., P. DEBACH, AND M. J. GARBER. 1970. Artificial selection for genetic
adaptation to temperature extremes in Aphytis lingnanensis Compere
(Hymenoptera: Aphelinidae). Hilgardia 40: 161-92.
WILKES, A. 1942. The influence of selection on the referendum of a chalcid (Micro-
plectron fuscipennis Zett.) and its significance in the biological control of an
insect pest. Proc. Roy. Ent. Soc. London (Ser. B) 130: 400-15.
1947. The effects of selective breeding on the laboratory propagation of insect
parasites. Proc. Roy. Ent. Soc. London (Ser. B) 134: 227-45.
1964. Inherited male-producing factor in an insect that produces its males
from unfertilized eggs. Science 144: 305-07.
WYLIE, H. G. 1967. Some effects of host size on Nasonia vitripennis and Mus-
cidifurax raptor (Hymenoptera: Pteromalidae). Canadian Ent. 99: 742-748.
1971. Observations on intraspecific larval competition in three hymenopterous
parasites of fly puparia. Canadian Ent. 103: 137-42.
1979. Sex ratio variability of Muscidifurax zaraptor (Hymenoptera:
Pteromalidae). Canadian Ent. 111: 105-109.
C-*- -- i- - -- -- - --* -- ^
CLUTCH SIZE, SEX RATIO AND MATING BY THE
POLYEMBRYONIC ENCYRTID COPIDOSOMA
M. R. STRAND
Department of Entomology
University of Wisconsin-Madison
Madison, WI 53706
Most parasitoids alter the clutch size and sex ratio of their progeny by facultatively
controlling the number of eggs they oviposit and fertilize. However, in polyembryonic
parasitoids clutch size and possibly sex ratio are also influenced by progeny develop-
ment, because the eggs of these parasitoids produce multiple offspring. In this study
the clutch size and sex ratio of the polyembryonic encyrtid Copidosoma floridanum
(Ashmead) were examined in relation to current theory. The majority of C. floridanum
broods collected from the field were comprised of males and females (i.e. mixed). No
differences were found in the secondary clutch sizes of all male, all female or mixed
broods. Mixed broods from the field and laboratory had median secondary sex ratios of
0.03 and 0.02 respectively. In mating experiments, 95% of females were mated when
the sex ratio was 0.03, suggesting that only enough males to mate all females were
produced in mixed broods. Laboratory studies demonstrated that C. floridanum pro-
duced a mixed brood by laying 1 female and 1 male egg. Thus, the highly female biased
sex ratio of a C. floridanum mixed brood was due to an alteration of the primary sex
ratio during the course of male and female egg development. Factors that might influ-
ence the sex ratio shift in mixed broods, and hypothetical conditions under which C.
floridanum might preferentially produce single sex and mixed broods are discussed.
La mayoria de parasitoides alteran facultativamente el tamafio de la masa de huevos
y la proporci6n del sexo de su prole, controlando el nfimero de huevos que ovipositan y
la fertilizaci6n de los mismos. Sin embargo en los parasitoides multiembri6nicos, el sexo
Insect Behavioral Ecology-'88 Strand
y el tamafo de la masa de huevos puede ser adicionalmente influenciado por el desarrollo
de la prole, porque estos parasitoides produce crias multiples. En este studio, el
tamafo de la masa de huevos y la proporci6n de sexos del encirtido multiembri6nico
Copidosoma floridanum (Ashmead) fueron comparados en relaci6n a la teoria actual.
La mayoria de las crias de C. floridanum colectatas en el campo estaban formadas por
una mezcla de hembras y machos. No se encontraron diferencias signicativas entire el
tamaio de las masas de huevos donde todas las crias fueron machos, hembras, o una
mezela. La proporci6n de sexos de las crias mixtas del campo y del laboratorio tuvieron
un intemedio de 0.03 y 0.02 respectivamente. En experiments de apareamiento, el 95%
de las hembras copularon cuando la proporci6n de sexos era0.03, lo que sugiere que
produce el nmmero just de marches para copular con las hembras de las crias mixtas.
En experiments de laboratorio se demostr6 que C. floridanum produce crias de ambos
sexos, depositando un huevo hembra y otro macho. La gran proporci6n de C.
floridanum hembras en las crias mixtas fue el resultado de la alteraci6n de la proporci6n
primaria del sexo durante el desarrollo de los huevos hembras y machos. Los factors
que pueden influir el cambio de la proporci6n de sexos en crias mixtas, y las condiciones
hipot6ticas donde C. floridanum produce uno o ambos sexos, son discutid6s en este
In recent years considerable theoretical interest has developed in understanding the
progeny and sex allocation strategies of parasitic Hymenoptera (Charnov et al. 1981,
Skinner 1985, Iwasa et al. 1984, Parker & Courtney 1984, Godfray 1987a, Strand &
Godfray 1989). The amount of resource available to each parasitoid offspring developing
in or on a host is a function of that host's quality and the number and sex of any other
progeny present. Thus, which hosts a parasitoid decides to parasitize, the clutch size
she lays, and for arrhenotokous species, the sex ratio she produces will all influence
progeny fitness (Waage 1986, Strand 1986). Recent studies indicate that at least some
parasitoids shift their clutch size and sex ratio in qualitative agreement with theoretical
predictions (Charnov 1982, Waage & Godfray 1985, Waage 1986, Godfray 1987b, Werren
1987, Strand 1988b).
The most prevalently studied factor influencing parasitoid sex ratios is Local Mate
Competition (LMC) (Hamilton 1967). Under conditions where males compete for mates
and mating takes place between offspring of one or a few mothers in isolated subpopu-
lations, the LMC model predicts that females will be selected to bias the sex ratio of
the eggs they lay towards females. The optimal sex ratio for a haplo-diploid organism
is described by the equation (n 1)(2n 1) /n (4n 1) (Hamilton 1979, Taylor & Bulmer
1980), where n is the number of females colonizing a patch of resource on which their
offspring mate randomly. Alternative descriptions have been forwarded to describe the
processes underlying LMC (Taylor 1981, Nunney 1985, Frank 1986), but the qualitative
predictions of the theory remain similar. Namely, when the level of LMC is low and n
is large an unbiased sex ratio is predicted, and when the level of LMC is high and n is
small a female biased sex ratio is predicted. When n = 1 and mating takes place exclu-
sively between siblings, a sex ratio of zero or only enough sons to ensure fertilization
of all daughters is the optimal sex ratio.
Although clutch size and sex ratio are usually studied independently (Waage 1986),
the two factors are obviously linked. The progeny of gregarious parasitoids or solitary
parasitoids of aggregated hosts often pupate in close proximity and mate before dispers-
ing. Under these conditions the level of LMC would be expected to be correlated with
clutch size (Griffiths & Godfray 1988, Strand 1988a). Thus, for a clutch size of 1, a sex
ratio of near 0.5 (proportion males) would be predicted since the possibility of sib-mating
would be very remote. However, as clutch size increases the possibility of sib-mating
also increases, and the sex ratio would be predicted to decline. If a single male could
mate all females, the predicted sex ratio would be the reciprocal of the clutch size.
34 Florida Entomologist 72(1) March, 1989
The vast majority of parasitoids alter the clutch size and sex ratios they produce by
facultatively controlling the number of eggs they oviposit and fertilize (Waage 1986).
However, for polyembryonic parasitoids clutch size and possibly sex ratio could also be
influenced by progeny development, because the eggs of these wasps produce multiple
offspring. The purpose of this study was to examine how LMC theory relates to polyem-
bryonic parasitoids. In the first portion of this paper I briefly describe the biology of
polyembryonic parasitoids. I will then compare some clutch and sex ratio data for the
polyembryonic encyrtid Copidosoma floridanum (Ashmead) with theoretical predic-
BIOLOGY OF POLYEMBRYONIC PARASITOIDS
Obligate polyembryonic development occurs in four families of parasitic Hymenop-
tera (Braconidae, Encyrtidae, Platygasteridae, Dryinidae) (Clausen 1940). Each of
these families is in a different superfamily suggesting that polyembryony has arisen
independently several times in the Hymenoptera. Because development within the dif-
ferent polyembronic families varies (Ivanova-Kasas 1972), I restrict my comments here
to the copidosomatine encyrtids. These wasps exhibit the most extreme form of polyem-
bryony with species producing broods of 50-2,000 adults per egg (Silvestri 1906, Patter-
son 1919, Leiby 1922, Stoner & Weeks 1976, Cruz 1986, Strand 1989a). It should be
noted that there is a considerable amount of inconsistency in the terminology used by
various authors in their descriptions of polyembryonic development. Much of this incon-
sistency is due to the timespan over which these studies have been conducted, but it is
also due to the fact that the development of these highly unusual insects does not
conform well to the events typically associated with insect embryogenesis (Retnakaran
& Percy 1985).
All polyembryonic encyrtids are minute (0.5-1.5 mm), egg-larval parasitoids of
Lepidoptera (Clausen 1940). After parasitism, the host egg hatches and the larva de-
velops up to its final instar. As the host development proceeds, the oviposited polyem-
bryonic egg divides by holoblastic cleavage producing aggregations of embryonic cells
enveloped by one or two serosal membranes. Collectively the embryonic and serosal
tissue is referred to as a polygerm (Silvestri 1906, Patterson 1921, Leiby 1922). The
polygerm itself then repeatedly subdivides, becoming distributed throughout the host's
body, and eventually forming a number of morulae. Each morula is enveloped by a
serosa, and large groups of morulae are held together by a second outer serosal mem-
brane. Such aggregates of morulae have been referred to as a polymorula or polyem-
bryonal mass (Leiby, 1922).
Several authors have noted that some polyembryonic encyrtids produce two larval
phenotypes from the same egg(s) originally laid into the host (Silvestri 1906, Patterson
1921, Doutt 1947, Cruz 1981, Strand 1989b). The first and by far most common
phenotype is the so called reproductive larva which develops and becomes an adult
wasp. Differentiation of morulae destined to form reproductive larvae begins after the
host molts to its ultimate instar (Strand 1989b, Strand et al. 1989). The reproductive
larvae consume the host and pupate within the remnant host larval cuticle forming a
"mummy". Other morulae give rise to the second phenotype which is referred to as a
precocious (= asexual) larva. Precocious larvae develop and are liberated from the
polyembryonic mass prior to the host's final instar. As few as 1 to as many as 100
precocious larvae per host have been reported (Parker & Thompson 1928, Fitzgerald
Simeone 1971, Cruz 1986, Strand 1989c). Precocious larvae do not molt, feed minimally
and die when the reproductive larvae consume the host (Cruz 1986, Strand 1989b). Cruz
(1981) suggests that precocious larvae function to defend the polygerm from interspecific
Insect Behavioral Ecology-'88 Strand
Polyembryonic encyrtids produce all male broods, all female broods and broods that
contain both sexes (i.e. mixed broods) in the laboratory (Patterson 1919, Leiby 1926,
Cruz 1986, Strand 1989a). Although it has been suggested that both males and females
of mixed broods could come from a common egg (Patterson 1919, Caltagirone 1970),
other data indicate that mixed broods are produced by the female laying more than one
egg into the host (Leiby 1926, Strand 1989a). Recent studies with C. floridanum de-
veloping in the soybean looper, Pseudoplusia includes (Walker), found that by observ-
ing female oviposition behavior, it was possible to determine the relationship between
the number and sex of the eggs a wasp lays into a host (hereafter referred to as the
primary clutch and sex ratio) and the number and sex of the emerging progeny (the
secondary clutch and sex ratio) (Strand 1989a). Using this procedure it was found that
C. floridanum females lay either 1 or 2 eggs per oviposition event. If 1 egg was laid
the emerging brood was all male or all female, but if 2 eggs were laid, the first egg laid
was always female, the second egg laid was always male and the emerging brood was
mixed. It was also found that females exhibited distinctly different oviposition gestures
when laying male (unfertilized) and female (fertilized) eggs.
Note that under these conditions all of the emerging females within a brood are
identical to one another as are all of the males. The secondary clutch size for C.
floridanum from P. includes in the laboratory was usually 1,000-1,300 for all male,
all female and mixed broods. However, while the primary sex ratio a female laid to
produce a mixed brood was 0.5 (1 male and 1 female egg), the mean secondary sex ratio
for an emerging mixed brood was 0.04 (Strand 1989b).
In the field C. floridanum likely experiences low levels of intraspecific competition
for hosts. P. includes moths lay their eggs solitarily on soybean leaves (Burleigh
1972). Thus, host eggs are highly dispersed on foliage, and wasp densities are probably
very low. C. floridanum discriminates against parasitized hosts (Strand, unpubl.) which
in the laboratory minimizes superparasitism. In laboratory observations of mixed
broods, C. floridanum males emerge immediately before females, and mating occurs at
the natal site before females disperse (Strand 1989a).
CLUTCH AND SEX RATIOS OF COPISODOMA FLORIDANUM
Assuming the complete absence of superparasitism, foundress number per host for
C. floridanum is 1, and LMC theory predicts a highly female biased sex ratio. However,
in the laboratory C. floridanum produces mixed and single sex broods. To determine
whether C. floridanum produces both brood types in the field and how they compare
with broods from the laboratory, the following study was conducted.
MATERIALS AND METHODS
P. includes larvae were collected in unsprayed soybean fields on the Edisto Experi-
mental Station, Blackville, SC. Field collections were taken weekly during September
and October 1987 using a standard beat cloth. Five samples 15 m apart were taken in
each soybean field with each sample consisting of one row meter of soybean plants. P.
includes larvae were brought to the laboratory and placed in 6 ml plastic cups half
filled with artificial diet according to the procedures of Greene et al. (1976). Larvae
were maintained at 280-2, 70% RH and 16L: 8D photoperiod until either pupation or
formation of a C. floridanum mummy. Mummies were placed individually in 18 cm glass
tubes in order to ensure that all of the progeny within a mummy were collected. After
progeny emergence the brood was killed in a 200C freezer, and the number and sex
of the wasps were counted.
Florida Entomologist 72(1)
The secondary clutch sizes and sex ratios of field-collected mixed broods were com-
pared with mixed broods produced in the laboratory. C. floridanum and P. includes
were reared in the laboratory as described by Strand (1989a). Individual wasps were
presented 24-h-old P. includes eggs, and their oviposition behavior was observed.
Based upon oviposition behavior (Strand 1989a), the primary clutch and sex ratio per
host was recorded. After egg hatching, the parasitized host larvae were placed individu-
ally in containers and reared on artificial diet. The mummies produced were placed in
glass tubes as described, and the number and sex of the emerging wasps were deter-
In total, 43 C. floridanum broods were collected from the field. Fifty-five percent
of the broods were mixed, 23% were all male, and 22% were all female. There was no
significant difference in the clutch size of the all male, all female and mixed broods
The secondary sex ratios of field-collected and laboratory reared mixed broods were
highly female biased with the median for each being 0.02 and 0.03 respectively (Fig. 1).
Three broods in both the field and laboratory samples had secondary sex ratios above
0.1, resulting in a positive skew for both samples. The majority of remaining broods
from the field and laboratory had sex ratios below 0.05. Excluding the broods with sex
ratios above 0.10, the mean sex ratio for the combined laboratory and field broods was
0.026 0.014. Although the secondary clutch sizes for field and laboratory broods
ranged from a low of 529 progeny to a high of 1,870, there was no correlation between
secondary clutch size and the arcsin transformed values for the secondary sex ratio
(y=6.2 x 105x +0.26; F(1,59)=1.06; p>0.25; r2=0.02) (Fig. 2). The bounds within
which 95% of observations are found if a sex ratio of 0.026 is binomially distributed are
also included in Fig. 2. Several broods fall outside of these bounds, suggesting that the
sex ratio variance is somewhat greater than binomial.
COPIDOSOMA FLORIDANUM MALE MATING CAPACITY
In the laboratory, the oviposition of 1 female and 1 male egg into a host results in
an emerging brood that is extremely female biased (Strand 1989a). It cannot be known
with certainty whether mixed broods from the field are also the progeny from individual
females; however, comparison of these broods with those from the laboratory support
such an assumption. The highly female biased sex ratio of C. floridanum mixed broods
is qualitatively consistent with the predictions of LMC theory. Reducing the number
of males in a brood would allow a larger number of females to be produced. Since all
males in a mixed brood are identical only enough males should be produced to mate all
females. To determine whether the number of adult males per mixed brood approaches
TABLE 1. NUMBER OF AND MEAN SECONDARY CLUTCH SIZE OF ALL MALE, ALL
FEMALE AND MIXED C. FLORIDANUM BROODS COLLECTED FROM THE
Brood type n Clutch size SD'
Male 10 1002.3 164.3
Female 9 1167.4 214.6
Mixed 24 1123.5 104.2
'(F (2, 37) = 0.76; p>0.25). One-way ANOVA
Insect Behavioral Ecology-'88 Strand 37
Field .,, ,, e
Laboratory "" ""* "-" *g *
0 0.01 0.02 0.03 0.04 0.05 0.1 0.2 0.4
Sex ratio (freq. males)
Fig. 1. Sex ratios for field and laboratory C. floridanummixed broods. The median
values for the two sets are indicated by the arrows. Note that while the majority of sex
ratios are < 0.05 three broods from the field and laboratory were above 0.10.
a value of only enough males to mate all females, experiments were conducted to esti-
mate the mating capacity of C. floridanum males.
MATERIALS AND METHODS
All male and all female C. floridanum broods were allowed to emerge in glass tubes
to ensure that only unmated males and females were used in the study. From 10-200
females were placed into 5.5 cm petri dishes at 250 + 2 and 50% RH. One male was then
introduced into each petri dish. Note that under these conditions 1/total number of
wasps per dish equals the sex ratio. The wasps were allowed to mate for 2 h which was
the time estimated for complete emergence of a C. floridanum mixed brood (Strand
unpubl.). To standardize the size of wasps used in the experiments, only individuals
with a hind tibia length of 0.3-0.4 mm long were used. After the 2 h mating period, the
petri dishes were placed into a 5C cooler. The females from each dish were then
dissected in saline, and their spermathecae examined by phase contrast microscopy for
the presence of sperm. The presence or absence of sperm rather than the absolute
number of sperm per female was the criterion used for whether a female was mated.
S 0.08- o laboratory
S 0.04- 0o o
0 0 o
0.02 o 0
U) 0O 0
400 800 1200 1600 2000
Fig. 2. The relationship between sex ratio and clutch size for laboratory and field
C. floridanum mixed broods. The solid lines show the bounds within which 95% of
broods lie if the sex ratio is binomially distributed with parameter 0.026.
Florida Entomologist 72(1)
Each male began to mate with females immediately after introduction into petri
dishes. In the unweighted regression, the values for the number of females per dish
were positively correlated with the arcsin transformed values for the frequency of
females per dish that were mated (y= -0.003x + 1.460, F(1,23)=52.8, p<0.005;
r"=0.67). Log transformation of the females-per-dish values slightly improved the
model fit (y= -0.654x + 2.312, F(1,23)=61.4; p<0.001; r"=0.73) (Fig. 3). Based on
the log transformed regression equation it was estimated that 95% of females were
mated when the ratio of females: males was 30:1. Taking the reciprocal of the total
number of wasps per dish, this value equaled a sex ratio of 0.033.
The vast majority of parasitic Hymenoptera shift their clutch sizes and sex ratios
by facultatively adjusting the number of eggs they fertilize and lay. Indeed, the compo-
nents of haplo-diploid sex determination and facultative control of oviposition and fertili-
zation have likely played a critical role in hymenopteran clutch and sex ratio evolution
(Charnov 1982, Waage 1986, Godfray 1987b, Werren 1987). Yet for polyembryonic
parasitoids like C. floridanum, clutch size and sex ratio are as influenced by progeny
development as they are by a female's facultative control of oviposition.
LMC theory predicts that the optimal brood sex ratio will shift with factors such as
foundress number, level of inbreeding, patch or host size, and host encounter rate
(Werren 1980, Waage & Ng 1984, Frank 1985, Herre 1985, Strand 1988a). Under com-
plete LMC where foundress number (n) equals 1, theory specifically predicts that only
enough sons to mate all daughters should be produced. Highly female biased sex ratios,
produced under complete LMC, have been reported (Hamilton 1967, Waage 1986, King
1987, Griffiths & Godfray 1988). For the parasitoid species which produce these sex
ratios, strong a priori reasons have been forwarded for why the majority of matings
1.00 0* *
0 0.40 -
0.5 1.0 1.5 2.0 2.5 3.0
Fig. 3. The relationship between the frequency of females mated and the number of
females present in petri dish experiments. A single male was present in each dish, and
the fit for the regression of log females x arcsin frequency of females mated is illus-
trated (y = 0.594x + 2.213, F(1,22) = 41.3; p<0.001; r2 = 0.69).
Insect Behavioral Ecology-'88 Strand
are assumed to occur between siblings. In each instance the female biased sex ratio is
a function of the proportion of eggs the ovipositing female fertilizes.
C. floridanum mixed broods are highly female biased as predicted, and as in other
studies, several factors contribute to the assumption that the majority of mating occurs
between sibilings. Wasps pupate within the remnant host cuticle, males usually emerge
before females, and mating occurs near the host mummy. Host patch size is effectively
1 since P. includes lay their eggs solitarily. At ca. 25C host eggs hatch in about 3
days, which combined with C. floridanum's discriminatory abilities, would minimize the
likelihood of superparasitism. The mating experiments support the prediction that only
enough males to mate all females are produced. The median sex ratio for laboratory
and field mixed broods were 0.03 and 0.02 respectively, and the sex ratio in which 95%
of females were mated was estimated to be 0.03.
What is unique for C. floridanum is that the highly female biased secondary sex
ratios are derived from a wasp laying 1 female and 1 male egg (i.e. a primary sex ratio
of 0.5). Strongly female biased sex ratios for the mixed broods of other polyembryonic
parasitoids are known (Patterson 1919, Leiby & Hill 1923, Leiby 1926, Flanders 1942,
Doutt 1947, Stoner & Weeks 1976, Cruz 1986). For Platygaster (Platygasteridae) and
Macrocentris (Braconidae) sp., there is limited evidence that all male broods are smaller
than all female broods, and that this difference may partially account for the female bias
in mixed broods (Leiby & Hill 1923, Leiby 1926, Clausen 1940, Flanders 1946). How-
ever, C. floridanum all male, all female and mixed broods are similar in size (Table 1),
and in previous studies no difference in development time was found between brood
types (Strand 1989a, Strand 1989b). Other studies of copidosomatine encyrtids that I
am aware of similarly do not indicate a sufficient difference in male and female brood
sizes or development time to account for the extreme female bias of mixed broods
(Patterson 1919, Leiby 1926, Doutt 1947, Cruz 1986). A second possibility for why such
a dramatic sex ratio shift occurs in mixed broods would be that female larvae outcompete
male larvae. This was specifically examined for C. floridanum, and was found to not
be responsible for the shift in sex ratio (Strand 1989b). Distortions of the primary sex
ratio by extrachromosomal or cytoplasmic factors have been reported or suggested for
a few gregarious parasitoids (Werren 1987). The possibility of such factors existing in
C. floridanum, and determining whether the sex ratios of C. floridanum from other
geographic regions are similar to those reported here are currently under investigation.
