Title: Florida Entomologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00098813/00079
 Material Information
Title: Florida Entomologist
Physical Description: Serial
Creator: Florida Entomological Society
Publisher: Florida Entomological Society
Place of Publication: Winter Haven, Fla.
Publication Date: 1987
Copyright Date: 1917
Subject: Florida Entomological Society
Entomology -- Periodicals
Insects -- Florida
Insects -- Florida -- Periodicals
Insects -- Periodicals
General Note: Eigenfactor: Florida Entomologist: http://www.bioone.org/doi/full/10.1653/024.092.0401
 Record Information
Bibliographic ID: UF00098813
Volume ID: VID00079
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access

Full Text

(ISSN 0015-4040)


(An International Journal for the Americas)

Volume 70, No. 4 December, 1987


Announcement 71st annual meeting ...................................... .............. .. i

FOSTER, R. E., and R. H. CHERRY-Survival of Fall Armyworm, Spodoptera
frugiperda (Lepidoptera: Noctuidae), Exposed to Cold Temperatures ...