Regardless of the mechanism responsible for the sex ratio shift, LMC theory pro-
vides a plausible explanation for the sex ratios observed for C. floridanum mixed
broods. However, two features of C. floridanum oviposition that are not well explained
by LMC theory is the variance in mixed brood sex ratio and the fact that C. floridanum
frequently produces single sex broods. Green et al. (1982) have suggested that
parasitoids will be selected to produce precise sex ratios with less than binomial vari-
ance, because such a strategy results in more inseminated females per brood. This
occurs in at least some parasitoids by wasps laying precise sequences of male and female
eggs (Green et al. 1982, Waage & Ng 1984, Strand 1988b). C. floridanum females are
extremely precise in the primary clutch and sex ratio they lay (always laying 1 female
and 1 male egg when producing a mixed brood), but the data suggest that the variance
of the secondary sex ratio is greater than binomial (Fig. 3). It is possible that the three
field broods which had sex ratios above 0.10 were due to superparasitism (Fig. 1).
However, this is not supported by the laboratory data where three broods also had sex
ratios above 0.10, even though superparasitism was prevented. Other mixed broods had
very low sex ratios of < 0.01. Thus, while C. floridanum may have facultative control
over whether it lays 1 or 2 eggs per host, the mechanism responsible for the sex ratio
shift in mixed broods is not as precise. This could result in some broods having an excess
of males while in other broods insufficient males to mate all females may occur.
40 Florida Entomologist 72(1) March, 1989
Single sex broods have been reported for several polyembryonic parasitoids, and are
produced along with mixed broods in the same host generation (Leiby & Hill 1923,
Leiby 1926, Clausen 1940). Several factors could influence a polyembryonic parasitoid
to produce single sex or mixed broods. One possibility is host availability, and the effect
it could have on the interaction between clutch size and sex ratio. Assume that C.
floridanum is constrained to lay either 1 egg and produce a single sex brood or 2 eggs
and produce a mixed, female biased brood. Gain-rate clutch size models predict that if
time or eggs are limiting, wasps should adjust clutch size with host availability; laying
more eggs per host when encounter rates are low and fewer eggs when encounter rates
are high (Parker & Courtney 1984, Iwasa et al. 1984, Skinner 1985). Concurrently,
LMC conditions for C. floridanum may be highest when host density is low since the
chance of mating between non-siblings would be greatest if many parasitized hosts were
in close proximity to one another. Increasing primary clutch size does not alter the
secondary clutch size or sex ratio (Strand 1989a). Thus, when hosts are rare, laying one
female and one male egg per host may be the best strategy for producing a mixed brood
under extreme LMC. However, if hosts are common a female could produce more
progeny over her lifetime by reducing the primary clutch size per host to 1. It currently
is not known to what extent females from single sex broods are mated, but intuitively
it would seem that as the number of parasitized hosts within a given area increases,
the possibility of mating between broods would also increase.
A second factor which could influence field sex ratios is the mating status of the
wasp population. If significant numbers of females in the population are unmated and
constrained to produce all male broods, mated females will be selected to progressively
bias their broods toward females (Godfray & Shaw 1987). As mentioned, some females
from single sex broods or mixed broods with too few males could remain unmated. How
these factors might influence C. floridanum population structure and progeny allocation
are currently under investigation.
I would like to thank H. C. J. Godfray and D. W. Tonkyn for useful discussions on
parasitoid progeny allocation. I would like to further thank D. L. Mahr and R. L.
Lindroth for reviewing the manuscript, and J. A. Johnson, G. Carner and D. Boethel
for their assistance during field collections of C. floridanum. This study was supported
by University of Wisconsin Hatch Project 3200 to M. R. S.
BURLEIGH, J. G. 1972. Population dynamics and biotic controls of the soybean looper
in Louisiana. Environ. Ent. 1: 290-94.
CALTAGIRONE, L. E. 1970. Gynandromorphism in the polyembryonic encyrtid Pen-
talitomastix plethoricus Calt. (Hymenoptera: Encyrtidae). Bull. Lab. Ent. Ag-
raria Portici 28: 98-112.
CHARNOV, E. L. 1982. The Theory of Sex Allocation. Princeton Univ. Press.
CHARNOV, E. L., R. L. LOSDEN HARTOGH, W. T. JONES, AND J. VAN DEN ASSEM.
1981. Sex ratio evolution in a variable environment. Nature 289: 27-33.
CLAUSEN, C. P. 1940. Entomophagous Insects. McGraw Hill, New York.
CRUZ, Y. P. 1981. A sterile defender morph in a polyembryonic hymenopterous para-
site. Nature 294: 446-47.
CRUZ, Y. P. 1986. Development of the polyembryonic parasite Copidosomopsis tanyt-
memus (Hymenoptera: Encyrtidae). Ann. Ent. Soc. America 79: 121-27.
DOUTT, R. L. 1947. Polyembryony in Copidosoma koehleri Blanchard. American Nat.
Insect Behavioral Ecology-'88 Strand 41
FITZGERALD, T. D., AND J. B. SIMEONE. 1971. Polyembryony in Paraleurucerus
bicolosipes (Hymenoptera: Encyrtidae). Ann. Ent. Soc. America 64: 774-76.
FLANDERS, S. E. 1942. Sex differentiation in the polyembryonic proclivity of the
Hymenoptera. J. Econ. Ent. 35: 108.
FLANDERS, S. E. 1946. Control of sex and sex-limited polymorphism in the
Hymenoptera. Q. Rev. Biol. 21: 135-43.
FRANK, S. A. 1986. Hierarchical selection theory and sex ratios. I General solutions
for structured populations. Theor. Pop. Biol. 30: 215-31.
GODFRAY, H. C. J. 1987a. The evolution of clutch size in parasitic wasps. American
Nat. 129: 221-33.
GODFRAY, H. C. J. 1987b. The evolution of clutch size in invertebrates. Pp. 117-54,
in P. H. Harvey and L. Partridge [eds.], Oxford Surveys in Evolutionary Biol-
ogy, Vol. 4.
GODFRAY, H. C. J., AND M. R. SHAW. 1987. Seasonal variation in the reproductive
strategy of the parasitic wasp Eulophus larvarum (Hymenoptera: Chalcidoidea:
Eulophidae). Ecol. Ent. 12: 251-56.
GREEN, R. F., G. GORDH, AND B. A. HAWKINS. 1982. Precise sex ratios in highly
inbred parasitic wasps. American Nat. 120: 653-65.
GREENE, G. L., N. C. LEPPLA, AND W. A. DICKERSON. 1976. Velvetbean caterpil-
lar: a rearing procedure and artificial medium. J. Econ. Ent. 69: 487-88.
GRIFFITHS, N. T., AND H. C. J. GODFRAY. 1988. Local mate competition, sex ratio
and clutch size in bethylid wasps. Behav. Ecol. Sociobiol. 22: 211-217.
HAMILTON, W. D. 1967. Extraordinary sex ratios. Science 156: 477-88.
HAMILTON, W. D. 1979. Wingless and fighting males in fig wasps and other insects.
In: Reproductive Competition and Sexual Selection in Insects. M. S. Blum and
N. A. Blum (eds.) Academic Press, New York, pp. 167-220.
HERRE, E. A. 1985. Sex ratio adjustment in fig wasps. Science 228: 896-97.
IVANOVA-KASAS, O. M. 1972. Polyembryony in insects. Pp. 243-72, in Developmental
Systems, Insects Vol. 2. S. J. Counce and C. H. Waddington (eds.) Academic
Press, New York.
IWASA, Y., Y. SUZUKI, AND H. MATSUDA. 1984. Theory of oviposition strategy of
parasitoids. 1. Effect of mortality and limited egg number. Theor. Pop. Biol. 26:
KING, B. H. 1987. Offspring sex ratios in parasitoid wasps. Q. Rev. Biol. 62: 367-86.
LEIBY, R. W. 1922. The polyembryonic development of Copidosoma gelechiae with
notes on its biology J. Morphol. 37: 195-285.
LEIBY, R. W. 1924. The polyembryonic development of Platygaster vernalis. J. Agric.
Res. 28: 829-39.
LEIBY, R. W. 1926. The origin of mixed broods in polyembryonic Hymenoptera. Ann.
Entomol. Soc. America 19: 290-99.
LEIBY, R. W. AND C. C. HILL. 1923. The twinning and monembryonic development
of Platygaster hiemalis, a parasite of the Hessian fly. J. Agric. Res. 25: 337-350.
NUNNEY, L. 1985. Female-biased sex ratios: individual or group selection? Evolution:
PARKER, G. A. AND S. P. COURTNEY. 1984. Models of clutch size in insect oviposition.
Theor. Pop. Biol. 26: 27-48.
PARKER, H. A. AND W. R. THOMPSON. 1928. Contribution a la biologie des chalci-
diens entomophages. Ann. Ent. Soc. France 97: 425-65.
PATTERSON, J. T. 1919. Polyembryony and sex. J. Hered. 10: 344-352.
PATTERSON, J. T. 1921. The development of Paracopidosomopsis. J. Morphol. 36:
RETNAKARAN, A. AND J. PERCY. 1985. Fertilization and special modes of reproduc-
tion. Pp. 231-93. in L. I. Gilbert and G. A. Kerkut (eds.), Comprehensive Insect
Physiology, Biochemistry and Pharmacology, Vol. 1 Embryogenesis and Repro-
duction. Pergamon Press, Oxford.
SILVESTRI, F. 1906. Contribuzioni alla conoscenza biological degli Imenotteri parassiti.
Biologia del Litomastix truncatellus (Dalm.) 20 nota preliminary). Ann. Regia
Sc. Agric. Portici. 6: 3-51.
42 Florida Entomologist 72(1) March, 1989
SKINNER, S. W. 1985. Clutch size as an optimal foraging problem for insects. Behav.
Ecol. Sociobiol. 17: 231-38.
STONER, A. AND R. E. WEEKS. 1976. Copidosoma truncatellum, a polyembryonic
parasite of Trichoplusia ni: age of host eggs parasitized, searching, fecundity,
and effectiveness. Environ. Ent. 5: 323-28.
STRAND, M. R. 1986. The physiological interactions of parasitoids with their hosts
and their influence on reproductive strategies. Pp. 97-136. in J. Waage and D.
Greathed [eds.], Insect Parasitoids. Academic Press, London.
STRAND, M. R. 1988a. Variable sex ratio strategy by Telenomus heliothidis
(Hymeoptera: Scelionidae): the influence of host and conspecific density.
Oecologia 77: 219-24.
STRAND, M. R. 1988b. Patterns of progeny and sex allocation by parasitic Hymenopt-
era, pp. 293-312, in V. Gupta [ed.], Advances in Parasitic Hymenoptera Re-
search. E. J. Brill, Leiden.
STRAND, M. R. 1989a. Oviposition behavior and progeny allocation by the polyem-
bryonic wasp Copidosoma floridanum. J. Insect. Behav. (in press).
STRAND, M. R. 1989b. Development of the polyembryonic parasitoid Copidosoma
floridanum in Trichoplusia ni. Ent. Exp. Appl. (in press).
STRAND, M. R., J. A. JOHNSON, AND B. A. DOVER. 1989. Ecdysteroid and juvenile
hormone esterase profiles of Trichoplusia ni parasitized by the polyembryonic
wasp Copidosoma floridanum. Arch. Insect Biochem. PhySiol. (in press).
STRAND, M. R., AND H. C. J. GODFRAY. 1989. Superparasitism and ovicide in parasi-
tic Hymenoptera: theory and a case study of the ectoparasitoid Bracon hebetor.
Behav. Ecol. Sociobiol. (in press).
TAYLOR, P. D. 1981. Intra-sex and inter-sex sibling interactions as sex ratio determin-
ants. Nature 291: 64-6.
TAYLOR, P. D., AND M. G. BULMER. 1980. Local mate competition and the sex ratio.
J. Theoret. Biol. 86: 409-19.
WAAGE, J. K. 1986. Family Planning in parasitoids: adaptive patterns of progeny and
sex allocation. Pp. 63-95. in J. Waage and D. Greathead [eds.], Academic Press,
WAAGE, J. K., AND W. M. NG. 1984. The reproductive strategy of a parasitic wasp.
I, Optimal progeny and sex allocation in Trichogramma evanescens. J. Anim.
Ecol. 53: 401-16.
WAAGE, J. K., AND H. C. J. GODFRAY. 1985. Reproductive strategies and population
ecology of insect parasitoids. Pp. 449-70. in R. M. Sibley and R. H. Smith [eds.],
Behavioral Ecology, Ecological Consequences of Adaptive Behaviour. Blackwell
WERREN, J. H. 1980. Sex ratio adaptations to local mate competition in a parasitic
wasp. Science 208: 1157-59.
WERREN, J. H. 1987. Labile sex ratios in wasps and bees. Bioscience 37: 498-506.
Insect Behavioral Ecology-'88 Wojcik 43
BEHAVIORAL INTERACTIONS BETWEEN ANTS
AND THEIR PARASITES
Insects Affecting Man & Animals Research Laboratory,
P. 0. Box 14565,
Gainesville, FL 32604
Ectoparasitic and endoparasitic arthropods of ants include species of Acarina (Anten-
nophoridae, Uropodidae, Macrochelidae), Strepsiptera (Myrecolacidae), Hymenoptera
(Formicidae, Diapriidae, Eucharitidae), and Diptera (Phoridae). Variation in the style
of ectoparasitism is illustrated by the different lifestages involved and differing effects
on the hosts by parasitic ants and eucharitids. Considerable variation in behavior occurs
between related genera and species of phorids emphasizing the danger of over-generali-
zation on the relationships of ants and their parasites.
Artrop6dos ecto- y endoparasitos de homigas incluyen species de Acarina (Anten-
nophoridae, Uropodidae, Macrochelidae), Strepsiptera (Myrmecolacidae), Hymenoptera
(Formicidae, Diapriidae, Eucharitidae), y Diptera (Phoridae). Variaci6n en el estilo de
ectoparasitismo es ilustrado por las diferentes etapas de vida involucrados, asi como
por los diferentes efectos en los hospederos causados por hormigas parasiticas y
eucharitidos. Considerable variaci6n en comportamiento ocurre entire generos y species
relacionadas de phoridos, enfatizando el peligro de sobre generalizaci6n en las relaciones
de hormigas y sus parasitos.
Arthropods that live with, or are associated with ants, are termed myrmecophiles.
This relationship can be permanent or temporary, obligatory or facultative, integrated
The generally accepted classification of myrmecophiles is based on studies by Was-
mann [translated into English by Wheeler (1910) ]. The 5 behavioral categories Was-
mann devised are: 1) synechthrans (persecuted guests), 2) synoeketes (indifferently
tolerated guests), 3) symphiles (true guests), 4) ectoparasites and endoparasites, and
5) trophobionts (guests that provide secretions to the ants). While the categories are
mutually exclusive, the myrmecophiles often can be placed in different Wasmannian
categories at different stages of their life-cycles (Wilson 1971). Several other classifica-
tion systems have been proposed (Kistner 1979), but none has become as generally
accepted as the Wasmannian system. In most cases, the true relationships of the myr-
mecophiles are not known, indeed many are still being discovered and their relationships
with their hosts defined.
Wilson (1971) divides myrmecophilous parasites into (1) ectoparasites that live on
the body surfaces of their hosts, licking up their oil secretions, steal food from them,
or bite through the exoskeleton and feed on their blood, and (2) endoparasites
(parasitoids) that penetrate and develop in the host's body. He notes that the behavior
of the internal parasites is not usually distinguishable from the behavior of similar
species which parasitize nonsocial insects. This is not necessarily the case as I will show.
Florida Entomologist 72(1)
Some inquilines use chemical mimicry as a means of integrating themselves into the
host colony. These organisms, several termitophilous staphylinids (Howard et al. 1980,
1982) and a myrmecophilous scarab (Vander Meer & Wojcik 1982), acquire the cuticular
hydrocarbons of the host. Studies by Vander Meer, Wojcik, and Jouvenaz (unpublished)
have shown chemical mimicry to be wide-spread in Solenopsis spp. myrmecophiles. I
discuss some of these myrmecophiles in this paper. Other myrmecophiles use body
shape, color, special morphological adaptions, and appeasement substances to become
integrated into ant colonies (Wilson 1971, Kistner 1979).
Most mites found in ant nests are phoretic and are scavengers, parasites, or pre-
dators on organisms other than the host ants (Wilson 1971). Some [e.g., Antennophorus
uhlmanni Haller (Antennophoridae)] are social parasties, and solicit food from the
worker ants (Lasius mixtus Nylander) in the same manner as the ants and symphiles
(Janet 1897). Other mites are ectoparasites. For instance, Cillibano comato (Berlese)
(Uropodidae) attaches to the gaster of the Lasius spp. host ants, feeds on the
hemolymph, and leaves feeding scars (Janet 1897). The mites associated with the New
World army ants are the best known, but most species are not parasites of the ant host
(Rettenmeyer 1961). Species of ectoparasitic mites usually are host-specific and may
even be specific to a part of the ant. Macrocheles rettenmeyeri Krantz (Macrochelidae),
for example, attaches to and feeds on the hind arolium of the host, Eciton dulcius
crassinode Borgmeier; and the ants seem unaffected and use the mite bodies as part of
their legs as they walk (Rettenmeyer 1962).
Strepsipterans of the family Mrymecolacidae are endoparasites of ants, and perhaps
Orthoptera (Riek 1970, Teson & de Remes Lenicov 1979). Genera of ants parasitized
include Eciton, Pseudomyrmex, Pheidole, Solenopsis, and Camponotus, which repre-
sent the majority of the major subfamilies of ants. Life cycles may be similar to those
of the non-myrmecophilous species (Clausen 1940a). Stylopized ants behave unusually,
lingering on the tips of grass stems even in bright sunlight (Riek 1970). Lingering on
the tips of grass stems increases the chance of being found by the short-lived male
parasites or puts them in a good position for male emergence. This type of behavioral
modification has also been reported in ants parasitized with mermithid nematodes (Riek
1970), trematode worms (Paraschivescu 1982), or fungi (Loos-Frank & Zimmerman
Parasitic wasps (Ichneumonoidea, Chalcidoidea, Bethyloidea) have been observed
ovipositing in the gasters of ants or have been found inside ant nests (Wheeler 1910,
Weber 1972, Kistner 1982), but little is known about their relationships with their hosts.
Some parasitic wasps are myrmecophiles even though they are not parasites of the host
ants. Paralypsis enervis Nees is known from the nests of Lasius niger (L.) in France
(Kistner 1982) and P. eikoae (Yasumatsu) and Aclitus sappaphis Takada & Shiga are
known from the nests of L. niger and Pheidolefervida F. Smith, respectively, in Japan
(Takada & Hashimoto 1985). The wasps live unmolested in the nests and solicit food
from the ants. The mouthparts of the wasps are modified to enable them to engage in
trophallaxis with the ants (Kistner 1982). These aphidiid wasps are parasites of the
aphids tended by the ants.
Insect Behavioral Ecology-'88 Wojcik 45
Social parasitism of ants by other ants runs the gamut from temporary residence by
a newly-mated queen, to slave-making, to obligatory parasitism (Buschinger 1986). For
example, the temporarily parasitic Lasius reginae Faber queen attacks (throttles) the
host L. alienus (Foerster) queen until she starves to death, while the Bothriomyrmex
decapitans queen cuts off the head of the Tapinoma spp. host queen. These temporary
parasites produce a worker caste and eventually usurp the host nest. The true parasitic
ants do not produce a worker caste and are totally dependent on the host workers
(Wilson 1971). An example of this group is Solenopsis (Labauchena) daguerrei Santschi,
whose queens attach to the neck of the host S. richteri Forel queen, and steal the food
the workers offer their mother queen (Silveira-Guido et al. 1968, 1973). If more than
one parasite queen is present, they distribute themselves over the host queen's body.
There is some evidence that the parasitic queen is fed preferentially to the host queen.
Eventually the host queen is deprived of enough food so that her egg production de-
clines, particularly when more than one parasite queen is present. The host queen
eventually dies and the host colony's vigor and size decreases. S. richteri colonies have
been found with over 3,300 parasite adults (males and females) and immatures (70% of
the collected colony) (Silveira-Guido et al. 1965, 1973). The method of invasion and
integration into the host colony is not understood, but good evidence exists that the
parasite queen has a host cuticular-hydrocarbon pattern which masks her own cuticular-
hydrocarbons (Vander Meer & Wojcik unpublished).
Most of the small wasps of the family Diapriidae are parasites of Diptera or Coleopt-
era. Species in several genera have been found in ant nests or running with army ants
on foraging trails (Huggert & Masner 1983, Masner 1959, 1976, 1977, Loiacono 1981).
The best known myrmecophilous species, Solenopsia imitatrix Wasmann, which was
first collected and observed in 1884 by Wasmann (reviewed by Wing 1951), has not
definitely been shown to be an ant parasite. The male and female wasps live in the ant
nests and solicit food from the ants. Although some species are thought to caste off
their wings before entering the ant nest (Masner 1976), other species have their wings
chewed off by the ants after they invade the ant nest (Masner 1959). Lachaud and
Passera (1982) demonstrated that Plagiopria passerai Masner parasitizes queen cocoons
of Plagiolepis pygmaea (Latreille).
All eucharitid wasps are parasitic on ants (Heraty 1985, 1988). Some species are
fastidious in their choice of hosts and oviposition sites, others are not. The females lay
their eggs in plant tissue away from the hosts. The specific plant and plant part in which
eggs are laid vary among species (Clausen 1940b, Parker 1937, 1942, Johnson et al.
1986). The genus Orasema and related genera (subfamily Oraseminae) have a highly
sclerotized scimitar-shaped ovipositor (Heraty 1985) and insert their eggs singly into
incisions in the leaf surfaces (Clausen 1940b); the remaining genera (subfamily
Eucharitinae) generally lack the highly sclerotized ovipositor (Heraty 1985) and usually
lay their eggs in cavities or spaces in floral or leaf buds or in cavities formed by leaf
scars (Clausen 1940b). The eggs of Schizaspidia tenuicornis Ashmead over-winter in
the buds (Clausen 1923).
After hatching, the planidia (first instar larvae) wait for an appropriate host. In
several species, an association of planidia with thrips has been reported (Clausen 1940a,
Florida Entomologist 72(1)
Wilson & Cooley 1972, Johnson et al. 1986). The planidia attach to the thrips and appear
to feed on them (Johnson et al. 1986). This association, and others like it, could enhance
the longevity of the planidia. After locating an appropriate host ant, the planidium,
which is legless (Heraty & Darling 1984), climbs on it, and rides phoretically back to
the ant nest attached by means of an anal sucker. In the ant nest, the planidium leaves
the adult ant and locates the ant brood (larvae and pupae). Upon locating an ant larva,
the planidium burrows beneath its cuticle (Wheeler & Wheeler 1937, Heraty & Wojcik
unpublished). The parasite does not seem to develop until the ant larva reaches the late
prepupal stage (Wheeler 1907, Clausen 1941, Kirkpatrick 1957). Most of the time, the
parasite feeds at least partially as an endoparasite (Wheeler & Wheeler 1937), resulting
in a characteristically deformed pupa called a phthisergate (Wheeler 1907, Van Pelt
1950). Phthisergates are found only in the ant subfamilies which have naked pupae
(Myrmicinae, Dolichoderinae, Pseudomyrmicinae, some Formicinae). In ants which pu-
pate in a cocoon (Ponerinae, some Formicinae), the host is attacked after the pupal case
is formed, and the deformed host is consumed so that only the exoskeleton remains
(Wheeler & Wheeler 1924, Parker 1932, Ayre 1962).
My studies of Orasema spp., parasitic on Solenopsis sp. in Mato Grosso, Brazil,
indicate that each individual parasite requires more than one host pupa. The size of the
mature parasite larvae and the variable size of the phthisergates indicate that each
Orasema larvae may feed on more than one victim. The parasites that feed on hosts
which pupate in cocoons can feed only on one host (Wheeler & Wheeler 1924, Parker
1932, Kirkpatrick 1957, Ayre 1962). Phthisergates that survive being fed upon by the
wasp larvae are not able to complete development and eventually die (Wheeler 1910,
Van Pelt 1950).
After the eucharitid larva terminates its feeding, it leaves the host (in those cases
where the host is not in a cocoon) and pupates, in the case of Orasema spp. with the
assistance of the worker ants (just as ant larvae are assisted in molting by worker ants)
(Silveira-Guido et al. 1964, Wojcik unpublished). After pupation, the parasites are mixed
in with the host pupae, and are cared for in the same manner as ant pupae (Silveira-
Guido et al. 1964, Williams 1980, Wojcik unpublished). Eclosion of the adult parasites
is assisted by worker ants in the same manner as they assist their own pupae (Silveira-
Guido et al. 1964, Wojcik unpublished). The adult parasites are integrated completely
into the ant society, being fed and groomed by the worker ants (Wheeler 1907, Williams
1980, Wojcik unpublished). When a nest is disturbed, the pupal and adult parasites are
rescued by the ants in preference to their own brood (Wheeler 1907, Mann 1914, Wojcik
unpublished). Generally the ants either ignore the parasites when they are leaving the
nest (Clausen 1923, 1941, Kirkpatrick 1957) or they actively remove them from the nest
(Ayre 1962). The adult parasites are sexually mature when they leave the nest as the
females all contain fully developed eggs (Wheeler 1907, Kirkpatrick 1957, Ayre 1962).
Mating takes place immediately upon exiting the ant nest, as the males hover over the
ant nest or rest on surrounding vegetation (Clausen 1923, 1941, Ayre 1962, Wojcik &
Jouvenaz unpublished). Studies have shown that Orasema sp. larvae, pupae and adults
possess only Solenopsis spp. host cuticular hydrocarbons while in ant nests; but that
Orasema sp. adults acquire species-specific cuticular hydrocarbons upon leaving the
nest (Jouvenaz et al. 1988, Vander Meer et al. in press).
Phorid flies are among the most abundant myrmecophiles known (Rettenmeyer &
Akre 1968). Some are scavengers, some predators, some symphiles, some parasites,
but most are of unknown status (Kistner 1982). The references to attempted or sus-
pected oviposition by phorid flies on ants are almost as numerous as the flies themselves.
Insect Behavioral Ecology-'88 Wojcik 47
In light of the work of Rettenmeyer and Akre (1968) which showed that many of the
species of phorids associated with army ants are not army ant parasites, the numerous
references to the behavior of these phorids must be reevaluated and will not be dis-
The site of oviposition by the parasites varies with the species of fly: ant head
(Eidmann 1937, Greene 1938, Disney 1980, 1982, 1986, Williams & Banks 1987), thorax
(Feener 1987), or gaster (Wasmann 1918, Donisthorpe 1927, Disney 1982, 1986). In only
a few instances have eggs (Williams & Banks 1987), larvae (Pergande 1901, Foresti &
Pereira da Silva 1970, Williams & Whitcomb 1974), or pupae (Wojcik et al. 1987) been
observed on or in the host. Generally, when the flies attack the ants, they do so after
hovering over foraging trails (Steyskal 1944, Eibl-Eibesfeldt & Eibl-Eibesfeldt 1967,
Disney 1981, 1986, Williams & Banks 1987), over nests during mating flights (Smith
1928), or over disturbed nests (Brues 1902, LaBerge 1953, Williams et al. 1973). A fly
darts down and quickly attempts to deposit an egg on the ant (Williams & Banks 1987)
or insert its ovipositor in the appropriate place (Feener 1987). In some cases the females
attempt to oviposit by landing on the ground, running up to the ants, and jumping on
them (Disney 1986). The ants sometimes appear stunned after oviposition (Smith 1928,
Williams et al. 1973). The ants defend themselves vigorously against the flies (Donis-
thorpe 1927, Smith 1928, Greene 1938, Weber 1972, Williams et al. 1973, Burges 1979,
Feener 1981, 1987, Williams & Banks 1987, Wojcik unpublished), often chasing the flies
away before they can oviposit (Williams et al. 1973, Burges 1979).
In attacking monomorphic ants, flies choose any worker present (Disney 1982, 1986),
but when attacking dimorphic or polymorphic ants, they attack majors preferentially
(Feener 1981, Disney 1982, Williams & Banks 1987). Atta cephalotes L. minima workers
ride on pieces of cut leaves being transported by major workers (Eibl-Eibesfeldt 1967,
Eibl-Eibesfeldt & Eibl-Eibesfeldt 1967); with their sole function being to prevent ovipos-
ition by Apocephalus sp. flies on the major workers. Phorids probably oviposit selec-
tively on major workers because this caste represents a more reliable food source,
because it is longer-lived (Feener 1981, 1987, Williams & Banks 1987). Alate ants are
much less common in colonies and are rarely exposed to parasitism. Only one instance
of phorid parasitism of alate ants is known (Wojcik et al. 1987).
Parasitism, or even the threat of parasitism, by phorids places the Pheidole spp.
(which are dimorphic) at a competitive disadvantage with competing fire ants (Feener
1981). A similar disadvantage may exist for all ants having majors which defend a
foraging arena. Attacks by phorid flies may cause Iridomyrmex cordatus (Smith) to use
covered foraging tunnels and Pheidologeton sp. to shift its foraging area daily (Disney
A different behavioral scenario has been reported for Apodicrania termilophila
(Borgmeier), a fly found in Solenopsis spp. nests in Brazil (Williams & Whitcomb 1974,
Williams 1980). The adult flies walk around on the disturbed nest and are ignored by
the ants. Immature fly larvae are endoparasitic in ant larvae while mature fly larvae
and pupae are found in the nest being tended by the ants (Williams 1980). The only
other report of a phorid attacking an ant immature is that of Apocephalus aridus
Malloch attacking a Pheidole dentata Mayr pupa (LaBerge 1953). At least some species
mate at the same sites where females oviposit (Disney 1986).
The true relationship of most myrmecophiles with their ant hosts is unknown. Much
of the older literature on myrmecophilous parasites needs reevaluation in light of recent
knowledge. There appears to be a great deal more variation in biology and behavior
between related genera and species than has been assumed in the past, and many
Florida Entomologist 72(1)
earlier generalizations are not valid. Detailed studies on the biology and behavior of
myrmecophiles will result in a better understanding of the relationships between ants
and their inquilines.
I would like to thank Drs. J. H. Frank and E. D. McCoy for providing the opportu-
nity to prepare and present this paper and for their useful and concise editorial com-
ments. Drs. C. S. Lofgren, R. S. Patterson, and A. H. Undeen provided valuable
AYRE, G. L. 1962. Pseudometagea schwarzii (Asm.) (Eucharitidae: Hymenoptera), a
parasite of Lasius neoniger Emery (Formicidae: Hymenoptera). Canadian J.
Zool. 40: 157-64.
BRUES, C. T. 1902. New and little-known guests of the texan legionary ants. Amer-
ican Nat. 36: 365-78.
BURGES, R. J. 1979. A rare fly and its parasitic behavior toward an ant (Diptera:
Phoridae, Hymenoptera: Formicidae). Florida Ent. 62: 413-4.
BUSCHINGER, A. 1986. Evolution of social parasitism in ants. Trends Ecol. Evol. 1:
CLAUSEN, C. P. 1923. The biology of Schizaspidia tenuicornis Ashm., a eucharid
parasite of Camponotus. Ann. Ent. Soc. America 16: 195-217, pl. 14-15.
1940a. Entomophagous insects. McGraw-Hill, New York, 688 p.
S1940b. The oviposition habits of the Eucharidae (Hymenoptera). J.
Washington Acad. Sci. 30: 504-16.
.1941. The habits of the Eucharidae. Psyche 48: 57-69.
DISNEY, R. H. L. 1980. A new species of Apocephalus (Diptera: Phoridae) that
parasitises Atta texana (Hymenoptera: Formicidae) in Texas. Z. Angew. Zool.
.1981. Apocephalus laceyi n.sp. (Diptera: Phoridae) attacking Cam-
ponotusfemoratum (F.) (Hymenoptera: Formicidae) in Brazil. Ent. Scandinavica
1982. Three new species of scuttle-fly (Diptera: Phoridae) that parasitize
ants (Hymenoptera: Formicidae) in North America. J. Zool., London 197: 473-
S1986. A new genus and three new species of Phoridae (Diptera) parasitiz-
ing ants (Hymenoptera) in Sulawesi. J. Nat. Hist. 20: 777-87.
DONISTHORPE, H. 1927. The guests of British ants. Routledge, London, 244 p.
EIBL-EIBESFELDT, J. 1967. On the guarding of leafcutter ants by minima-workers.
Naturwissenschaften 54: 346.
EIBL-EIBESFELDT, I., and E. EIBL-EIBESFELDT. 1967. Das Parasitenabwehren der
minima-Arbeiterinnen der blattschneider-Ameise (Atta cephalotes). Z.
Tierpsychol. 24: 278-81.
EIDMANN, H. 1937. Die Gaste und Gastverhaltnisse der Blattschneiderameise, Atta
sexdens L. Z. Morphol. Oekol. Tiere 32: 391-462.
FEENER, D. H., JR. 1981. Competition between ant species: outcome controlled by
parasitic flies. Science 214: 815-7.
.1987. Size-selective oviposition in Pseudacteon crawfordi (Diptera:
Phoridae), a parasite of fire ants. Ann. Ent. Soc. America 80: 148-51.
FORESTI, L. AND V. PEREIRA DA SILVA. 1970. Nota s6bre a ocorr&ncia de um
forideo parasite de formigas sauivas (Diptera: Phoridae). Proc. 22nd Meet. Soc.
Bras. Progres. Cien., p. 287.
Insect Behavioral Ecology-'88 Wojcik
GREENE, C. T. 1938. A new genus and two new species of the dipterous family
Phoridae. Proc. U.S. Natl. Mus. 85: 181-5.
HERATY, J. M. 1985. A revision of the nearctic Eucharitinae (Hymenoptera: Chal-
cidoidea: Eucharitidae). Proc. Ent. Soc. Ontario 16: 61-103.
. 1988. Of eucharitids and ants (Hymenoptera): biological conservatism and
morphological explosion. Proc. 18th Int. Congr. Ent., Abstr.: 12.
AND D. C. DARLING. 1984. Comparative morphology of the planidial
larvae of the Eucharitidae and Perilampidae (Hymenoptera: Chalcidoidea). Syst.
Ent. 9: 309-28.
HOWARD, R. W., C. A. MCDANIEL, AND G. J. BLOMQUIST. 1980. Chemical mimicry
as an integrating mechanism: cuticular hydrocarbons of a termitophile and its
host. Science 210: 431-3.
-,, AND 1982. Chemical mimicry as an integrating
mechanism for three termitophiles associated with Reticulitermes virginicus
(Banks). Psyche 89: 157-67.
HUGGERT, L., AND L. MASNER. 1983. A review of myrmecophilic-symphilic diapriid
wasps in the Holarctic realm, with descriptions of new taxa and a key to genera
(Hymenoptera: Proctotrupoidea: Diapriidae). Contrib. American Ent. Inst. 20:
JANET, C. 1897. Etudes sur les fourmis, les gu&pes et les abeilles. Note 13: sur le
Lasius mixtus 'Antennophorus ulhmanni etc. H. Ducourtieux, Limoges, 62 p.
JOHNSON, J. B., T. D. MILLER, J. M. HERATY, AND F. W. MERICKEL. 1986. Obser-
vations on the biology of two species of Orasema (Hymenoptera: Eucharitidae).
Proc. Ent. Soc. Washington 88: 542-9.
JOUVENAZ, D. P., R. K. VANDER MEER, AND D. P. WOJCIK. 1988. Chemical
mimicry in a parasitoid, Orasema sp., (Hymenoptera: Eucharitidae) of fire ants.
Proc. 18th Int. Congr. Ent., Abstr.: 243.
KIRKPATRICK, T. W. 1957. Insect life in the tropics. Longmans, Green, London, 311
KISTNER, D. H. 1979. Social and evolutionary significance of social insect symbionts,
Pp. 339-413 in H. R. Hermann [ed.], Social Insects. Academic Press, New York,
Vol. 1, 437 p.
1982. The social insects' bestiary, Pp. 1-244, in H. R. Hermann [ed.],
Social Insects. Academic Press, New York, Vol. 3, 459 p.
LABERGE, W. E. 1953. A note on a phorid parasite of Pheidole dentata Mayr. J.
Kansas Ent. Soc. 26: 69.
LACHAUD, J. P., AND L. PASSERA. 1982. Donnies sur la biologie de trois Diapriidae
myrm6cophiles: Plagiopria passerai Masner, Solenopsia imitatrix Wasmann et
Lepidopria pedestris Kieffer. Insect. Soc. 29: 561-7.
LOIACONO, M. S. 1981. Diapridos ectofilos sinfilos de la region Neotropical
(Hymenoptera Proctotrupoidea Diapriidae). Rev. Soc. Ent. Argentina 40:
LOOS-FRANK, B., AND G. ZIMMERMAN. 1976. Uber eine dem Dicrocoelium-befall
analoge Verhaltensanderung bei Ameisen der Gattung Formica durch einen Pilz
der Gattung Entomophthora. Z. parasitenkd. 49: 281-9.
MANN, W. M. 1914. Some myrmecophilous insects from Mexico. Psyche 21: 171-84.
MASNER, L. 1959. A revision of ecitophilous diapriid-genus Mimopria Holmgren
(Hym., Proctotrupoidea). Insect. Soc. 6: 361-7.
.1976. Notes on the ecitophilous diapriid genus Mimopria Holmgren
(Hymenoptera: Proctotrupoidea, Diapriidae). Canadian Ent. 108: 123-6.
1977. A new genus of ecitophilous diapriid wasps from Arizona
(Hymenoptera: Proctotrupoidea, Diapriidae). Canadian Ent. 109: 33-6.
50 Florida Entomologist 72(1) March, 1989
PARASCHIVESCU, D. 1982. Freiland- und Laboruntersuchungen zum En-
twicklungskreislauf von Dicrocoelium dendriticum (kleiner Leberegel) in einem
Wald/Feld-Biotop bei Wiirzburg (Unterfranken) unter besonderer Be-
riicksichtigung des 2. Zwischenwirtes Formica pratensis. Waldhygiene 14: 141-
PARKER, H. L. 1932. Notes on a collecting spot in France and a chalcid larva (Stibula
cynipiformis Rossi) (Hymenop.: Eucharidae). Ent. News 43: 1-6.
S1937. On the oviposition habits of Stibula cynipiformis Rossi (Hymen.,
Eucharidae). Proc. Ent. Soc. Washington 39: 1-3.
.1942. Oviposition habits and early stages of Orasema sp. Proc. Ent. Soc.
Washington 44: 142-5.
PERGANDE, T. 1901. The ant-decapitating fly. Proc. Ent. Soc. Washington 4: 497-
RETTENMEYER, C. W. 1961. Behavior, abundance and host specificity of mites found
on Neotropical army ants. (Acarina; Formicidae; Dorylinae). Proc. XI Int. Congr.
Ent. 1: 610-3.
.1962. Notes on the host specificity and behavior of myrmecophilous mac-
rochelid mites. J. Kansas Ent. Soc. 35: 358-60.
-, and R. D. AKRE. 1968. Ectosymbosis between phorid flies and army ants.
Ann. Ent. Soc. America 61: 1317-26.
RIEK, E. F. 1970. Strepsiptera, Pp. 622-35, in The Insects of Australia. Melbourne
Univ. Press, 1,029 p.
SILVEIRA-GUIDO, A., J. CARBONELL, AND C. CRISCI. 1973. Animals associated
with the Solenopsis (fire ants) complex, with special reference to Labauchena
daguerrei. Proc. Tall Timbers Conf. Ecol. Anim. Control Habitat Manage. 4:
J. CARBONELL-BRUHN, C. CRISCI, AND P. SAN-MARTIN. 1968.
Labauchena daguerrei Santschi como parasito social de la hormiga Solenopsis
saevissima richteri Forel. Agron. Trop. (Maracay, Venezuela) 18: 207-9.
C. CRISCI-PISANO, P. SAN-MARTIN, AND J. CARBONNELL-BRUHN.
1965. Investigations on the biology and biological control of the fire ant, Sol-
enopsis saevissima richteri Forel in Uruguay. Final report. Depto. Sanid. Veg.,
Fac. Agron., Univ. Repub., Montevideo, Uruguay, 70 p., 36 pl.
P. SAN-MARTIN, C. CRISCI-PISANO, AND J. CARBONNELL-BRUHN.
1964. Investigations on the biology and biological control of the fire ant, Sol-
enopsis saevissima richteri Forel in Uruguay. Third report. Depto. Sanid. Veg.,
Fac. Agron., Univ. Repub., Montevideo, Uruguay, 67 p.
SMITH, M. R. 1928. Plastophora crawfordi Coq. and Plastophora spatulata Malloch
(Diptera: Phoridae), parasitic on Solenopsis geminata Fabr. Proc. Ent. Soc.
Washington 30: 105-8.
STEYSKAL, G. 1944. A new ant-attacking fly of the genus Pseudacteon, with a key
to the females of the North American species (Diptera, Phoridae). Occas. Pap.
Mus. Zool., Univ. Michigan 489: 1-4.
TAKADA, H., AND Y. HASHIMOTO. 1985. Association of the root aphid parasitoids
Aclitus sappaphis and Paralipsis eikoae (Hymenoptera, Aphidiidae) with the
aphid-attending ants Pheidole fervida and Lasius niger (Hymenoptera, For-
micidae). Kontyfi 53: 150-60.
TESON, A., AND A. M. M. DE REMES LENICOV. 1979. Estrepsipteros parasitoides
de hymenopteros (Insecta Strepsiptera). Rev. Soc. Ent. Argentina 38: 115-22.
VANDER MEER, R. K., D. P. JOUVENAZ, AND D. P. WOJCIK. Chemical mimicry in a
parasitoid (Hymenoptera: Eucharitidae) of fire ants (Hymenoptera: Formicidae).
J. Chem. Ecol. (in press.)
Insect Behavioral Ecology-'88 Wojcik 51
AND D. P. WOJCIK. 1982. Chemical mimicry in the myrmecophilous beetle
Myrmecaphodius excavaticollis. Science 218: 806-808.
VAN PELT, A. F. 1950. Orasema in nests of Pheidole dentata Mayr (Hymenoptera:
Formicidae). Ent. News 41: 161-3.
WASMANN, E. 1918. Zur Lebensweise und Fortpflanzung von Pseudacteon for-
micarium Verr. Biol. Zentbl. 38: 317-28.
WEBER, N. A. 1972. Gardening ants, the attines. Mem. American Philos. Soc. 92:
WHEELER, G. C., AND E. H. WHEELER. 1924. A new species of Schizaspidia
(Eucharidae), with notes on a eulophid ant parasite. Psyche 31: 49-56.
AND 1937. New hymenopterous parasites of ants (Chal-
cidoidea: Eucharidae). Ann. Ent. Soc. America 30: 163-73, pl. 1-2.
WHEELER, W. M. 1907. The polymorphism of ants, with an account of some singular
abnormalities due to parasitism. Bull. American Mus. Nat. Hist. 23: 1-93, pl. 1-6.
.1910. Ants: Their Structure Development and Behavior. Columbia Univ.
Press, New York, 663 p.
WILLIAMS, D. F., AND W. A. BANKS. 1987. Pseudacteon obtusus (Diptera: Phoridae)
attacking Solenopsis invicta (Hymenoptera: Formicidae) in Brazil. Psyche 94:
WILLIAMS, R. N. 1980. Insect natural enemies of fire ants in South America with
several new records. Proc. Tall Timbers Conf. Ecol. Anim. Control Habitat Man-
age. 7: 123-34.
-, J. R. PANAIA, D. GALLO, AND W. H. WHITCOMB. 1973. Fire ants at-
tacked by phorid flies. Florida Ent. 56: 259-62.
-, AND W. H. WHITCOMB. 1974. Parasites of fire ants in South America.
Proc. Tall Timbers Conf. Ecol. Anim. Control Habitat Manage. 5: 49-59.
WILSON, E. 0. 1971. The Insect Societies. Belknap, Cambridge, MA, 548 p.
WILSON, T. H., AND T. A. COOLEY. 1972. A chalcidoid planidium and an entomophilic
nematode associated with the western flower thrips. Ann. Ent. Soc. America 65:
WING, M. W. 1951. A new genus and species of myrmecophilous Diapriidae with
taxonomic and biological notes on related forms. Trans. Roy. Ent. Soc. London
WOJCIK, D. P., D. P. JOUVENAZ, AND C. S. LOFGREN. 1987. First report of a
parasitic fly (Diptera: Phoridae) from a red imported fire ant (Solenopsis invicta)
alate female (Hymenoptera: Formicidae). Florida Ent. 70: 181-2.
52 Florida Entomologist 72(1) March, 1989
CHAOS AND NATURAL SELECTION
J. C. ALLEN
Department of Entomology and Nematology
University of Florida
Gainesville, Florida 32611-0143 U.S.A.
The nonlinear dynamics view of population interactions emphasizes three critical
points: 1) instability does not imply extinction, 2) very complicated behavior is possible
from very simple systems, and 3) if density-dependence occurs as lagging nonlinear
feedback, then it is the primary cause of instability and chaos and does not stabilize
populations in contrast to the traditional view.
Discrete time ("Nicholsonian") host-parasitoid models are used to illustrate that a
patch-efficient parasitoid is destabilizing when searching a patchy host distribution.
When some very crude genetics are added in terms of parameter phenotypes (patchy
and random hosts and patchy and random parasitoid search strategies), then there is a
much wider variety of dynamic behavior. In general, selection does not seem to work
for or against chaotic parameter phenotypes. Instead, it appears to increase the likeli-
hood of chaos (i.e., at lower parameter values) and decrease the likelihood of extinction
of chaotic systems if they occur.
El punto de vista no lineal y dinamico de interacciones de poblaciones hace incapi6
en tres punto critics: 1) inestabilidad no implica extinci6n; 2) un comportamiento muy
complicado es possible en sistemas muy simples; y 3) si la densodependencia ocurre como
un retardo no lineal retroactive, entonces esta es la primera causa de inestabilidad y
caos, lo que no estabiliza las poblaciones contrariamente al punto de vista traditional.
Los models hospedero-parasitoide de tiempo discrete ("Nicholsoniano"), son
utilizados para ilustrar que un parasitoide parche eficiente estg es destabilizador cuando
6ste busca la distribuci6n en parches del hospedero. Una amplia variedad de compor-
tamiento dinAmico en t6rminos de parAmetros fenotipicos (hospederos al azar o en
parches y estrategias de bfsqueda al azar o en parches), ocurre cuando algunos aspects
de la gendtica basica son adicionados. En general, la selecci6n pare que no trabaja para
o en contra de los parametros de fenotipos ca6ticos. Por el contrario, parece incrementar
la probabilidad de caos (esto es, a bajos valores del parametro) y decrece la probabilidad
de extenci6n de sistemas ca6ticos si ellos occurieran.
The subtle and clever mechanisms by which insect parasitoids locate their hosts are
a continuing source of wonder to biologists (Cade 1975, Walker 1986, Lewis & Tumlinson
1988). Modeling the population dynamic consequences of efficient non-random search
behavior by parasitoids ("aggregation to host density") is the subject of a large and
active literature (Hassell & May 1973, 1974, Beddington et al. 1978, May 1978, May &
Hassell 1981, Heads & Lawton 1983, Waage 1983, Hogarth & Diamond 1984, Chesson
& Murdoch 1986, Walde & Murdoch 1988). Much of this work was probably stimulated
by Nicholson's early comment that his elegant but unstable host-parasitoid model might
be stabilized by "the breaking up of the species-population into numerous small widely
separated groups which wax and wane and then disappear, to be replaced by new
groups in previously unoccupied situations" (Nicholson & Bailey 1935).
Insect Behavioral Ecology-'88 Allen 53
Many complications and misunderstandings have beset the work on host-parasitoid
"stability" from the beginning. First of all stability came to be used in a narrow sense
to mean attraction to a fixed point equilibrium. When contrasted with Nicholson's oscil-
lations of increasing amplitude, one gets the notion that the only biological alternative
to an attracting fixed point is extinction (at least locally). Obviously, a more detailed
and sophisticated view of dynamic behavior is required for an understanding of ecolog-
ical systems. A more refined view of dynamic motion has recently arisen from the field
of nonlinear dynamics with particular reference to ecology (Schaffer & Kot 1985, 1987,
Allen, in press). Even the engineers have been somewhat narrow in their approach to
dynamics, and "tutorials" on the dynamic possibilities have been written from the en-
gineering perspective (Parker & Chua 1987), so it is scarcely surprising that ecologists
have had a problem. Briefly stated, some of the dynamic possibilities are:
Attracting point: a fixed point equilibrium to which the system is attracted.
Attracting or limit cycle: an unvarying cycle with constant period and amplitude into
which the system is attracted.
Toroidal flow: motion on the surface of a torus (i.e., a doughnut) resulting from the
interaction of two (or more) cycles (one around the axis the other around the center of
the torus). Two possibilities have been recognized in terms of the periods of the cycles:
1) The ratio of the periods is rational (a whole number ratio). In this case the motion
will be truly periodic. 2) The ratio of the periods is irrational (not a whole number ratio).
In this case the motion is said to be quasiperiodic or almost periodic. A good example
is the "wagon wheel effect" in old western movies when the wheels appear to be rotating
slowly forward or backward due to the interaction of the film cycle and the wheel cycle.
A much more rigorous technical definition is given by Parker & Chua (1987).
Chaos: This can be loosely defined as "none of the above." It is bounded attracting
dynamic behavior that is not an attracting point, not periodic and not quasiperiodic. It
is sensitive to initial conditions in that initially close points do not stay close but separate
at an exponential rate while staying in the bounded region (the "attractor").
This short discourse on non-linear behavior is included here to emphasize its impor-
tance when considering ecological systems. Even in the most sophisticated mathematical
models of host-parasitoid interactions (e.g., Chesson & Murdoch 1986, Murdoch et al.
1987), the principle focus is still on stable vs. unstable behavior with the implication
that instability and extinction go hand-in-hand. There is nothing in the nonlinear be-
haviors defined above which dictates extinction of species having "unstable" dynamics,
and even chaos does not necessarily imply extinction. It could be argued that what we
are talking about here is "mathematical" stability vs. "biological" stability, but since the
mathematics is an attempt to describe the biology, the two should be compatible. Insta-
bility and even local extinction could be tolerated easily by species with high rates of
movement, and this may be the rule rather than the exception in host-parasitoid systems
with high rates of attack and reproduction. Thus, nonlinear dynamics, instability and,
even chaos at the local population level do not imply extinction of the species (i.e.,
biological disaster). In fact, natural selection acting on individuals to produce high repro-
ductive and attack rates may actually promote unstable or even chaotic dynamics at the
local level. We will investigate this in some models below.
One final word about density-dependence as a "stabilizing" mechanism in ecology: it
is difficult to envision density-dependence (i.e., nonlinear density feedback) as an instan-
taneous process. That is, most such processes (e.g., reproductive feedback, predator
numerical responses, disease transmission, etc.) logically involve some sort of associated
lag in the process before its feedback effects are felt by the system. But lagging non-
linear feedback along with periodic forcing are the primary causes of instability and
chaos in nonlinear systems (Allen, in press). Thus, expecting to find direct density-de-
pendence in field data on real systems seems a bit simplistic considering the complexity
Florida Entomologist 72(1)
of the situation (see, for example, Schaffer & Kot 1986). In fact, density-dependence
may be present, it may go undetected because of the lags involved, and it may be
destabilizing the system anyway and even producing chaos at the local level if it is
strong enough. It is, therefore, no surprise that one can show such seemingly odd things
as inverse density-dependence stabilizing a population (Hassell 1984, Chesson & Mur-
doch 1986) and that one can easily illustrate systems in which density-dependence is
present but nearly impossible to detect (Hassell 1985). Thus, lagging density-depen-
dence could be all around us but simply be very difficult to detect by conventional
approaches to field data, and it would probably tend to destabilize populations rather
than stabilize them anyway. When the full range of nonlinear behaviors is considered,
the traditional search for stabilizing density dependence seems much too narrow in its
approach. Better perhaps would be to search for any sort of attracting behavior in some
space-time window which we find convenient or which seems biologically meaningful.
HOST-PARASITOID MODELS INCORPORATING SPATIAL BEHAVIOR
BY THE PARASITOID
In this section some extensions of Nicholson-Bailey discrete generation host-
parasitoid models will be considered which attempt to include parasitoid behavior and
genetics. Full analysis will be avoided and the emphasis will be on methods and ideas
about behavior and/or genetics in these models. The basic general form will be
Ht1 = FHf() 1)
Pt+ = H [1 f(*)]
in which f(*) = f(Ht,Pt) is the fraction of hosts (Ht) which escapes the attack of the
parasitoid (Pt). F is the host's reproductive function, which is often assumed constant,
but if F is allowed to express density dependence in the host population, i.e., F(Ht) =
exp[r(1-H,/k)] as per Beddington et al. (1975, 1978) then the dynamic behavior covers
a wider range of possibilities (including chaos). One simplified version we will consider is
xt+i = xt exp[r(1-xt) (yyt)1-m]
yt+l = xJl exp(-yyt)"m
(Allen, in press). This relationship incorporates Beddington's host density-dependence
and also the notion of parasite mutual interference (competition) (m) after Hassell &
Varley (1969). The equilibrium point for host and parasite densities will be stable (i.e.,
"attracting" for nearby densities) if the largest root (X) of the characteristic equation of
the system is less than one in absolute value, i.e., lKlm < 1 (Allen, in press). In Figure
1, KIm is plotted over the (r, m) plane for different y values. For unstable parameter
combinations (1Xm >1), the computer set IXIm = 1, producing the effect of an unstable
"plateau" for these values of (r, m, y) and a stable "valley" for stable regions. Chaotic
dynamics appear in parts of the unstable regions of this model (Allen, in press). Since r
and Y are the host reproductive rate and parasite attack rate, we have the result that
high r and/or y are destabilizing and may cause high amplitude chaotic oscillations and
local extinctions (particularly for high y). For r = 2, m = 0 and y = 7 (a fairly efficient
parasite) the model is chaotic (Fig. 2), and it appears that such an intense interaction
might cause local extinction.
The phenomenon of chaos is interesting enough in its own right, but in biological
population systems it poses some very fundamental and interesting questions. For
example, could natural selection acting on individuals produce behavioral phenotypes
having chaotic population parameters? Would these populations then have a high likeli-
hood of local extinction?
Insect Behavioral Ecology-'88 Allen
.05 .23 .4 7 .11.12
05 .3. 1 .05 .23 .
.77 .9 .1 .59 .77 .9 1
m r m r
Fig. 1. Stability surface for the host-parasitoid model eqs. (2). The model is stable
for lklm <1 (the "valley" in the figure) and unstable when 1XIm >1 (the "plateau" in the
figure). The unstable plateau to the left of the stable valley increases in size with
increasing parasite efficiency (-). Chaotic and quasiperiodic dynamics occur for parame-
ter values in this region.
In an attempt to answer such questions consider a modification of eqs. (2) (with m
= 0 for simplicity) with x', y' representing the t + 1 values of xt and yt (again to simplify
the writing). This gives
x' = xer(1-x)--y (4)
y x-1e Y(4)
y' = x(l-e-y)
where the fraction of the host found per parasite (-y) is given by
y = ak (5)
where a is the fraction of the area searched/parasitoid (Nicholson's "area of discovery")
and k is the environmental carrying capacity of the host (Oster 1976; Allen, in press).
As illustrated in Figures 1 and 2 the model tends to be destabilized by increasing
parasitoid efficiency (y).
Florida Entomologist 72(1)
Host Parasitoid Model
0 PARASITOID 0.75
Fig. 2. Dynamics of the host-parasitoid model eqs. (4) for r = 2, y = 7 (or eqs. (2)
with m = 0). Chaotic oscillations occur in both host and parasitoid (a,b) and a chaotic
or strange attractor is present in the parasitoid-host plane (c). With y = 7 this parasitoid
is so effective that it would probably cause local extinction.
Suppose now that the host changes its spatial distribution from random to patches
of area a (Fig. 3). If the parasitoid's original search covered an area s (s>o) the fraction
of this area containing hosts is now Nr/s where N is the number of patches in area s.
Insect Behavioral Ecology-'88 Allen
Host Patches within
Parasitoid Search Window
area = o
area = s
Fractional Host Area = No/s
(N = Patch Density)
Fig. 3. A parasitoid searching an area of size s containing N host patches each of
area a. The fraction of the host found by a randomly searching parasite is Nc/s.
If the parasitoid searches this new host distribution at random, its effective area of
discovery is reduced to a(Na/s), and the host carrying capacity in s becomes k(Nr/s).
The new y for the patchy host distribution is then given by
Ys = y(Nu/s)2 (6)
Thus, if a parasitoid searches a large area s at random for hosts in small patches of
area a, y, will be greatly reduced over a parasitoid that confines its search to the
patches. This suggests that a parasitoid which was strongly attracted to host patches
would increase and tend to destabilize an otherwise stable interaction. If we let s, the
area actually searched by the parasitoid, be XNa (X <1), then we can think of 1/X as a
parasitoid patch-concentrating factor and eq. (6) becomes
's = YE2 (7)
where e = 1/X can be interpreted as parasitoid patch efficiency (0
eqs. (4), can now be written as
X' = xer(l-x) -2ye
y' = x(l-e-Y ) (8)
Florida Entomologist 72(1)
where, as we might suspect, the model is destabilized by parasites with high patch
efficiency (Fig. 4). This is not meant to imply that parasite aggregation to host density
(Chesson & Murdoch 1986, Walde & Murdoch 1988) will always be destabilizing. More
complex models with more explicit patch-level dynamics might have different behavior.
It does seem intuitive, however, that a parasitoid which is highly effective in locating
its host will tend to eliminate the host from local areas and hence to have unstable local
dynamics (regardless of how the parasitoid finds its host).
Effect of Parasitoid Patch Efficiency
0 100 0 100
Fig. 4. Effect of parasitoid patch-finding efficiency (e) when searching for a host
which is distributed in patches (eqs. (8), r = 2, y = 7). The dynamics are chaotic for e
= 1.0 (a,b), quasiperiodic cycles occur when E = 0.75 (c,d). Attracting fixed points
occur when E = 0.5 (e) and E = 0.4 (f). Graphs (b) and (d) plot points only for a very
long time (10,000 points). For E<0.3, the parasitoid dies out and the host has persistent
cycles of its own.
Insect Behavioral Ecology-'88 Allen
TWO-PHENOTYPE HOST-PARASITOID INTERACTION
Surely the evolutionary battle between hosts and parasitoids must play an important
role in their dynamical interaction. Several authors have considered this battle in both
predator-prey type models (Pimentel 1961, Lomnicki 1971, Stewart 1971, Levin 1972)
and competition models (Le6n 1974, Lawlor & Maynard-Smith 1975, Rocklin & Oster
1976). Little attention has been given, however, to the interaction between natural
selection and chaos, i.e., whether natural selection might favor chaotic parameter
phenotypes in some situations.
Using the Nicholsonian framework, eqs. (4), and the previous discussion of a "patch-
efficient" parasitoid, we impose a minimum level of genetic complexity: a "patchy" host
phenotype and a "random" host phenotype having reproductive rates rl, and r2 at
frequencies p and (l-p). In addition, we assume a "patchy" parasitoid phenotype and a
"random" parasitoid phenotype at frequencies u and (1-u). The situation is illustrated
in Figure 5 where different attack rates (y,1, y12, y21 and y22) have been assigned to
Type Search Search
Frequency u 1 u
S Patchy p r ,711 r, ,71
Random 1 p r2 O, 21 r2 22
Fig. 5. Two parasitoid phenotypes: patchy search and random search present at
frequencies u and (1-u) attacking two host phenotypes: patchy distribution and random
distribution present at frequencies p and (l-p). We assume that the patchy parasitoid
is more efficient against the patchy host and conversely, so that y11>Y21 and y22>y12.
The patchy host and random host have reproductive rates ri and r2, respectively.
60 Florida Entomologist 72(1) March, 1989
the four possible types of encounters. We tacitly assume that the "patchy" parasitoid
phenotype is a patch efficient parasitoid as described in the previous section and, con-
versely, the random parasitoid phenotype is more efficient at finding randomly distrib-
uted hosts, i.e., in the convention of Figure 5, that y11 >721 and y22 >'12. Using the
phenotype frequencies we an now write the system from eqs. (4) as a weighted sum of
mutually exclusive proportions:
x' = pxer(l-x) [ue- "Y + (1-u)e '12Y] + (1-p)xer2 (1-x) [ue-21Y + (1-u)e-Y2Y]
y' = x[pu(1-e-1"Y)+ p(1-u)(1-e-72Y) + (1-p)u(1-e- Y) + (1-p)(1-u)(1-e -Y2)]
p' = pxerl(-x) [ue-YlY + (1-u)e -Y2Y]/x'
u' = ux[p(1-e-"'Y)+(1-p)(-e -21Y)]/y'
where x and y are the host and parasitoid, and p and u are the proportions of patchy
hosts and patch efficient parasitoids, respectively. Notice that p' and u' are simply the
fraction of the total reproductive output which is patchy host and patchy parasitoid
phenotype, respectively. Thus, eqs. (9) represent an extension of eqs. (4) which crudely
incorporate the dynamics of natural selection on phenotypes.
While eqs. (9) still omit many relevant details, and have the limitation of a discrete
time framework, their behavior is remarkably complicated and varied. One can find
virtually any kind of dynamical behavior simply by varying the 6 parameters and initial
conditions. While it is tempting to illustrate them all, there is neither time nor room
here, and only two examples are illustrated in Figures 6 and 7. Both densities and
phenotype frequencies apparently can be chaotic (Fig. 6), and initial conditions alone
can determine the kind of behavior which is observed (Fig. 7). In this regard, Figure
7 illustrates what appears to be chaos (but may be very complicated quasicycles) (Fig.
7a) in host-parasitoid densities and a beautiful attractor for phenotype frequencies (Fig.
7b), which may be chaotic or a very complex quasicycle (it is difficult to say which
without further analysis). When initial conditions are changed, one finds much simpler
quasicycles (Fig. 7c-d). This is not to say that other attracting behaviors do not exist
elsewhere in the initial condition space. I do not know.
Quasicycles in the phase plane (host vs. parasite) have the appearance of loops of
string, and during simulation, plotting points only, one sees the points processing
around on this loop until (if one plots many of them) they fill the loop completely (Fig.
7c-d). Phase plane portraits of chaotic attractors, using points only, tend to show a
discontinuous structure (Fig. 6b-c).
DISCUSSION AND CONCLUSIONS
From these results we cannot be sure if selection for chaotic parameter phenotypes
is likely to occur in host-parasitoid systems. It does seem fairly certain in this conceptual
framework that increasing parasitoid patch-finding efficiency destabilizes the interac-
tion, making chaos and local patch extinction more likely (Fig. 4), and in a general way
this makes intuitive sense. When genetics are added to the system in the form of two
simple parameter phenotypes for each species, dynamic behavior is more complex, with
extinction seeming somewhat less likely even for chaotic oscillations (Figs. 6-7).
Local patch extinction (whether the result of chaotic oscillations or whatever) does
not imply species extinction. One has, in addition to intrapatch dynamics, an interpatch
dynamics, so the question of whether patch chaotic oscillations would cause patch extinc-
tion is a moot point. The question really is: could one have chaotic interpatch dynamics
with frequent patch extinction without intrapatch (species) extinction. In such a system
we would see local chaos, but the species would endure.
As it was constructed here, the genetic model did not always go to fixation of the
highest reproductive rate host and the most efficient parasitoid (although this can hap-
Insect Behavioral Ecology-'88 Allen
Host-Parasitoid "Genetic" Model
500 1000 0 0.75
500 1000 0 1.0
TIME PATCHY PARASITOID FREQ
Fig. 6. Host-parasitoid "genetic" model, eqs. (9), r1 = 1, r2 = 2, Y11 = 5, Y12 = 2,
21 = 3, y,2 = 6. Initial conditions are: x = 0.1, y = 0.1, p = 0.5, u = 0.5. Apparent
chaotic osIIIations occur for both density and phenotype frequency. On the left (a,c)
are short time period graphs after allowing 500 time points for transient behavior to
decay. On the right (b,d) are 10,000 points plotted in the phase plane.
pen in some cases). Instead, the system exhibits a great variety of dynamical behaviors
depending upon the parameters and initial conditions. It is interesting in this regard
that the system in Figure 6 simply has an attracting 6-point cycle for the highest
reproductive and attack rates (r = 2, y = 6), but it is chaotic in density and phenotype
frequency once both phenotypes are present (even though the added phenotypes have
lower reproductive and attack rates).
A similar kind of phenomenon occurs in the example of Figure 7. For the highest
reproductive and attack rates there (r = 1.5, y = 7), a simple quasicycle is observed
which goes so close to zero as to suggest local extinction. When both phenotypes are
present, neither densities nor phenotype frequencies go to extinction, and simple
quasicycles coexist with apparently chaotic behavior determined by initial conditions
From both of these examples it appears that the addition of simple genetic selection
to the interaction has not indicated selection for or against chaotic parameter
phenotypes. Instead, the addition of genetics seems to have increased the likelihood of
chaos (i.e., at lower parameter values) and decreased the likelihood of extinction of
chaotic systems if they occur.
Florida Entomologist 72(1)
Sensitivity to Initial Conditions
ATCHY PARASITOID FREQ
Fig. 7. Different qualitative behaviors obtained from the host-parasitoid genetic
model, eqs. (9), by shifting the initial conditions. (In all graphs 10,000 points are plotted
in the phase plane.) Parameters are the same for all graphs: r1 = 1, r2 = 1.5, -y1 = 5,
Y12 = 2, Y21 = 3, Y22 = 7. The initial conditions (i.c.) for (a,b) are x = 0.1, y = 0.1, p
= 0.5, u = 0.5. For (c,d) the i.c. are x = 1, y = 0.05, p = 0.1, u = 0.9. Thus, one
can have behavior (a,b) or behavior (c,d) simply by an initial condition change.
I am indebted to Sandra J. Walde and J. H. Frank for helpful reviews. I also thank
Barbara Hollien for her patient typing and Susan Marynowski for help in preparing the
figures. This is Florida Agricultural Experiment Station Journal Series No. 9685.
ALLEN, J. C. Are natural enemy populations chaotic?, in L. L. MacDonald, J. A.
Lockwood, and J. A. Logan (eds.), Estimation and analysis of insect populations.
Springer, New York (in press).
BEDDINGTON, J. R., C. A. FREE, AND J. H. LAWTON. 1975. Dynamic complexity
in predator-prey models framed in difference equations. Nature 255: 58-60.
1978. Modelling biological control: on the characteristics of successful natural
enemies. Nature 273: 573-79.
CADE, W. 1975. Acoustically orienting parasitoids: fly phonotaxis to cricket song.
Science 190: 1312-13.
CHESSON, P. L., AND W. W. MURDOCH. 1986. Aggregation of risk: relationships
among host-parasitoid models. American Nat. 127: 696-715.
Insect Behavioral Ecology-'88 Allen 63
HASSELL, M. P. 1984. Parasitism in patchy environments: inverse density dependence
can be stabilizing. IMA J. Math. Appl. Med. Biol. 1: 123-33.
1985. Insect natural enemies as regulating factors. J. Anim. Ecol. 54: 323-34.
--, AND R. M. MAY. 1973. Stability in insect host-parasite models. J. Anim. Ecol.
- 1974. Aggregation in predators and insect parasites and its effect on stability.
J. Anim. Ecol. 43: 567-94.
---, AND G. C. VARLEY. 1969. New inductive population model for insect para-
sites and its bearing on biological control. Nature 223: 1133-36.
HEADS, P. A., AND J. H. LAWTON. 1983. Studies of the natural enemy complex of
the holly leaf-miner: the effects of scale on the detection of aggregative responses
and the implications for biological control. Oikos 40: 267-76.
HOGARTH, W. L., AND P. DIAMOND. 1984. Interspecific competition in larvae be-
tween entomophagous parasitoids. American Nat. 124: 552-60.
LAWLOR, L. R., AND J. MAYNARD-SMITH. 1975. The coevolution and stability of
competing species. American Nat. 110: 79-99.
LEON, J. A. 1974. Selection in contexts of interspecific competition. American Nat.
LEVIN, S. A. 1972. A mathematical analysis of the genetic feedback mechanism.
American Nat. 106: 145-64.
LEWIS, W. J., AND J. H. TUMLINSON. 1988. Host detection by chemically mediated
associative learning in a parasitoid wasp. Nature 331: 257-59.
LOMNICKI, A. 1971. Animal population regulation by the genetic feedback mechanism:
A critique of the theoretical model. American Nat. 105: 413-21.
MAY, R. M. 1978. Host-parasitoid systems in a patchy environment: a phenomenolog-
ical model. J. Anim. Ecol. 47: 833-44.
- AND M. P. HASSELL. 1981. The dynamics of multiparasitoid-host interactions.
American Nat. 117: 234-61.
MURDOCH, W. W., R. M. NISBET, S. P. BLYTHE, W. S. C. GURNEY, AND J. D.
REEVE. 1987. An invulnerable age class and stability in delay-differential
parasitoid-host models. American Nat. 129: 263-82.
NICHOLSON, A. J., AND V. A. BAILEY. 1935. The balance of animal populations.
Proc. Zool. Soc. London 3: 551-98.
OSTER, G. F. 1976. The dynamics of nonlinear models with age structure, p. 411-38
in S. A. Levin (ed.), Studies in mathematical biology. Part II: Populations and
communities. Studies in mathematics, vol. 16. Math. Assoc. of America.
PARKER, T. S., AND L. O. CHUA. 1987. Chaos: A tutorial for engineers. Proc. IEEE
PIMENTEL, D. 1961. Population regulation by the genetic feedback mechanism. Amer-
ican Nat. 95: 65-79.
ROCKLIN, S., AND G. F. OSTER. 1976. Competition between phenotypes. J. Math.
Biol. 3: 225-61.
SCHAFFER, W. M., AND M. KOT. 1985. Do strange attractors govern ecological sys-
tems. BioScience 35: 342-50.
1986. Chaos in ecological systems: the coals that Newcastle forgot. Trends in
Ecol. & Evol. 1: 58-63.
1987. Differential systems in ecology and epidemiology, in A. V. Holden (ed.),
Chaos. Princeton Univ. Press, Princeton, N.J.
STEWART, F. M. 1971. Evolution of dimorphism in a predator-prey model. Theor.
Pop. Biol. 2: 493-506.
WAAGE, J. K. 1983. Aggregation in field parasitoid populations: foraging time alloca-
tion by a population of Dragema (Hymenoptera, Ichneumonidae). Ecol. Ent. 8:
64 Florida Entomologist 72(1) March, 1989
WALDE, S. J., AND W. W. MURDOCH. 1988. Spatial density-dependence in
parasitoids. Annu. Rev. Ent. 33: 441-466.
WALKER, T. J. 1986. Monitoring the flights of field crickets (Gryllus spp.) and a
tachinid fly (Euphasiopteryx ochracea) in north Florida. Florida Ent. 69: 678-85.
THE PYCOMERINI (COLEOPTERA: COLYDIIDAE)
OF THE WEST INDIES
MICHAEL A. IVIE
Department of Entomology
Montana State University
Bozeman, MT 59717 USA
STANISLAW A. SLIPINSKI
Institute of Zoology
Polish Academy of Sciences
ul. Wilcza 64
00-679 Warsaw, POLAND
Eight species of Pycnomerus Erichson are recorded from the West Indies. A lectotype
is designated for Penthelispa longior Grouvelle, which is placed in synonymy with
Penthelispa infima Grouvelle as Pycnomerus infimus NEW SYNONYMY, NEW COMBI-
NATION. A lectotype is designated and the type locality restricted to Puerto Rico for
Penthelispa corpulenta Reitter, with Penthelispa aequicollis Reitter placed as a junior
synonym, under the name Pycnomerus corpulentus NEW SYNONYMY, NEW COMBINA-
TION. A lectotype is designated for Pycnomerus biimpressus Reitter. The following
species are described as NEW SPECIES: P. annae (Jamaica), P. darlingtoni (Jamaica),
P. hottae (Haiti), P. uniforms (Guadeloupe), and P. valentine (Hispaniola). A key and
illustrations are provided for the identification of adults of West Indian Pycnomerus
Se registran ocho species de Pycnomerus Erichson en las Indias Occidentales. Se
design el lectotipo de Penthelispa longior Grouvelle, cual especie esta reconocida como
sin6nimo de Penthelispa infima Grouvelle NUEVO SIN6NIMO, y esta listada como
Pycnomerus infima (Grouvelle) NUEVO COMBINACION. Se design el lectotipo de
Penthelispa corpulenta Reitter, y la localidad del tipo esta limitada a Puerto Rico.
Penthelispa aequicollis Reitter se reconoce como sin6nimo de esta especie y aparace
como Pycnomerus corpulentus (Reitter) NUEVO SIN6NIMO, NUEVA COMBINACION.
Se design el lectotipo de Pycnomerus biimpressus Reitter. Se described cinco NUEVAS
ESPECIES: P. annae (Jamaica), P. darlingtoni (Jamaica), P. hottae (Haiti), P. uni-
formis (Guadalupe), y P. valentinei (La Espafiola). Hay ilustraci6nes y una tabla dian6s-
tica para facilitar la identificaci6n de los adults de las species de Pycnomerus de las
Ivie & Slipinski: West Indian Pycnomerini
The Colydiidae are a small family of Tenebrionoidea, poorly understood at all
taxonomic levels, due in large part to the artificiality of the family in classical usage,
and the large number of taxa transferred in and out of the family in recent years (Ivie
& Slipifiski, in press a). The recent redefinition of the family (Lawrence 1980) and its
tribes (Slipifski & Burakowski 1988) and the cataloging of included genera (Ivie &
Slipifiski, in press a) provide a framework within which to produce taxonomic data on
poorly studied faunas.
The colydiids of the Neotropics are virtually unknown, and the West Indian species
are no exception. No comprehensive treatment of any group of colydiids has been pub-
lished for the West Indies, and only isolated descriptions, many inadequate, are available
for identification purposes. We hope this will be the first in a series of papers eventually
covering all West Indian members of this family.
The Pycnomerini are a small tribe, with 3 currently recognized genera. Only the
cosmopolitan Pycnomerus occurs in the West Indies. For a definition of the tribe see
Slipifski & Burakowski (1988), and for a key to the genera, Ivie & Slipifiski (in press b).
Penthelispa Pascoe, here considered a synonym of Pycnomerus, has been recognized
as distinct from Pycnomerus on the basis of a 2 versus 1 segmented antennal club. The
fallacy of this division was recognized by Sharp (1894) and Grouvelle (1908) who
synonymized them. The fact that the 1 segmented club is actually a very close association
of segments 10 and 11, and that intergrades between the loose and tight clubs occur,
makes it impossible to recognize 2 monophyletic lineages on this single character.
Hetschko (1930) and subsequent workers ignored Sharp's action, until Penthelispa was
returned to synonymy by Slipifski (1984). Dajoz' (1977: 175) assertion that the 2 are
separated by a ciliate fovea on the submentum of Pycnomerus is proved incorrect by
the presence of this sexual character in American loose-clubbed species, and none of his
key characters purporting to separate the 2 have any usefulness at the generic level.
Blackwelder (1945) listed 6 species of Pyconomerus and Penthelispa from the West
Indies. One of these, P. exaratus (Chevrolat) has been moved to Philothermus
(Cerylonidae) (Slipifski, in press); 1 (armata Erichson 1845: 291) is a nomen nudum; 2
others have been found to be synonyms (see below); and 1 additional species, listed by
Blackwelder (1945) from "Central America" has been found to be West Indian. To these
3 remaining described species, we add 5 more. These 8 species constitute a major portion
of the 22 currently recognized Neotropical species, although there are undoubtedly many
mainland species yet to be described.
Of the West Indian species, 2 are widespread both on several islands and the Neo-
tropical mainland, with 5 of the remaining 6 apparently being highland forms, each
endemic to a single island. The last occurs on 2 of the high islands of the Lesser Antilles.
Champion's (1898) record of Penthelispa sp. from St. Vincent remains of unknown
identity, as we saw no specimens from that island.
Pycnomerus are associated with dead, decaying wood in moist forests. They can
often be found in considerable numbers in such habitats, but they are often overlooked
due to their small size, brown color, and sluggish movements. Although specimens have
been seen from only the larger islands of the Greater and Lesser Antilles, Pycnomerus
probably occurs on any island supporting moist tropical forest. The general absence of
specimens from Cuba is of interest (only 2 specimens have been seen from Cuba). This
large, moist, forested island must be home to more species than are reported here.
The actual food of all Pycnomerus is probably fungi (Slipifiski 1984), and neither
adult nor larval morphology (Dajoz' 1977, Nikitsky & Belov 1980) give any indication
of support for Dajoz's (1977: 178) assertion that Pycnomerus "c'est sans doute un pr&-
dateur de Microarthropods". The mandibles and other mouthparts are clearly of a form
useful in crushing fungi or decayed wood, and are not fitted for a predatory lifestyle.
Examination of gut contents of preserved material showed no arthropod parts, but
66 Florida Entomologist 72(1) March, 1989
considerable wood and/or fungal material. Label data recorded below and personal ob-
servation (MAI) support the fungivorous habit of this genus.
Material utilized for this study is deposited in several collections and is cited in the
text with the following acronyms: AMNH-American Museum of Natural History, New
York, USA. CASC-California Academy of Science, San Francisco, USA. CNCI-
Biosystematics Research Centre, Agriculture Canada, Ottawa, Canada. FSCA-Florida
State Collection of Arthropods, Gainesville, USA. IJAM-Institute of Jamaica, Kings-
ton, Jamaica. INHS-Illinois Natural History Survey, Urbana, USA. IZPN-Institute
of Zoology, Polish Academy of Science, Warsaw, Poland. HFCM-H. Franz, private
collection, Mbdling, Austria. MAIC-Michael A. Ivie, private collection, Bozeman, Mon-
tana, USA. MCZC-Museum of Comparative Zoology, Cambridge, Massachusetts,
USA. MHND-Museo Nacional de Historia Natural, Santo Domingo, Dominican Repub-
lic. MNHN-Museum National d'Histoire Naturelle, Paris, France. NMNH-United
States National Musuem of Natural History, Washington. RSMC-Richard S. Miller,
private collection, Columbus, Ohio, USA.
KEY TO THE SPECIES OF WEST INDIAN PYCNOMERUS
1. Antennomeres 10 and 11 fused into a 1-segmented club (Fig. 1-4); pronotal
disk with a pair of distinct longitudinally impressed lines (Fig. 1-4); prono-
tum at base with a pair of confluent irregular foveae opposite elytral striae
3 and 4 (Fig. 1-4) .............................................................................. 2
1'. Antennomeres 10 and 11 free, forming a loose 2-segmented club (Fig. 8-10,
12); pronotal disk with at most faint depressions (Fig. 8-11); pronotum with
basal punctures either obsolete or uniform ............................................ 5
2. Elytron parallel-sided, at least 2.2X as long as wide and 2.3X as long as pro-
notum; interval 8 reaching anterior margin (Fig. 2); eyes normal, extending
onto ventral portion of head. Widespread ...... Pycnomerus biimpressus Reitter
2'. Elytron rounded laterally, less than 2X as long as wide and 2X as long as
pronotum; with intervalf 8 not reaching anterior margin (Fig. 1, 3, 4); eyes
reduced, entirely lateral. Mountains of Greater Antilles ............................ 3
3. Interstriae 6 and 8 fused at level of metacoxa or 1st visible sternite (Fig.
16); lateral margin of pronotum, in lateral view, markedly sinuate (Fig. 5).
Mountains of Jamaica .......................................... Pycnomerus annae n. sp.
3'. Interstriae 6 and 8 fused at level of third visible sternite (Fig. 14); lateral
margin of pronotum, in lateral view, nearly straight (Fig. 6-7) ................... 4
4. Pronotum with transverse medial sub-basal sulcus (Fig. 3); intervals costi-
form and uniform in area of fusion of 4, 5, and 6, not pustulate (Fig. 14);
penultimate visible sternite impunctate medially (Fig. 15). Blue Mountains
of Jamaica .................................................. Pycnomerus darlingtoni n. sp.
4'. Pronotum not bordered postero-medially (Fig. 4); intervals confused and
indistinct in area of fusion of 4, 5, and 6, pustulate; penultimate visible
sternite punctate medially (Fig. 17). La Hotte Peninsula of Haiti ..........
............................ .....Pycnomerus hottae n. sp.
5. Pronotum coarsely punctate, basal margin with line of coarse punctures, at
least toward sides (Fig. 8-9) ............................................................... 6
5'. Pronotum finely punctate; basal margin without line of punctures, not de-
lim ited (Fig. 10-11) ........................................................................... 7
6. Second elytral interval reaching anterior margin (Fig. 8), all intervals of
equal prominence; strial punctures on elytral disk elongate, narrowed
medially (Fig. 8, inset); last visible sternite flat, pronotum rounded laterally
(Fig. 8). Hispaniola, Cuba? ............................ Pycnomerus valentinei n. sp.
Ivie & Slipinski: West Indian Pycnomerini
Fig. 1-7. Pycnomerus spp. 1-4) dorsal view. 1) P. annae; 2) P. biimpressus; 3) P.
darlintoni; 4) P. hottae. 5) P. annae, lateral view of pronotum. 6-7) shape of pronotal
lateral margin in lateral view. 6) P. hottae; 7) P. darlingtoni.
6'. Second elytral interval not reaching anterior margin (Fig. 9), intervals 3, 5,
and 7 more prominent than 4 and 6, with 6 especially less prominent on disk;
strial punctures on elytral disk round (Fig. 9, inset); last visible sternite with
arcuate concavity; pronotum weakly bisinuate laterally (Fig. 9). Puerto
Rico .................................................... Pycnomerus corpulentus (Reitter)
7. Pronotal punctures round (Fig. 10); elytra less than 2.3X as long as wide,
less than 2.5 X as long as pronotum; last visible sternite concave; length
greater than 3 mm. Guadeloupe ...................... Pycnomerus uniforms n. sp.
Fig. 8-13. Pycnomerus spp. 8-10)
10) P. uniforms. 11-13) P. infimus.
mental pit of male.
\ 0O' 0 0 ,
dorsal view. 8) P. valentinei; 9) P. corpulentus:
11) pronotum; 12) antenna; 13) mentum and sub-
7'. Pronotal punctures shallowly elongate (Fig. 11); elytra more than 2.5X as
long as wide, more than 2.7X as long as pronotum; last visible sternite flat;
length less than 3 mm. Lesser Antilles ......... Pycnomerus infimus (Grouvelle)
Florida Entomologist 72(1)
Ivie & Slipinski: West Indian Pycnomerini 69
Pycnomerus biimpressus Reitter
(Fig. 2, 22)
Pycnomerus biimpressus Reitter 1877: 355 [Puerto Rico, Lectotype in MNHN].
Hetschko 1930: 61. Blackwelder 1945: 472. Wolcott 1951: 316.
Pycnomerus exaratus Champion 1898: 401 [not Chevrolat]. Wolcott 1951: 316.
Distribution: Cuba, Jamaica, Hispaniola, Puerto Rico, Tortola, Guadeloupe, Dominica,
Martinique, St. Vincent, Grenada, Central and South America.
From the type series, a single, partial, card-mounted specimen, missing the head
and prothorax, is in the MNHN. A second specimen, formerly mounted on the same
pin is missing, but the card remains. This pin bears 4 labels as follows: "Brit. Mus./
Portorico/ typ Reitter/ biimpressum Reitter". This partial specimen is here designated
lectotype, and has been so labeled.
This widespread species is 1 of 4 in the West Indies belonging to the tight-clubbed
group of Pycnomerus. The domed, biimpressed pronotum links all 4 of these West Indian
species. The other 3 species (P. hottae, P. annae, and P. darlingtoni) are narrow-en-
demics to high mountains of the Greater Antilles, and it seems possible that biimpressus
is the ancestor of the others, all of which share the incomplete elytral interval 8 and
reduced eyes. Of these shared characters, the first is associated with loss of wings, and
both loss of wings and reduction of eyes is common for montane West Indian endemics
in a variety of groups. These characters are shared with an undescribed Andean species
from Ecuador and Peru [CASC], which seems to belong to a different species group.
No real phylogenetic conclusions can be drawn without a revision of the entire genus,
an undertaking far beyond the scope of this work.
Diagnosis: The tight antennal club, with antenomeres 10 and 11 appearing as a single
ball, the biimpressed pronotal disk (Fig. 2), the large eyes extending onto the ventral
portion of the head, the elytra 2.2-2.3X as long as wide and 2.3-2.5X as long as the
pronotum, and elytral interval 8 reaching anterior margin, will distinguish individuals
of this species from those of all other known West Indian Pycnomerus spp. Some Greater
Antillean specimens have the eyes somewhat smaller and the pronotum more strongly
impressed than the Lesser Antillean populations, but this variation seems at most clinal
in nature, as expected in a widespread species with island populations, and is not indi-
cative of different taxa. Length 3.0-4.8 mm, male genitalia as in Fig. 22.
Material examined, in addition to type: BRAZIL: 1-[IZPN]. PERU: 1-Huanuco,
2500 m, Chinchao, 25 km below Carpish, 06 IX 1946, F. Woytkowski [AMNH].
GUATEMALA: 7-Coban, 4000 ft., 31 VII 1947, C. & P. Vaurie [AMNH, IZPN, MAIC].
CUBA: 1-Prov. Oriente, Gran Piedra Range, 30-31 V 1936, 2000-3000 ft.,
Darlington [MCZC]. JAMAICA: 4-Pt. Antonio, 7 I, A. E. Wright [IZPN, MCZC].
5-Whitfield Hall, Blue Mts., nr. 4500 ft., 13-20 VIII 1934, Darlington [MCZC]. 1-
Gordon Tn., 4 II 1937, sta. 382, Blackwelder and Chapin [NMNH]. 1-Whitfield Hall,
Blue Mts, nr 4500 ft., 13-20 V111 1934, Darlington [IZPN]. 1-Chincona, 26 II 1911
[AMNH]. 1-Hardwar Gap, 13 XI 1966, A. B. Gurney [NMNH]. 21-ibid., 4000 ft.,
5-29 VII 1966, Howden and Becker [CNCI]. 22-ibid., 10-17 VII 1966, A. T. Howden
[CNCI, IZPN, MAIC]. 14-ibid., 16 XII 1972, J. Peck [CNCI, IZPN, MAIC]. 3-ibid.,
16 XII 1973, S. & J. Peck, under bark [CNCI]. 5-Portland Par., Hardwar Gap, 5 XII
1975, C. W. & L. O'Brien & Marshall [CASC]. 14-St. Andrew Par., Hardwar Gap,
6-8 XII 1975, G. F. Hevel [IZPN, MAIC, NMNH]. 1-St. Andrew Par., St. Peters, 18
VII 1966, A. T. Howden [CNCI]. 6-Blue Mountains, H. Franz [HFCM, IZPN]. HIs-
PANIOLA: Haiti: 9-Furcy, W. M. Mann [IZPN, MCZC]. 1-Desbarriere, Massif La
Hotte, nr. 4000 ft., 12-14 X 1934, Darlington [IZPN]. 6-La Visite & vic., La Selle
Range, 5-7000 ft., 16-23 IX 1934, Darlington [MCZC]. 9-Dept. Sud-Oueste, Parc Na-
tional La Visite, nr. Headquarters, 1880 m., 10 V 1984, M. C. Thomas, in rotting log
Florida Entomologist 72(1)
14 o Dooo0oooooo00
oo%1 oo 17
op o 0i 0 17
Fig. 14-20. Pycnomerus spp. 14-15) P. darlingtoni. 14) elytra in 3/4 lateral view; 15)
visible sternites III-V. 16) P. annae, elytra in 3/4 lateral view. 17) P. hottae, visible
sternites III-V. 18-20) male genitalia. 18) P. valentine; 19) P. corpulenta; 20) P. hottae.
,." ;0 00 0
'a 00 C
Fi.1-0 ynmrssp 41)P alngoi 4 ltai / aea iw 5
viibe trnte IIV.1) nne eytain3/ atra ve. 7)P htte vsil
striesIIV.1-0 ml entla.1)P vlnini 9)P opuet; 0 ote
Ivie & Slipinski: West Indian Pycnomerini 71
of Pinus occidentalis [FSCA, IZPN, MAIC]. 5-Dept. Sud-Oueste, Parc National La
Visite, ca. 1km S. Roche Plat, 22 V 1984, M. C. Thomas [FSCA, IZPN, MAIC]. 1-Dept.
Sud-Oueste, Massif de La Selle, Morne d'Enfer, 1850 m., 16 V 1984, M. C. Thomas
[FSCA]. 5-Morne Gimby, 22 km SE Fond Verrettes, 8 VII 1956, 6500', B. & B. Valen-
tine, Foret des Pines, hardwood cloud forest, beating [MAIC, IZPN]. 1-Dept. l'Ouest,
Fond Verrettes to Refuge, 28 V 1950, H. B. Mills [INHS]. HISPANIOLA: Dominican
Republic: 1-Loma Vieja, ca. 6000 ft., S. Constanza, VIII 1938, Darlington [MCZC].
2-Constanza, 3-4000 ft. VIII 1938, Darlington [IZPN, MCZC]. 3-San Jose de las
Matas, 1-2000 ft., VI 1938, Darlington [IZPN, MAIC, MCZC]. PUERTO RICO: 1-El
Yunque, c. 3000 ft., V 1938, Darlington [MCZC]. 9-El Yunque Sta., Luquillo Forest,
6-16 VII 1969, H. & A. Howden [CNCI, MAIC, IZPN]. 1-Cerro Dona Juana, Ponce,
28 XII 1966, ex Polyporus zonalis, S. Peck [IZPN]. 2-Maricao For. Res., Hwy 120,
KllH8, 26 VII 1979, O'Briens & Marshall [IZPN, MAIC]. 5-Maricao For. Res., Hwy
120, K19H8, 25 VII 1979, L. O'Brien & Marshall [IZPN, MAIC]. 1-Caribbean Nat.
For., El Yunque Hwy (191), K11H4, 29 VII 1979, G. B. Marshall, under bark [IZPN].
3-El Yunque, Rt. 191, km 9.7, 16 VIII 1961 [NMNH]. 1-Guilarte For. Res., Hwy
131 & 158, 23 VII 1979, L. B. O'Brien [MAIC]. 1-Caribbean National Forest, 1km. S.
Palmer, 200 m, 23 IX 1987, M. A. Ivie, under bark [MAIC]. TORTOLA (BRITISH
VIRGIN IS.): 1-Mt. Sage Nat. Park, 460m, 7-8 XII 1985, S. & P. Miller [NMNH].
GUADELOUPE: 4-Gourbeyre [AMNH, MAIC]. DOMINICA: 19-Long Ditton, 14-19 VI
1911 [AMNH, IZPN, MAIC]. 1-St. Joseph Par., Wet Area Exp. Sta., 800 ft, 31 X11
1978, M. A. & L. L. Ivie [MAIC]. 2-4 mi S. Salisbury, 19 VIII 1986, C. W. & L. B.
O'Brien [IZPN, MAIC]. 1-6 mi E. Dublanc, 16 VIII 1986, C. W. & L. B. O'Brien
[MAIC]. 1-6 mi E. Salisbury, Morne Apion, 2500', 19 VIII 1986, C. W. & L. B.
O'Brien. 1-N. Pont Cass6, 1500 ft., 25 VI 1969, P. J. Darlington, Jr. [MCZC]. MAR-
TINIQUE: 4-12 km. N. Fort de France (N3), 23 VIII 1986, C. W. & L. B. O'Brien
[IZPN, MAIC]. 1-5 km SE Morne Rouge, 24 VIII 1986, Forest Rd., C. W. & L. B.
Pycnomerus annae NEW SPECIES
(Fig. 1, 5, 16, 21)
This is an extremely well marked species, apparently restricted to the mountains of
Jamaica. The AMNH series from Montego Bay may represent a lowland population,
but the meaning of the code number F2192 on the under side of the labels is unknown.
The specimens were probably collected by L. B. Woodruff, but the location of his
notebooks is unknown. These specimens are probably from the mountains south-east of
Pycnomerus annae is sympatric with P. biimpressus at Hardwar Gap, elev. 1212
m., on the border of the Parishes of St. Andrew and Portland, and at Chincona, 1515
m., but seems to occupy the middle elevational habitats in Jamaica, with P. biimpressus
below and P. darlingtoni above.
Diagnosis: the compact antennal club (Fig. 1), reduced eyes (Fig.1), elytral intervals
2 (Fig. 1) and 8 (Fig. 16) not reaching anterior margin, and the fusion of intervals 6-8
at the level of metacoxa to 1st visible sternite (Fig. 16) will distinguish individuals of
this species from those of all other West Indian Pycnomerus.
Description. Male: elongate oval, rufo-castaneous, wingless.
Head (Fig. 1) with surface finely punctate anterior to antennal insertions, coarsely
umbilicate punctate behind; anterior margin weakly emarginate, angles square; frons
with distinct depressions media of supra-antennal ridges; supra-antennal ridges dis-
tinctly punctate, narrow anterior to frontal depressions, wider laterad of depressions.
Antenna (Fig. 1) with compact club of antennomeres 10 and 11; 10 slightly larger than
Florida Entomologist 72(1)
Fig. 21-25. Pycnomerus spp., male genitalia. 21) P. annae; 22) P. biimpressus 23)
P. darlingtoni; 24) P. infimus; 25) P. uniforms.
11. Eye reduced (Fig. 1), approximately 13-15 facets, not extending below ventral edge
of antennal insertion; antennal groove wide, shallow, well marked; submentum with
small median ciliate fovea at anterior margin.
Ivie & Slipinski: West Indian Pycnomerini 73
Pronotum (Fig. 1) nearly as wide as long, sides weakly bisinuate in dorsal view,
distinctly sinuate in lateral view (Fig. 5); lateral margins distinctly crenulate, bordered
above by a narrow sulcus, which is wider medially, making raised portion of disk nar-
rowed medially; sulcus extending around anterior angles onto anterior margin; anterior
angles distinctly produced, posterior angles rounded, not distinct; anterior margin dis-
tinctly bisinuate; posterior margin broadly rounded; disk distinctly impressed medially,
with a longitudinal raised nearly impunctate callus dividing the depression into 2 impre-
ssed lines; a pair of large, confluent foveae, connected across base by a narrow sulcus
present opposite elytral striae 3 and 4. Prosternum uniformly coarsely umbilicate
punctate; metasternum short, subequal in length to mesofemur. Elytra 2X as long as
broad, sides weakly rounded to subparallel from just behind humeri to declivity; hum-
eral angles microdenticulate; strial punctures elongate, some slightly narrowed medially
(Fig. 1, inset); intervals 2 and 8 not reaching anterior margin (Fig. 16); intervals 6-8
fused at level of hind coxae (Fig. 16); 4, 5, and 6 distinct and smooth in area of fusion
Abdomen coarsely umbilicate-punctate, punctures less coarse and dense medially on
visible sternites 1-3; 4th visible sternite with swelling postero-medially, 5th distinctly
concave. Genitalia as in Fig. 21.
Female: differs from male in lacking a ciliate fovea on submentum.
Length: 3.5-4.6 mm.
Holotype (male): Jamaica, 4000'; Hardwar Gap; VII-10-1966; A. T. Howden [CNCI].
Paratypes: JAMAICA: 9-same data as holotype [CNCI, IZPN, MAIC]. 1-ibid., 24
VII 1966, A. T. Howden [CNCI]. 1-ibid., 17 VII 1966, A. T. Howden [IZPN]. 1-ibid.,
7 VII 1966, Howden & Becker [CNCI]. 13-ibid., 10 VII 1966, Howden & Becker
[CNCI, IZPN, MAIC]. 1-ibid., 29 VII 1966, Howden & Becker [CNCI]. 61-St. An-
drew Par., 16 XII 1972, J. Peck [CNCI, IJAM, IZPN, MAIC]. 5-ibid., 16 XII 1973,
S. & J. Peck, under bark [CNCI, IZPN, MAIC]. 4-Portland Par., Hardwar Gap, 5
XII 1975, C. W. & L. O'Brien & Marshall [CASC]. 19-St. Andrew Par., Hardwar Gap,
6-8 XII 1975, G. F. Hevel [IZPN, MAIC, NMNH]. 7-Hardwar Gap, 13 XI 1966, A.
B. Gurney [NMNH]. 11-Montego Bay, 15 III 1911, F2192 [AMNH, IZPN, MAIC].
2-Cinchona, 26 II 1911, F2141 [AMNH]. 1-ibid., 24 II 1911, F2134 [AMNH]. 1-ibid.,
10 V 1941, Chapin [NMNH]. 4-ibid., 5000' I 1912, C. T. Brues [MCZC]. 1-ibid., 16
VIII 1934, Darlington [MCZC].
Etymology: This species is named in honor of Anne T. Howden, of Ottawa, Canada,
in honor of her extensive collecting of Jamaican beetles, including a portion of the type
series of this species, and in recognition of her many courtesies over the years.
Pycnomerus hottae NEW SPECIES
(Fig. 4, 6, 17, 20)
Discussion. This species is known only from the heights of Mt. La Hotte, an isolated
peak near the western end of the La Hotte Peninsula of southern Haiti. This is one of
the most unique and remote areas in the West Indies, and undoubtedly worthy of
protection from environmental degradation.
Pycnomerus hottae shares many similarities with P. darlingtoni of Jamaica, but
whether this reflects phylogeny or convergence due to reduction of eyes and loss of
wings is a matter of conjecture.
Diagnosis: The compact antennal club (Fig. 4), reduced eyes (Fig. 4), elytral inter-
vals 2 (Fig. 4) and 8 not reaching the anterior margin, intervals 6-8 fused at the level
of the 3rd visible sternite, and the confused, pustulate interval where 4-6 fuse will
distinguish individuals of this species from those of all others known from the West
74 Florida Entomologist 72(1) March, 1989
Description. Male: elongate oval, convex, shining rufo-castaneous, probably wing-
less, entire ventral surface with coarse umbilicate punctures.
Head (Fig. 4) with surface coarsely umbilicate-punctate; anterior margin straight,
anterior angles subangulate; frons with distinct depressions media of suprantennal
ridge; suprantennal ridge nearly impunctate, narrow before frontal depressions, wide
between depressions and lateral margins. Antennae (Fig. 4) with compact club of anten-
nomeres 10 and 11; 10 and 11 equal in size. Eye reduced (Fig. 4), approximately 10
facets; not extending below level of ventral edge of antennal insertion; antennal groove
wide, shallow, and relatively high due to small eye. Submentum with medio-anterior
Pronotum (Fig. 4) as long as broad; sides weakly diverging to anterior 1/5, then
weakly convergent; lateral margins crenulate, bordered above by narrow sulcus, weakly
sinuate in lateral view (Fig. 6); anterior angles square; posterior angles obtuse; anterior
margin bisinuate, without submarginal groove; posterior margin weakly lobed; disk
coarsely umbilicate-punctate as on head; disk with median depression, divided near base
by median longitudinal impunctate callus, resulting in 2 impressed lines on either side;
a pair of very coarse, confluent foveae present opposite elytral striae 3 and 4. Proster-
num uniformly punctate; metasternum shorter than mesofemur. Elytra approximately
2X as long as wide and 1.7-1.8X as long as pronotum, widest behind humeri, sides
weakly arcuate; anterior margin not bordered; humeral angles microdenticulate; strial
punctures elongate, narrowed medially on each side by a small denticle; intervals 2 and
8 not connected to anterior margin; intervals 6-8 fused at level of third visible sternite
as in P. darlingtoni (Fig. 14); 4, 5 and 6 confusedly fused, intervals pustulate in this
Abdomen (Fig. 17) with swelling on hind margin of 4th visible sternite, 5th distinctly
concave. Genitalia as in Fig. 20.
Length: 4.0-4.2 mm.
Holotype (male): Mt. La Hotte; 5-7800 ft.; Oct 6-7/ Haiti; 1934; Darlington. [MCZC].
Paratype: 1 male, same data as holotype [MCZC].
Etymology: Named for Mont La Hotte, on the La Hotte Peninsula of southern Haiti.
Pycnomerus darlingtoni NEW SPECIES
(Fig. 3, 7, 14, 15, 23)
Known only from the highest parts of the Blue Mountains of Jamaica, P. darlingtoni
shares that island with P. annae and P. biimpressus. There seems to be some eleva-
tional stratification, with the widespread biimpressus in the lower areas, overlapping
at the upper end of its range with the endemic mid-elevation annae and the endemic
darlingtoni restricted to the highest elevations. Further collecting is required to docu-
ment the full ranges of Pycnomerus species on Jamaica.
Diagnosis: the compact antennal club (Fig. 3), reduced eyes (Fig. 3), elytral intervals
2 (Fig. 3) and 8 (Fig. 14) not reaching anterior margin, and smooth intervals 4-6 fusing
at the level of visible sternite 3 (Fig. 14) will distinguish members of this species from
all others known to occur in the West Indies.
Description. Male: elongate oval, shining rufo-castaneous, probably wingless.
Head (Fig. 3) with surface moderately coarsely umbilicate-punctate above, more
densely punctate to subrugose below; anterior margin very weakly bisinuate, angles
weakly produced; frons with distinct depressions media of suprantennal ridges; supran-
tennal ridges nearly impunctate, narrow anterior to frontal depressions, wide between
depressions and lateral margins. Antenna (Fig. 3) with compact club of antennomeres
10 and 11, 10 slightly larger than 11. Eyes (Fig. 3) reduced, approximately 20 facets,
Ivie & Slipinski: West Indian Pycnomerini
not extending below level of ventral edge of antennal insertion; antennal grove wide,
shallow, and relatively high due to small eye. Submentum medio-anteriorly with ciliate
Pronotum (Fig. 3) slightly longer than wide, sides diverging slightly from base to
near anterior margin; lateral margins nearly straight in lateral view (Fig. 7), smooth to
subcrenulate; bordered above by a narrow sulcus, which is wider medially, making
raised portion of disk narrowed medially; this sulcus extending around anterior angles
onto anterior margin; anterior and posterior angles rounded to angulate; anterior margin
bisinuate; posterior margin nearly straight; disk coarsely umbilicate-punctate, medial
punctures larger than those on head; disk with median depression, divided near base
by median longitudinal callus, resulting in an impressed line on each side; a pair of very
coarse, confluent foveae present opposite elytral striae 3 and 4, connected across base
by a narrow sulcus. Prosternum uniformly coarsely umbilicate-punctate. Metasternum
slightly shorter than mesofemur. Elytra 1.9X as long as wide, 2X as long as pronotum;
sides weakly rounded, widest just behind humeri; humeral angles microdenticulate;
strial punctures elongate, slightly or not narrowed medially; intervals 2 (Fig. 3) and 8
(Fig. 14) not connected to anterior margin; intervals 6-8 fused at level of third visible
sternite (Fig. 14); 4, 5, and 6 distinct and smooth in area of fusion.
Abdomen (Fig. 15) coarsely umbilicate-punctate, with punctures less dense medially
on 3rd visible sternite, medially impunctate on 4th; 4th visible sternite with swelling
medially at posterior margin; 5th visible sternite deeply concave. Genitalia as in Fig. 23.
Female: differs from the male in lacking ciliate fovea on submentum.
Length: 4.4-4.8 mm.
Holotype (male): Main Range; Blue Mts; 5-7388 ft.; Aug 17-19/Jamaica; 1934; Dar-
Paratypes: 11 same data as holotype [IZPN, MCZC]; 4-Blue Mt.; Peak, Jan.; 12-13-
90 [IZPN, MAIC, MCZC].
Etymology: This species is named for the collector of the holotype, Prof. P. J.
Darlington, Jr., whose contributions to our knowledge of the beetle fauna of the Greater
Antilles are so great as to require no further comment.
Pycnomerus corpulentus (Reitter) NEW COMBINATION
(Fig. 9, 19)
Penthelispa corpulenta Reitter 1877: 351 [Type locality "America mer?", here restricted
to Puerto Rico; lectotype in MNHN]. Hetschko 1930: 65. Blackwelder 1945: 472.
Penthelispa aequicollis Reitter 878: 23 [Puerto Rico; Berlin?]. Hetschko 1930: 65.
Blackwelder 1945: 472. Wolcott 1951: 316. NEW SYNONYMY.
Penthelispa aequeicolle; Leng and Mutchler 1914: 413 [part].
Distribution: Mountains of Puerto Rico.
Two syntypes of P. corpulenta are housed in the MNH, on the same pin, labeled
"Penthis.; America/ typ Reitter; corpulenta m.". The upper specimen is here designated
lectotype, and the lower paralectotype, and are so labeled. These specimens are identi-
cal to specimens from Puerto Rico.
Penthelispa aequicollis is synonymized on the basis of the description and type
locality, as no types have been seen. The type should be in Berlin.
This species is related to the tight-clubed group, including P. biimpressus, with
which it shares Puerto Rico, by the the form of the elytral intervals. Pycnomerus
corpulentus is superficially similar to P. valentinei from Hispaniola, but can be easily
distinguished by the characters in the key and diagnoses.
Diagnosis: the loose antennal club (Fig. 9), coarsely punctate pronotum (Fig. 9), and
2nd elytral interval not reaching the anterior margin (Fig. 9) will distinguish this
76 Florida Entomologist 72(1) March, 1989
species. Further characters include the concave last visible sternite, weakly bisinuate
lateral pronotal margins (Fig. 9), and the 6th elytral interval lower than the 5th and
7th, although this last may be weak in some individuals. It is a relatively large species,
length: 3.1-4.6 mm, male genitalia as in Fig. 19.
Material examined, in addition to types: PUERTO RICO: 10--El Yunque, 21 II
[MCZC]. 5--Oeste Peak, 18 II 1968, L. Herman, rotting stump [AMNH, IZPN, MAIC].
1-El Yunque, 2,100-2,200 ft., 15-24 II 1969, T. & B. Hlavac & L. Herman Jr. [AMNH].
5-El Yunque Sta., Luquillo Forest, 2-16 VII 1969, H. & A. Howden [CNCI, IZPN].
1-El Yunque, 16-17 VII 1958, M. W. Sanderson [INHS]. 1-Caribbean Nat. Forest,
El Yunque, 2 VII 1979, M. A. Ivie [MAIC]. 1-Caribbean Nat. Forest, El Yunque trail,
610-1050 m, 23 IX 1987, M. A. Ivie, beating [MAIC]. 2-5 mi NE Jayuya, 17-19 VII
1969, H. & A. Howden [CNCI, MAIC]. 20-El Yunque, c. 3,000 ft. V 1938, Darlington
[IZPN, MAIC, MCZC]. 2-Caribbean Nat. For., El Toro Negro D., Hwy 143, k16-
18H4, 21 VII 1979, L. O'Brien & G. B. Marshall [IZPN]. 2-Villalba, Ins. Gov. Finca,
9 VII 1934, R. G. Oakley, in decaying wood.
Pycnomerus valentinei NEW SPECIES
(Fig. 8, 18)
The type series of P. valentinei is from various localities above 1,500 m in the Massif
de la Selle of south-eastern Haiti and the Cordillera Central of the Dominican Republic.
It is narrowly sympatric with P. biimpressus. A single male specimen from the lowland
locality of Soledad (Cienfuegos Prov.), Cuba [MCZC], is 1 of only 2 representatives of
the genus seen by us from Cuba (the other being P. biimpressus). It has the basic
appearance of the upland Hispaniolan P. valentinei, and externally can be separated
from that species only by the less strongly punctate prosternum. In the absence of more
Cuban specimens, it is impossible to ascertain the extent of variation in such a poten-
tially plastic character, and therefore this specimen is placed here, but not made part
of the type series. One factor in this case is the fact that the provenance is a botanical
garden, and as such, subject to adventive populations of introduced species. The El
Aceitillar paratypes were taken on an unidentified fungus between the laminae of a wet,
rotten pine log, ca. 1/2 meter in diameter, and associated with Clinidium corbis Bell
(Rhysodidae) and phrenapatine tenebrionids near Diodeus.
Diagnosis: the loose antennal club (Fig. 8), flat elytral intervals all reaching the
anterior margin (Fig. 8), and coarse pronotal punctation (Fig. 8) will distinguish this
species. Further, the pattern of fusion of elytral intervals in the female is unique (see
below). The flat last sternite characteristic of the species is subject to some variation.
One male from Constanza, Dominican Republic [MCZC], has the last visible sternite
slightly concave, but not nearly so distinctly so as in P. corpulenta.
Description. Male: elongate, parallel sided; rufo-brunneus; fully winged.
Head (Fig. 8) densely punctate, punctures moderately coarse; anterior margin dis-
tinctly emarginate; anterior angles rounded; frons on each side with distinct depressions
media of antennal insertions, without differentiated supra-antennal ridge. Antenna
(Fig. 8) with loose club of antennomeres 10 and 11, 10 subequal to 11. Eyes normal Fig.
8), easily visible from below. Antennal groove narrow, short, distinct, directed 450
below anterior-posterior line. Submentum with a large median ciliate fovea at anterior
Pronotum (Fig. 8) as wide as long, sides rounded to sinuate, lateral margins smooth;
anterior angles indistinct to moderately produced, rounded; posterior angles obtuse;
anterior margin nearly straight to weakly bisinuate, not bordered; posterior margin
straight on each side, weakly obtusely angulate medially, narrowly bordered, with 1
(sometimes 2) larger separate punctures opposite stria 3 to 4; disk coarsely punctate,
Ivie & Slipinski: West Indian Pycnomerini 77
with narrow impunctate longitudinal callus on posterior 2/3. Prosternum coarsely
punctate, metasternum longer than mesofemur. Elytra 2.1-2.2X as long as wide, 2.3-
2.5X as long as pronotum; sides subparallel; humeral angles distinct, smooth; strial
punctures elongate, narrowed medially; all intervals reaching anterior margin (Fig. 8);
intervals 4-8 fused as in uniforms.
Abdomen coarsely, uniformly punctate; 4th visible sternite with very slight swelling
medially on hind margin; 5th flat. Genitalia as in Fig. 18.
Female: differs from male in lacking fovea on submentum, in having the pattern of
fusion of elytral intervals ((((8 + 7) + 6) +5) + 4) rather than ((8 + 7 + 6)+5 + 4), and in
more strongly punctate head and pronotum.
Length: 4.2-5.2 mm.
Holotype male: Morne Guimby, 22 km.; SE Fond Verrettes; Haiti 18 JUL 1956;
6500' B. & B. Valentine/ Foret des Pines; Hardwood cloud; forest, beating [from MAIC,
deposited in NMNH].
Paratypes: HISPANIOLA: Haiti: 1-Morne Guimby, Foret des Pines, 22 km SE Fond
Verrettes, 6500' 18 VII 1956, B. &. B. Valentine, beating in hardwood cloud forest
[MAIC]. 1-Dept. l'Ouest, Fond Verrettes to Refuge, 28 V 1950, H. B. Mills [INHS].
2-Dept. Sud-Oeste, Massif del La Selle, Morne d'Enfer, 1850 m., 16 V 1984, M. C.
Thomas [FSCA, MAIC]. 14-Dept. Sud-Oeste, Parc National La Visite, vicinity park
hdqtrs, 1880 m., 10 V 1984, M. C. Thomas, in rotten logs of Pinus occidentalis [FSCA,
IZPN, MAIC]. 8-Dept. Sud-Oeste, Parc National La Viste, near park hdqtrs, 1880
m., 10 V 1984, M. C. Thomas, in rotten logs of Pinus occidentalis [FSCA, IZPN,
MAIC]. 2-Dept. Sud-Oeste, Parc National La Viste, ca 1 km S. Roche Plat, 22 V 1984,
M. C. Thomas, [FSCA, IZPN]. 1-La Visite & vie., La Selle Range, 5-7000 ft., 16-23
IX 1934, Darlington [MCZC]. HISPANIOLA: Domincan Republic: 1-vic. Valle Nuevo,
VIII 1938, ca. 6000' cloudforest, Darlington [MCZC]. 2-Constanza to Jarabacoa, VIII
1938, Darlington [MCZC]. 4-LaVega Prov., 28 km SE Constanza, 4 VIII 1979, C. W.
O'Brien [IZPN, MAIC, MHND]. 2-Prov. Pedernales, ca. 35 km N. Cabo Rojo, El
Aceitillar, 1363 m. 26 VIII 1988, in rotten pine log, M. A. Ivie [MCZC].
Other Material Examined: 1-Cuba, Cienfuegos, Soledad, V-VI 1939, C. Parsons
[MCZC] (see discussion above).
Etymology: Named in honor of its collector, Prof. Barry D. Valentine of The Ohio
State University, and in recognition of his generous help in MAI's study of West Indian
Pycnomerus uniforms NEW SPECIES
(Fig. 10, 25)
Penthelispa aequeicolle; Leng and Mutchler 1914: 413 [not Reitter] [part].
This species is unique among West Indian species in lacking any peculiar sculptural
structure, and has a basically generalized form for the genus. This is the only 1 of the
3 known Lesser Antillean species (others are P. biimpressus and P. infimus) not known
to occur on the South American mainland.
Diagnosis: the loose antennal club, finely punctate pronotum with round punctures,
concave last visible sternite, and size (3.3-3.7 mm) will distinguish members of this
Description. Male: elongate, parallel sided; rufo-brunneus, probably winged.
Head (Fig. 10) evenly covered with moderate punctures of medium density; anterior
margin emarginate, angles rounded; frons on each side with distinct depressions media
of antennal insertions, lacking differentiated supra-antennal ridges. Antenna (Fig. 10)
with moderately loose club of antennomeres 10 and 11; 11 longer but narrower than 10.
Eyes normal (Fig. 10), easily visible from below; antennal groove narrow, short, di-
78 Florida Entomologist 72(1) March, 1989
rected 450 below anterior-posterior line; submentum with very small median ciliate
fovea just behind anterior margin.
Pronotum (Fig. 10) nearly as long as wide, sides nearly parallel, only slightly
rounded; lateral margin smooth; disk moderately punctate, plane, evenly rounded to
lateral margins; anterior angles weakly produced, posterior angles rounded; anterior
margin weakly bisinuate, not bordered; posterior margin rounded, not bordered. Pros-
ternum shinning, with sparse punctures, most coarse laterally and posteriorly; metas-
ternum longer than mesofemur. Elytra with sides parallel, 2.2-2.3X as long as wide,
2.4-2.5X as long as pronotum; humeral angles prominent, smooth; strial punctures elon-
gate, linear; all intervals reaching anterior margin, intervals 4-5-6 and 6-7-8 fused at
the level of the 3rd and 4th visible sternites respectively.
Abdomen coarsely, sparsely punctate, 4th visible sternite with a slight swelling
medially on posterior margin, 5th concave. Genitalia as in Fig. 25.
Female: differs from male in lacking submental fovea.
Length: 3.3-3.7 mm.
Holotype (male): Ace. 4860; Gourbeyre; Guadeloupe [AMNH].
Paratypes: 5 (2 males, 3 females) same data as holotype [AMNH, IZPN, MAIC]. 2
(1 male, 1 female)-Guadeloupe, Trois Rivieres, Dufau [NMNH]. 2 (1 male, 1 female)-
Dominica, St. Joseph Par., Wet Area Exp. Sta., 800 ft., 31 XII 1978, M. A. & L. L.
Ivie [MAIC, RSMC].
Etymology: The name refers to the flat and uninteresting form of this species.
Pycnomerus infimus (Grouvelle) NEW COMBINATION
(Fig. 11, 12, 13, 24)
Penthelispa infima Grouvelle 1902: 464 [Martinique; MNHN?]. Leng & Mutchler 1914:
Penthelispa longior Grouvelle 1913: 294. [Guadeloupe; lectotype in MNHN]. Leng and
Mutchler 1914: 413. NEW SYNONYMY.
Pycnomerus longior; Blackwelder 1945: 472.
Penthelispa infirma Blackwelder 1945: 472 [misspelling].
Distribution: Guadeloupe, Dominica, Martinique, Brazil.
Grouvelle (1902) based infima on 1 specimen each from Martinique and Bahia, Brazil,
indicating the Martinique specimen as "d6f.", apparently a valid indication of holotype.
This specimen has not been found in the Paris collection, but the Bahia paratype has
been studied, and the synonymy based upon it. Of longior, 6 syntypes on 3 pins in the
MNHN are labeled "Guadeloupe Dufau/type/ Penthelispa longior ty. Grouv." The left-
hand specimen on the first card is here designated lectotype, the other 5 paralectotypes,
and are so labeled.
This species can be expected on at least the high islands of the Windward Islands
from St. Lucia to Grenada.
Diagnosis: the small size alone (2.0-2.4 mm) will distinguish individuals of this species
among West Indian Pycnomerus. Further, the loose antennal club (Fig. 12); fine, elon-
gate pronotal puncation (Fig. 11), and flat last visible sternite make this species highly
Material examined, in addition to types: GUADELOUPE: 12-Guadeloupe, Dufau
[MNHN]. DOMINICA: 3-5 mi E. Dublanc, 16 VIII 1986, C. W. & L. B. O'Brien
[MAIC]. 1-Pt. Casse, ca 1500', 14 VIII 1986, C. W. & L. B. O'Brien [MAIC]. 1-3 mi
NE Pt. Casse, June. Rosalie & Castle Bruce Rd., 18 VIII 1986, C. W. & L. B. O'Brien
[MAIC]. 1-5 mi E. Dublanc, 1250', 20 VIII 1986, C. W. & L. B. O'Brien [MAIC]. 1-6
mi E. Dublanc, 1250', 20 VIII 1986, C. W. & L. B. O'Brien [MAIC].
Ivie & Slipinski: West Indian Pycnomerini 79
We would like to thank the following persons for their help with this project.
Curators who loaned material in their care: L. Herman (AMNH), N. Berti (MNHN),
J. M. Kingsolver, U.S.D.A. (NMNH), H. Franz (HFCM), N. Penny and D. H.
Kavanaugh (CASC), D. Webb (INHS), A. Newton, S. Shaw and M. Thayer (MCZC),
M. C. Thomas (FSCA), R. S. Miller (RSMC), and A. Smetana (CNCI). Valuable mate-
rial was donated to the project by B. D. Valentine (The Ohio State University) and C.
W. and L. B. O'Brien (Florida A & M University). The Spanish resumn and a very
careful review were done by S. M. Clark, Montana State University. The manuscript
was further reviewed by D. Bartell, K. Stephan, M. C. Thomas and H. V. Weems, Jr.
A special thanks to G. Waxmonsky, Attache for Scientific Affairs of the U. S. Embassy
in Warsaw for logistic help. Field work by MAI in Puerto Rico and the Dominican
Republic was funded by a grant from MONTS-NSF. Extensive assistance with field
work by MAI in the Dominican Republic was provided by J. and M. Brodzinsky (Santo
Domingo); L. Dominquez, R. O. Rimoli, and A. Jimenez (MHND); and T. K. Philips
and K. A. Johnson (Montana State University). Our collaboration was funded by the
U. S. National Academy of Sciences-Polska Academia Nauk Interacademy Exchange
Program. This is contribution J-2239 of the Montana Agricultural Experiment Station.
BLACKWELDER, R. E. 1945. Checklist of the coleopterous insects of Mexico, Central
America, the West Indies, and South America. Bull. United States Nat. Mus.
185(3): i-iv + 343-550.
CHAMPION, G. C. 1898. A list of the clavicorn Coleoptera of St. Vincent, Grenada,
and the Grenadines. Trans. R. Entomnol. Soc. London 1898: 393-412.
DAJOZ, R. 1977. Coleopteres Colydiidae et Anommatidae Pal6arctiques. Faune de
l'Europe et du Bassin M6diterran6en 8: 1-275.
ERICHSON, W. F. 1845. Naturgeschichte der Insecten Deutschlands. I. Coleoptera 3
GROUVELLE, A. H. 1902. Voyage de M. le Dr. Ed. Bugnion au V6nezuela, en Colom-
bie et aux Antilles. Col6opteres clavicornes. II. Nitidulidae, Colydiidae,
Cucujidae, Cryptophagidae, Tritomidae et Dryopidae. Ann. Soc. Entomol.
France 71: 461-467.
GROUVELLE, A. H. 1908. Coleopteres de la region indienne. Rhysodidae,
Trogositidae, Nitidulidae, Colydiidae, Cucujidae (lermemoire). Ann. Soc. En-
tomol. France 78: 315-495.
GROUVELLE, A. H. 1913. in A. H. Grouvelle and A. Raffray. Supplement des collop-
teres de la Guadeloupe (3' supplement). Ann. Soc. Entomol. France  81:
HETSCHKO, A. 1930. Colydiidae. in S. Schenkling, Coleopterorum Catalogus 107:
IVIE, M. A., AND S. A. SLIPINSKI. (in press a). Cataloge of the genera of the Colydi-
idae (Coleoptera). Polskie Pismo Entomol.
IVIE, M. A., AND S. A. SLIPINSKI. (in press b). Key to the genera of New World
Colydiidae. Coleopt. Bull.
LAWRENCE, J. F. 1980. A new genus of Indo-Australian Gemphylodini with notes on
the constitution of the Colydiidae (Coleoptera). J. Australian Entomol. Soc. 19:
LENG, C. W., AND A. J. MUTCHLER. 1914. A preliminary list of the Coleoptera of
the West Indies as recorded to January 1, 1914. Bull. American Mus. Nat. Hist.
NIKITSKY, N. B., AND V. V. BELOV. 1980. [Larvae of cylindrical bark-beetles (Col-
eoptera, Colydiidae) of the European part of the U. S. S. R. and Caucasus, with
Florida Entomologist 72(1)
comments on the taxonomy of the family. 2.] Zoologitscheskiy Zyrnal [also as
Zoologicheskii Zhurnal] 59 (9): 1328-1333. [in Russian with English summary].
REITTER, E. 1877. Beitrag zur Kenntniss der Colydier. Entomol. Zeit. Stettin 38(1-
REITTER, E. 1878. Neue Colydiidae des Berliner Museums. Deutsche Entomologische
Zeitschrift 22 (1): 113-125.
SHARP, D. 1894. Colydiidae. Biologia Centrali-Americana, Coleoptera II (1): 443-498.
SLIPINSKI, S. A. 1984. Studies on the African Colydiidae (Coleoptera). Part II. Gen-
era: Afrorthocerus Pope and Pycnomerus Erichson. Ann. Zool. (Warsaw) 38 (6):
SLIPIrSKI, S. A. (in press). A monograph of the World Cerylonidae (Coleoptera:
Cucujoidea). Part I-Introduction and higher Classification. Ann. Mus. Civ.
Stor. Nat. Genova.
SLIPINSKI, S. A., AND B. BURAKOWSKI. 1988. A review of the genus Rhopalocerus
W. Redtenbacher of the World (Coleoptera, Colydiidae). Ann. Zool. (Warsaw)
42 (2): 75-118.
WOLCOTT, G. N. 1951. Coleoptera. Insects of Puerto Rico. J. Agric. Univ. Puerto
Rico  32: 225-416.
FLASH BEHAVIOR AND ECOLOGY OF THAI LUCIOLA
FIREFLIES (COLEOPTERA: LAMPYRIDAE)
JAMES E. LLOYD AND STEVEN R. WING
Department of Entomology and Nematology
University of Florida
Gainesville, FL 32611-0143, USA
Pesticide Research Laboratory
Department of Agriculture
Bangkhen, Bangkok 9, Thailand
The advertising flashes and general ecology of four species of Luciola fireflies from
the Bangkok, tidal region of Thailand are reported, and comparisons made with other
species. The general signal systems and aquatic life-histories found fit with what has
been reported for other Asian and Australian Luciolinae.
Se report sobre los destellos de anuncio y de la ecologia general de cuatro species
de Luciola en la region donde hay cambios de mareas en Bangkok, Tailandia, ye se
comparan con otras species. El sistema general de sefales e histories de la vida acuAtica
se ha encontrado que concuerdan bien con lo que se ha reportado de otros Luciolinos
de Asia y Australia.
Thailand fireflies are known almost exclusively for the synchronous flashing behavior
of a single, "superstar" species, Pteroptyx malaccae Gorham, which, at least before its
Lloyd et al.: Ecology & Behavior of Luciola
habitat had been extensively altered, occurred in aggregations of thousands on man-
groves along estuaries and in inland swamps within the tidal reaches of large rivers.
However, Thailand has several other species of Lampyridae, some of which we saw
during field work in the vicinity of Bangkok (Lloyd & Wing 1981, Wing et al. 1983,
Lloyd et al. 1989). We record here observations made on Luciola (Luciolinae) species
from 19 July to 13 August 1980.
Luciola japonica (Thunb.) was the most commonly seen firefly during our study,
and was always associated with standing water. It occurred around and over ponds,
flooded grasslands and rice paddies, and along roadside ditches and canals. Luminescing
adults frequently were seen landing and perching on, and taking off from emergent
vegetation in these habitats. The larvae probably are aquatic, an ecology known for
several other Asian Luciolinae (e.g., Luciola cruciata Motsch., Yuma 1984). Occasion-
ally larval glows could be seen on and under standing water, and lampyrid larvae were
collected there. In captivity, these fed on snails from their habitats, including Hebledia
crosseana and siamensis; Lymnaea (= Radix) auriculasia and rubiginosa; and Filop-
ludina spp. Salinity at one road-side ditch site was 13.2 ppt, showing a tidal influence
even two miles inland from a large river and at a point several miles up river from the
Gulf (LaMotte test kit, 0.2-.40 ppt).
Males began flying and flashing 33-37 min after sunset (n = 2 evenings), that is,
1.5-1.7 crep units after sunset (Nielson 1961). The number of active males gradually
increased for the next hour or so, was reduced considerably by 3 hours after sunset
(ss + 3 hrs), and by ss +4, only occasional individuals were seen. Males flew 1-2 meters
above the vegetation in their sites, emitting long trains of short, bimodal flashes (Fig.
1). The first peak in each flash was much less intense than the second, and was somewhat
variable in form (Fig. 2a-d); the variation seen in recordings is probably partially due
to the movement (orientation) of flying males relative to the PM-recorder (photo-multip-
lier recording methods are described in Lloyd 1973). The duration of 87 recorded flashes
of 9 males ranged from 160 to 180 mSec (26.1 C); height ratio (ratio of relative intensity
of first peak to that of second peak) ranged 0.09 to .34; and the modulation rate of the
two peaks ranged from 12.5 to 17.9 Hertz (Table 1). Note that with respect to these
last two parameters, there was less variability than might be concluded by the range
extremes. Table 1 gives values for the flashes of five males as examples.
Flash trains were of three forms: (1) bright flashes were emitted at a fast rate (Fig.
1, center of trace; e.g., 2.5 Hertz at 26.1C); (2) flashes were emitted at half the fast
rate; (3) flashes were emitted at the fast rate, but alternate flashes were much reduced
in intensity (Fig. 1, beginning and end of trace shown) in a manner reminiscent of that
noted in Luciola huonensis Ballantyne in New Guinea (Lloyd 1973). During late even-
TABLE 1. DATA FOR THE FLASHES OF FIVE LUCIOLA JAPONICA MALES; FLASHES
WERE RECORDED WITH PHOTO-MULTIPLIER SYSTEM AT 26.10C. DURA-
TIONS IN MILLISEC AND MODULATION FREQUENCIES IN HERTZ. MEANS
AND STANDARD DEVIATIONS GIVEN. DATA ARE FOR FLASH DURATION
AT BASE AND AT HALF MAXIMUM INTENSITY, RELATIVE INTENSITY OF
THE TWO PEAKS, RATE OF THE TWO PEAKS, AND NUMBER OF FLASHES
MEASURED FOR EACH MALE.
Male Base Dur. Dur. Max/2 Mod. Ratio Mod. Freq. N
1 170 10 60 00 .23 .05 14.5 1.3 15
3 160 20 50 00 .16 .05 16.7 1.4 11
11 170 10 50 10 .13 .07 14.2 1 15
13 170 10 60 10 .11 .06 14.9 .9 11
14 170 10 60 10 .12 .03 14.9 .7 9
Florida Entomologist 72(1)
Fig. 1-6. Chart traces of firefly flashes. Horizontal axis, time as indicated by time-
lines; vertical axis, relative intensity. Flashes were recorded at 26.1C. (1-3) L.
japonica: (1) Train of male flashes, fast rate, but see text. (2) Sample of variation in
bimodal flashes. (3) Flicker of landing male. (4-5) Flashes of L. cinqulata: (4) Male flash
pattern (advertising flash). (5) Two sections of a sputtery flicker emitted by a male after
he received a response to his advertising flash. (6) Simulated flash pattern of male L.
nr carinata. Actually only the second flash was recorded; for illustration, a first flash,
based on the profile of the recorded second flash, was sketched in position, with an
appropriate adjustment made in the falling (AFC) baseline to connect the two.
ing, the slow rate seemed to be more common than the fast, possibly relating to lower
male density, that is, reduced male competition. Table 2 gives values in PM-recordings
of the flash periods of six males as examples.
Lloyd et al.: Ecology & Behavior of Luciola
TABLE 2. FLASH PERIODS OF SIX MALES, ONE AT HALF RATE, PHOTOMULTI-
PLIER-RECORDED AT 26.1C. MEAN INTERVAL AND STANDARD DEVIA-
TION IN SEC; N IS NUMBER OF INTERVALS MEASURED.
Male Interval N
3 .38 .01 30
8 .39 .01 30
13 .40 .01 22
14 .39 .01 20
17 .39 .01 17
22A .78 (half) .01 4
Males flying within 10 meters of each other sometimes synchronized their flashes.
Their synchrony appeared to be as precise as occurs in Pteroptyx malaccae, rather than
imprecise, as seen in certain other fireflies (e.g., in Photinus carolinus Green, in the
Appalachian Mountains of eastern North America, Lloyd 1966, unpub. obs. J. Sivinski,
JEL, SRW). Synchrony was noted at both fast and slow rates. Landing males, and
males approaching and hovering near attractive sources of light on the substrate, emit-
ted flickers (Fig. 3) or apparent glows (not electronically recorded) that may actually
have been subliminally (for human eyes) modulated. Landing and flying females also
emitted glows and flickers.
Numerous apparent conspecific specimens in two Thai collections (Acknowledge-
ments) suggest that adults of this species are active throughout the year, and that the
species is widely distributed in Thailand.
Luciola cinqulata E. Oliver (tentative identification) was seen flying over flooded
grassland and rice paddies. It was never observed before ss + 147 min, and was noted
as late as ss + 252 min. Males flew 3-9 meters above the ground, and emitted one-third
sec long, flare-like flashes at 3.5 sec intervals (s = 0.29, range = 2.9-3.9 sec, n 8,
stopwatch; 25.8-27.2C). Two PM-recordings of advertising flashes reveal a rather sharp
ON transient (Fig. 4), which was not suspected from field observations. In the field,
the flashes are reminiscent of those of Florida's Pyractomena ecostata (LeConte), a
species also found in open, wet grassland (JEL unpub. obs.).
Males could be attracted toward a half-sec flash of a penlight emitted immediately
after each of their flashes, from as far away as 45 meters (n = 12 of 12 males). When
they had approached to 1-4 meters, they began to emit a short "sputtery" flicker after
the penlight flash (Fig. 5). This behavior was previously noted in males of four species
of New Guinea Pteroptyx (Lloyd 1973). Males would not approach the flash of the
penlight further unless it was left ON (n = 4 of 4 males), suggesting that responding
females may glow between their response flashes, or perhaps begin glowing during the
final moments of a male's approach. Males landed within a few inches of the glowing
decoy. This experiment was first suggested by the approach of a male to the orange
glow of the light-emitting diodes of an electronic stopwatch.
Numerous possibly conspecific specimens in Thai collections were labeled L. chinen-
sis and L. substriata, and suggest that adults of this species are active throughout the
year and that the species occurs throughout Thailand.
Luciola nr carinata: Three males of this species, judging from flash pattern, were
seen flying with L. cinqulata. They flew slowly 2-6 meters above the ground, and
emitted a two-flash pattern at 3 sec intervals. The flashes of a pair were spaced at an
estimated 0.6-1.0 sec interval, and appeared to be without sharp intensity transients.
However, the single PM-recording shows an asymmetrical, 620 mSec flash whose rise
time is only 60 percent as long as its decay (Fig. 6). The recorded male was attracted
84 Florida Entomologist 72(1) March, 1989
briefly toward a short penlight flash that was emitted immediately after the second
flash of his pattern.
Luciola circurndata Motsch.: Although collections indicate that this species occurs
throughout the year, and throughout Thailand, we saw few (<5). Males emitted single,
short flashes in continuous trains, while flying 2-5 meters above the ground. Flash
period was estimated to be about 0.7 sec, and flash duration about 200 mSec. The
general appearance of flashing is similar to that of Photuris frontalis LeConte in eastern
North America, and Luciola salomonis limbatipennis Pic on Guadalcanal (Lloyd 1973
and unpub.), though the former's flash is much shorter. These fireflies irregularly inter-
rupt their trains by omitting single or several flashes.
Behavior voucher specimens (Lloyd 1966) are in the JEL collection, and will eventu-
ally be deposited in the Florida State Collection of Arthropods.
Observations on the sexual signals of Asian Luciolinae, in particular those made in
Japan over the past decade by Nobuyoshi Ohba (1983 and refs.), along with the excellent
and comprehensive taxonomic work on the group by Lesley Ballantyne (1968, 1987 and
in prep.), will permit questions to be asked of the flashing data that are of broad
ecological and higher systematic significance, to complement our current use of flashes
for species-level characterization and taxonomy. Ohba (1983) has made progress along
these lines, addressing, for example, signaling systems, seasonal and geographic occurr-
ence, and "reproductive isolation." Future comparisons and analyses of differences be-
tween the signal systems of Asian and New World fireflies, the latter having the high-
impact influence of Photuris spp. predation by signal-targeting involving both aerial
attack and attractive signal mimicry, and the former often involving considerable mate
choice opportunity (Lloyd 1984), should reveal important facts about signal evolution
(see Burk 1982).
We thank: Drs. Prayoon Deema and Nuan Chan for their considerable assistance
during our study, for without it, little if anything could have been accomplished; the
curator of the collection at Chulalongkorn University, Prof. Siriwat Wongseri, and at
the National Collection, Dept. of Agriculture, at Karsetsart University; Lesley Ballan-
tyne of the Riverina School of Agriculture, Wagga Wagga, Australia, for identifying
our specimens; the National Geographic Society for financial support through Grant No.
221680; John Sivinski and Joe Cicero for commenting on the manuscript; and Barbara
Hollien for technical preparation of the mansucript. Florida Agric. Exp. Sta. Journal
Series No. 9273.
BALLANTYNE, L. A. 1968. Revisional studies of Australian and Indomalayan
Luciolini (Coleoptera: Lampyridae: Luciolinae). Univ. Queensland Papers. 2:
1987. Lucioline morphology, taxonomy and behavior: a reappraisal (Coleopt-
era, Lampyridae). Trans. American Entomol. Soc. 113: 171-188.
BURK, T. 1982. Evolutionary significance of predation on sexually signaling males.
Florida Entomol. 65: 90-104.
LLOYD, J. E. 1966. Studies on the flash communication system in Photinus fireflies.
Univ. Michigan Mus. Zool. Misc. Pub. No. 130. 95 p.
Chan & Linley: Immature Atrichopogon Development
1973. Fireflies of Melanesia: bioluminescence, mating behavior, and synchron-
ous flashing (Coleoptera: Lampyridae). Environ. Entomol. 2: 991-1008.
1984. On deception, a way of all flesh, and firefly signaling and systematics.
Pages 48-84 in R. Dawkins and M. Ridley, Eds. Oxford Surveys in Evolutionary
Biology 1. Oxford.
- AND S. R. WING. 1981. Photo story (copulation clamp). Florida Entomol. 64:
-- S. R. WING, AND T. HONTRAKUL. 1989. Ecology, flashes and behavior of
congregating Thai fireflies. Biotropica. (in press)
NIELSON, E. T. 1961. Twilight and the "crep" unit. Nature 190: 878-879.
OHBA, N. 1983. Studies on the communication system of Japanese fireflies. Sci. Rpt.
Yokosuka City Mus., No. 30. 62 p.
WING, S. R., J. E. LLOYD, AND T. HONGTRAKUL. 1983. Male competition in Pterop-
tyx fireflies: wing-cover clamps, female anatomy, and mating plugs. Florida En-
tomol. 66: 86-91.
YUMA, M. 1984. Egg size and viability of the firefly, Luciola cruciata (Coleoptera,
Lampyridae). Kontyu 52: 615-629.
LABORATORY STUDIES ON THE IMMATURE STAGES OF
ATRICHOPOGON WIRTHI (DIPTERA: CERATOPOGONIDAE)
KAI LOK CHAN
Department of Zoology
National University of Singapore
Kent Ridge, Singapore 0511
JOHN R. LINLEY
Florida Medical Entomology Laboratory
University of Florida
200 9th Street S.E.
Vero Beach, FL 32962
In the laboratory, at 24-260C and 90 + % relative humidity, the immature stages of
Atrichopogon wirthi Chan & Linley (Diptera: Ceratopogonidae) were completed in 2-3
weeks. The mean durations (days) of the individual stages were, respectively: egg 3.3,
1st instar 1.7, 2nd instar 2.0, 3rd instar 2.3, 4th instar 4.1, pupa 3.0. The mean overall
duration was 16.4 days.
Se completaron en 2-3 semanas en el laboratorio a 24-26C y 90 + % humedad relative,
las etapas immaduras de Atrichopogon wirthi Chan y Linley (Diptera:
Ceratopogonidae). El average de la duraci6n (dias) de las etapas individuals fueron,
respectivamente: huevo 3.3, ler estadio 1.7, 2nd estadio 2.0, 3er estadio 2.3, 4to estadio
4.1, y pupa 3.0. El average total de la duraci6n fue de 16.4 dias.
86 Florida Entomologist 72(1) March, 1989
We recently described all stages of Atrichopogon wirthi Chan & Linley from speci-
mens collected from leaves of the water lettuce, Pistia stratiotes L. (Chan & Linley
1988). Sufficient numbers of the immature stages were obtained to allow the complete
immature life cycle to be studied in the laboratory. In the literature, comparable infor-
mation exists for only two other Atrichopogon species, A. geminus Boesel (as A. levis)
(Boesel & Snyder 1944) and A. jacobsoni (de Meijere) (Drake 1971). In view of this very
meager information on the immature biology of the genus, we collected the data here
MATERIALS AND METHODS
Pistia plants that yielded immatures of A. wirthi were collected from Chinese Farm,
a disused aquaculture project adjacent to Old Dixie Highway, about 5 km south of the
Florida Medical Entomology Laboratory, Vero Beach, Indian River County, Florida.
All rearing was done in a laboratory maintained at 24-260C, on cut pieces of Pistia
leaf placed in petri dishes lined on the bottom with damp paper towel to ensure high
humidity (90 + % RH). Potential predators, such as dragonfly and beetle larvae, mites
and spiders were removed from the leaf surfaces to ensure survival of the Atrichopogon
immatures. Eggs were transferred to dishes on the pieces of leaf to which they had
been attached by ovipositing females. In all, 82 eggs from 18 individual egg batches
were hatched and reared. In addition, larvae in various stages were also collected.
These were occasionally left on their original leaves, but more usually were transferred
to other pieces selected specifically for the presence of an abundant microflora diatomss,
algae, bacteria, fungi), which constitutes the larval food. Every 3-4 days, additional food
material was scraped from freshly collected leaves and added to the cultures. The
natural food was also supplemented by a 1:1 lactalbumin/brewer's yeast mixture sprink-
led very sparingly on the leaf surfaces and replenished every 3-4 days.
Each petri dish containing immatures was examined twice daily, from 0800-1100 hr
and again from 1500-1700 hr. Careful notes were kept of developmental progress and
behavior, and checks made to ensure the presence of adequate food. Only specimens
that successfully completed each stage were included in the final analysis of data. Abnor-
mal or moribund larvae were discarded.
The mean durations (days) of the individual immature stages (ranges in parentheses)
and of the entire immature developmental period were: egg 3.3 (3-4), 1st larva 1.7 (1-3),
2nd larva 2.0 (1-4), 3rd larva 2.3 (1-4), 4th larva 4.1 (2-7), pupa 3.0 (2.5-4), entire 16.4
(14-20). At 3.3 days, the mean duration of the egg stage was relatively long. After
hatching, the durations of the larval stages increased progressively through the instars,
with the pupal stage lasting 3.0 days, on average, and the entire immature life cycle
requiring 16.4 days under the stipulated laboratory conditions.
The distributions of the developmental periods for each stage are shown in Fig. 1
as the percentages of individuals that daily completed that stage. Hatching of the eggs
was distributed over 2 days, most on day 3, the remainder on day 4. Through the larval
stages, as the time required to complete each stage became progressively longer, the
distributions were spread over more days (respectively 3,4,4 and 6), probably because
of differences in the rate and efficiency of feeding among individuals. As would be
expected in a non-feeding stage, developmental times among pupae were much more
closely grouped (Fig. 1).
* E .
1 2 3 4 5 6 7
Fig. 1. Distributions of times required by A. wirthi to complete each of its immature
Chan & Linley: Immature Atrichopogon Development
Florida Entomologist 72(1)
Although the laboratory environment provided in these studies was certainly more
constant than conditions in the field, the mean overall development period of 16.4 days
is in close agreement with estimates from field-collected plants. Chan and Linley (in
press) recorded the distribution of immature A. wirthi on all the leaves taken from
individual Pistia plants and, from estimates of leaf age, deduced that the immature life
cycle ofA. wirthi probably required 12-18 days. Atrichopogon geminus took 12-13 days
under summer laboratory conditions in Ohio (Boesel & Snyder 1944) and A. jacobsoni
required 24-27 days at 200 C (Drake 1971). Similar immature life cycles are found in the
genus Forcipomyia of the same subfamily (Forcipomyiinae). Forcipmyia (Dacnofor-
cipomyia) anabaenae Chan & Saunders from Singapore took 12-27 days to complete
development, averaging 19 days under laboratory conditions (Chan & Saunders 1965).
Forcipmyia (Lasiohelea) taiwana (Shiraki) from Taiwan required 16-18 days to com-
plete the larval stages and 3-5 days for the pupal stage, for an overall oviposition to
adult emergence period of 21-26 days (Sun 1965). All these periods are brief and indicate
that development is generally relatively rapid in the Forcipomyiinae as compared to
other genera in other subfamilies.
The rapid growth of A. wirthi larvae indicates that their food is abundant on the
Pistia leaves. Many larvae were seen feeding on the leaf surfaces, almost invariably in
areas covered with algae, diatoms, fungi and bacteria. Dissections of the gut contents
of several larvae indicated that these organisms were ingested. There appeared to be
a particular association between A. wirthi and a pyralid caterpillar (and perhaps also a
small leafhopper), both of which feed on the Pistia leaves. Atrichopogon wirthi eggs
were not seen on leaves devoid of lepidopteran larvae and were often deposited in close
association with fecal pellets left by the caterpillars, or near holes and tracks made by
them in the leaf surface. Conceivably, gravid female A. wirthi oviposit selectively where
nearby fecal material and its associated bacteria will ensure a source of food for the
newly hatched larvae.
We thank D. Duzak for assistance in collecting water lettuce plants from the field
site. This paper is Institute of Food and Agricultural Sciences, University of Florida
Experiment Stations Journal Series No. 9375.
BOESEL, M. W., AND E. G. SNYDER. 1944. Observations on the early stages and life
history of the grass punky, Atrichopogon levis (Coquillett) (Diptera: Heleidae).
Ann. Entomol. Soc. Am. 37: 37-46.
CHAN, K. L., AND J. R. LINLEY. 1988. Description of Atrichopogon wirthi new
species (Diptera: Ceratopogonidae) from leaves of the water lettuce (Pistia
stratiotes) in Florida. Florida Entomol. 71: 186-201.
CHAN, K. L., AND J. R. LINLEY. The distribution of immature Atrichopogon wirthi
(Diptera: Ceratopogonidae) on leaves of the water lettuce, Pistia stratiotes. En-
viron. Entomol. (In press.)
CHAN, K. L., AND L. G. SAUNDERS. 1965. Forcipmyia (Dacnoforcipomyia)
anabaenae, a new bloodsucking midge from Singapore, described in all stages
(Diptera, Ceratopogonidae). Canadian J. Zool. 43: 527-540.
DRAKE, E. F. 1971. Life cycle and laboratory diet for Atrichopogon jacobsoni (de
Meijere) (Diptera: Ceratopogonidae). Proc. Hawaiian Entomol. Soc. 21: 63-66.
SUN, W. K. C. 1965. Study of a biting midge, Forcipomyia (Lasiohelea) taiwana
(Shiraki) (Diptera: Ceratopogonidae) 1. Description of the complete life cycle of
the midge reared in the laboratory. J. Formosan Med. Assoc. 64: 76-84.
Habeck et al.: New U.S. Nitidulid
COLOPTERUS POSTICUS (ERICHSON), A
NITIDULID BEETLE NEW TO THE UNITED STATES
D. H. HABECK
University of Florida, IFAS
Gainesville, FL 32611
J. NAGEL AND J. E. PENA
IFAS, Univ. of Florida Tropical Res. & Educ. Center
Homestead, FL 33031
C. posticus (Erichson) is recorded from the United States for the first time. It has
been collected from various fruits and flowers in Homestead, Florida.
Colopterus posticus (Erichson) se encontr6 en los Estados Unidos por primera vez.
Se ha recolectado en varias flores y frutas en Homestead, Florida.
An unfamiliar nitidulid beetle was collected on fallen loquat (Eriobotrya japonica
Lindl) fruit in April, 1987 by J. Nagel of the University of Florida Tropical Research
and Education Center, Homestead, FL. It was later identified (DHH) as Colopterus
posticus (Erichson), a species not known to be established in the United States. Sub-
sequently 7 additional specimens were collected at the research center on fallen loquat
fruit and one on atemoya (Annona squamosa x A. cherimola) in April and 16 specimens
on jack fruit (Artocarpus integrifolia Forst) on June 6, 1987. Although other fruits
(mango, many varieties of citrus, Barbados and Surinam cherries (Malpighia sp.), za-
pote (Calocarpum sapota Merr.) and other tropical species) were growing nearby, no
C. posticus have been found on them. In June, 5 more specimens were collected from
old flowers of (Philodendron selloum C. Koch) at the research center and another
specimen was collected by R. M. Baranowski from old flowers of torch ginger (Nicolaia
elatior (Jack) Horan) from a site 2 miles away. In March, 1988 another specimen was
collected indicating that the species survived the winter and is established.
Colopterus posticus is easily recognized by the orange body with the apical half of
the elytra black. No other North American species has similar coloration. All male fore
tibiae are distinctly shaped (Fig. 1) although there is some variation in the angle and
depth of the groove. The purpose of the odd-shaped tibiae is not known, but they may
be used to grasp and hold the female during copulation. Females have normal fore tibiae.
Colopterus posticus is a tropical New World species ranging from Mexico to Brazil
and Peru. The list of host associations is long, but it apparently is of little or no economic
importance. It has been found on rotting oranges, mangos, papayas, cacao pods,
cherimoya, bananas, lemons, grapefruit, pineapple, guava, avocado, limes, maize and
legume pods as well as flowers of cactus (Cereus sp.) Philodendron, torch ginger, and
Enterolobium. It has been intercepted at many U.S. ports on a variety of produce
including Monstera deliciosa seeds, paprika seeds, cut flowers, Alocasia roots, papaya
fruit, and bromeliads or orchids. These records are based on specimens in the collection
of Lorin Gillogly, the Field Museum of Natural History in Chicago, and the D. H.
Florida Entomologist 72(1)
Fig. 1. Fore tarsus of C. posticus male.
The arrival of C. posticus increases the described species of Colopterus in the U.S.
to 10. Most of the species are fairly widespread but C. posticus is currently known only
from the Homestead area of Florida.
Another tropical species, Colopterus amputatus (Erichson) (identified by L. Watr-
ous) was established in Florida at least temporarily. Colopterus amputatus, widespread
throughout the new world tropics, was collected March 7-8, 1919 on Marathon Key,
Florida by E. A. Schwarz. To our knowledge these six adults and a nearly mature larva
collected from a gumbo limbo stump, are the only specimens ever collected in the United
The assistance of Dr. R. M. Baranowski (collecting) and Dr. K. Langdon (plant
names) is appreciated. Mr. Lorin Gillogly kindly gave permission to use data from his
private collection. Mr. Don Weisman arranged the loan of C. amputatus from the U.S.
National Museum. Florida Agric. Exp. Sta. J. Series No. 9328.
'*.. ^ _
Deyrup et al.: Ants of Florida 91
A PRELIMINARY LIST OF THE ANTS OF FLORIDA
Archbold Biological Station
P. O. Box 2057
Lake Placid, Florida 33852
Department of Zoology
University of Florida
Gainesville, Florida 32611
GEORGE C. WHEELER AND JEANETTE WHEELER
3358 NE 58th Avenue
Silver Springs, Florida 32688
Research Associate, Florida State Collection of Arthropods
Division of Plant Industry
Florida Department of Agriculture and Consumer Services
Gainesville Florida 32602
A list is given of ants known to occur in Florida. Except for widespread species the
known distribution by counties is recorded. Species which require further study are
listed in an appendix. This list is preliminary to a book which will treat fully the ant
fauna of the state.
Se present una lista de las de las hormigas sabidas a hallarse en Florida. Sin las
species extensamente distribuidas, se ofrece la distribuci6n conocida por condado. Las
species requiriend6s studio adicional se presentaqn en un apendice. Esta lista es
preliminary i un libro sobre la fauna de las hormigas del Estado.
This study presents an up-to-date list of Florida ants and their general distribution
within the state. The list is a preliminary step by the authors toward a book-length
coverage of identification, taxonomy, and natural history. A comprehensive review of
the Florida ant fauna has been slow in developing. The geographic location has produced
a rich fauna where north temperate, subtropical, endemic, and exotic species co-exist.
Many species are rare, secretive, or localized in distribution.
This list is based on (1) all trustworthy records discovered in the literature, (2)
specimens in the collection of the Archbold Biological Station and the Florida State
Collection of Arthropods (FSCA), and 3) specimens in the authors' collections.
Subfamilies and tribes are arranged according to Wheeler and Wheeler 1985. Genera
follow Smith's catalog (1979), with a few exceptions. Species are arranged alphabetically
under the genus. Under each species we have listed alphabetically the counties in which
that species has been collected. If the number of counties exceeds 25, they are not
listed, but we have given the number followed by "Widely distributed." In our book we
will add localities within counties and cite records in the literature.
Appended is a list of species reported for Florida, but concerning which we have
doubts. These may include dubious published records, species in unpublished studies,
or species reported as new but not yet published.
Florida Entomologist 72(1)
De Soto E-7
Indian River D-14
Palm Beach E-13
St. Johns C-3
St. Lucie E-5
Santa Rosa A-2
Fig. 1. Diagrammatic map of Florida to facilitate location of counties.
Hopefully myrmecologists and other entomologists will bring to our attention over-
sights, errors, and unpublished records that doubtless are to be found in many collec-
THE ANTS OF FLORIDA
Genus AMBLYOPONE Erichson
pallipes (Haldeman). Alachua, Brevard, Clay, Columbia, Dixie, Flagler, Highlands,
Jackson, Levy, Liberty, Madison, Marion, Monroe, Pinellas, Polk, Putnam, St.
Johns, Suwanee, Volusia, Wakulla.
Deyrup et al.: Ants of Florida
Genus PRIONOPELTA Mayr
antillana Forel. Marion.
Genus PLATYTHYREA Roger
punctata (F. Smith). Broward, Collier, Dade, Highlands, Indian River, Martin, Monroe,
Genus GNAMPTOGENYS Roger
aculeaticoxae (Santschi. Dade.
Genus PROCERATIUM Roger
croceum (Roger). Highlands, Jefferson, Putnam, Volusia.
pergandei (Emery). Alachua, Baker, Citrus, Columbia, Jefferson, Leon, Marion, Nas-
silaceum Roger. Alachua, Calhoun, Gadsden, Putnam.
Genus DISCOTHYREA Roger
testacea Roger. Alachua, Broward, Citrus, Columbia, Flagler, Highlands, Jefferson,
Lake, Leon, Liberty, Marion, Martin, Monroe, Nassau, Putnam.
Genus PACHYCONDYLA F. Smith
stigma (Fabricius). Broward, Collier, Dade, Highlands, Lee, Martin, Monroe, Palm
Genus CRYPTOPONE Emery
gilva (Roger). Alachua, Columbia, Jackson, Jefferson, Leon, Levy, Marion, Putnam.
Genus PONERA Latreille
exotica Smith. Columbia, Flagler, Highlands, Putnam, St. Johns, Wakulla.
pennsylvanica Buckley. Alachua, Columbia, Jefferson, Lake, Marion, Santa Rosa.
Genus HYPOPONERA Santschi
inexorata (Wheeler). Dade, Highlands, Indian River, Marion, Monroe, Palm Beach.
opaciceps (Mayr). 34 counties. Widely distributed.
opacior (Forel). 54 counties. Widely distributed.
punctatissima (Roger). Alachua, Broward, Collier, Dade, De Soto, Glades, Highlands,
Jefferson, Levy, Liberty, Manatee, Monroe, Palm Beach, Pinellas, Polk, Put-
nam, Sarasota, Suwannee, Volusia.
Genus LEPTOGENYS Roger
elongata (Buckley). Alachua, Baker, Columbia, Dade, Dixie, Gadsden, Highlands, In-
dian River, Levy, Marion, Monroe, Putnam, Volusia.
Florida Entomologist 72(1)
Genus ANOCHETUS Mayr
mayri Emery. Dade.
Genus ODONTOMACHUS Latreille
brunneus (Wheeler). 27 counties. Widely distributed.
clarus Roger. Highlands, Polk.
ruginodis Wheeler. Broward, Collier, Dade, Glades, Highlands, Hillsborough, Indian
River, Martin, Monroe, Pinellas, Sarasota, St. Lucie.
Genus NEIVAMYRMEX Borgmeier
carolinensis (Emery). Alachua, Lake, Leon, Levy, Madison.
opacithorax (Emery). Alachua, Charlotte, Dade, Highlands, Lake, Marion, Monroe, St.
texanus (Emery). Alachua, Clay, Leon, Polk, Santa Rosa, Volusia, Wakulla.
Genus PSEUDOMYRMEX Lund
cubaensis Forel. Collier, Dade, Highlands, Lake, Manatee, Martin, Monroe, Osceola,
Palm Beach, Polk, Sarasota.
ejectus F. Smith. Alachua, Broward, Collier, Dade, Hardee, Hernando, Highlands,
Hillsborough, Indian River, Leon, Levy, Marion, Monroe, Orange, Palm Beach,
Pinellas, Polk, Sarasota, Sumter, Taylor, Volusia.
elongatus Mayr. Broward, Collier, Dade, Highlands, Lee, Monroe, Pinellas, Sarasota.
leptosus Ward. Alachua, Monroe.
mexicanus Roger. Alachua, Broward, Collier, Dade, De Soto, Duval, Glades, Hardee,
Highlands, Manatee, Marion, Martin, Monroe, Osceola, Palm Beach, Pasco,
Pinellas, Polk, Putnam, St. Johns, St. Lucie, Sarasota, Sumter.
pallidus (F. Smith). Alachua, Collier, Duval, Glades, Highlands, Leon, Monroe, Polk,
seminole Ward. Alachua, Dade, De Soto, Highlands, Indian River, Marion, Monroe,
Pinellas, Polk, Sarasota.
simplex F. Smith. Charlotte, Collier, Dade, Highlands, Monroe, Pinellas, Polk,
Genus Pogonomyrmex Mayr
badius (Latreille). 29 counties. Widely distributed.
Genus APHAENOGASTER Mayr
ashmeadi Emery. Alachua, Clay, Columbia, Franklin, Hamilton, Hernando, Highlands,
Leon, Levy, Marion, Okaloosa, Pasco, Polk, Putnam, St. Johns, Suwannee,
carolinensis Wheeler. 37 counties. Widely distributed.
flemingi Smith. Alachua, Citrus, Columbia, Dade, Highlands, Monroe, Putnam, St.
Deyrup et al.: Ants of Florida
floridana Smith. Alachua, Clay, Escambia, Flagler, Gadsden, Hamilton, Highlands,
Leon, Levy, Okaloosa, Polk, Putnam, Santa Rosa, St. Johns, Suwannee,
fulva Roger. Alachua, Columbia, Dade, Escambia, Gilchrist, Highlands, Levy, Marion,
Orange, Pasco, Pinellas, Putnam, Santa Rosa, Sumter, Suwannee, Union, Wal-
lamellidens Mayr. Alachua, Duval, Hernando, Highlands, Jackson, Levy, Marion,
Okaloosa, Osceola, Putnam, Sarasota, St. Lucie, Sarasota.
mariae Forel. "Florida"
tennesseensis (Mayr). Jackson, Leon, St. Johns, Suwannee.
treatae Forel. Alachua, Collier, Hamilton, Highlands, Leon, Levy, Monroe, Polk, Put-
Genus PHEIDOLE Westwood
adrianoi Naves. Alachua, Brevard, Collier, Highlands, Marion, Polk.
carrolli Naves. Alachua, Leon.
crassicornis Emery. Alachua, Leon.
dentata Mayr. 38 counties. Widely distributed.
dentigula Smith. 39 counties. Widely distributed.
floridana Emery. 35 counties. Widely distributed, mostly southern.
lamia Wheeler. Leon.
littoralis Cole. Highlands, Polk, Sarasota.
megacephala (Fabricius). Broward, Collier, Dade, Highlands, Hillsborough, Indian
River, Monroe, St. Johns, Sarasota.
metallescens Emery. Alachua, Clay, Dade, De Soto, Duval, Hardee, Hernando, High-
lands, Hillsborough, Lake, Leon, Levy, Marion, Pasco, Putnam, Suwannee,
moerens Wheeler. 27 counties. Mostly southern.
morrisi Forel. Alachua, Broward, Citrus, Clay, Collier, Dade, Gilchrist, Hernando,
Highlands, Jackson, Leon, Levy, Marion, Martin, Orange, Pinellas, Polk, Put-
nam, St. Johns, St. Lucie, Suwannee.
tysoni Forel. Alachua.
Genus CARDIOCONDYLA Emery
emeryi Forel. Alachua, Broward, Collier, Dade, Hardee, Hernando, Highlands, Hill-
sborough, Indian River, Manatee, Martin, Monroe, Okeechobee, Polk, Pasco,
nuda (Mayr). Alachua, Broward, Dade, De Soto, Glades, Hernando, Highlands, Lake,
Marion, Monroe, Pasco, Polk, Putnam, Sarasota.
venustula Wheeler. Broward, Highlands, Manatee, Marion, Monroe, Sarasota.
wroughtoni (Forel). Alachua, Dade, De Soto, Glades, Highlands, Manatee, Marion,
Monroe, Orange, Pasco, Putnam.
Genus CREMATOGASTER Lund
ashmeadi Mayr. 33 counties. Widely distributed.
atkinsoni Wheeler. Alachua, Dade, Holmes, Jackson, Lee, Leon, Levy, Marion, Mon-
roe, Palm Beach, Sarasota, Sumter, Union.
Florida Entomologist 72(1)
cerasi (Fitch). Alachua, Bay, Franklin, Gilchrist, Hernando, Highlands, Leon, Marion,
lineolata (Say). Alachua, Calhoun, Escambia, Gulf, Leon, Nassau, Putnam.
minutissima Mayr. Alachua, Calhoun, Collier, Columbia, Dade, Highlands, Leon,
Levy, Liberty, Madison, Marion, Monroe, Orange, Pinellas, Polk, Putnam, Vol-
pilosa Emery. Alachua, Citrus, Dade, Gilchrist, Highlands, Leon, Levy, Madison,
Marion, Nassau, Pinellas, Putnam, Suwannee.
vermiculata Emery. Alachua, Dixie, Jefferson, Leon, Putnam.
Genus MONOMORIUM Mayr
destructor (Jerdon). Manatee, Monroe, Nassau, Pinellas.
ebeninum Forel. Monroe.
floricola (Jerdon). Broward, Collier, Dade, De Soto, Highlands, Leon, Marion, Martin,
Monroe, Palm Beach, Putnam, St. Lucie, Sarasota.
minimum (Buckley). Alachua, Dade, Brevard, Broward, Monroe, Putnam.
pharaonis (Linnaeus). Alachua, Collier, Dade, Highlands, Monroe, Okeechobee, St.
viride Brown. Alachua, Broward, Columbia, Dade, Flagler, Hernando, Highlands, In-
dian River, Levy, Marion, Polk, Suwannee.
Genus SOLENOPSIS Westwood
carolinensis Forel. Alachua, Columbia, Flagler, Highlands, Leon, Levy, Liberty, Mar-
ion, Putnam, St. Johns, Suwannee.
corticalis Forel. Monroe.
geminata (Fabricius). 26 counties. Widely distributed.
globularia littoralis Creighton. Alachua, Duval, Franklin, Highlands, Martin, Monroe,
Putnam, St. Lucie, Sarasota, Volusia.
invicta Buren. Widely distributed.
nickersoni Thompson. Alachua, Collier, Gilchrist, Highlands, Indian River, Leon, Mar-
ion, Orange, Polk, Putnam, Sarasota, Sumter.
pergandei Forel. Alachua, Escambia, Highlands, Levy, Marion, Putnam, Suwannee.
picta Emery. Alachua, Dade, Highlands, Jackson, Lee, Leon, Monroe, Putnam, St.
tennesseensis Smith. 27 counties. Widely distributed.
xyloni McCook. Escambia, Marion, Santa Rosa.
Genus XENOMYRMEX Forel
floridanus Emery. Alachua, Collier, Dade, Highlands, Monroe, Palm Beach, Polk,
Pinellas, Sarasota, Volusia.
Genus MYRMECINA Curtis
americana Emery. Alachua, Bay, Bradford, Broward, Calhoun, Columbia, Dade, High-
lands, Hillsborough, Jefferson, Lake, Leon, Levy, Liberty, Marion, Monroe,
Nassau, Pasco, Santa Rosa, Suwannee, Volusia.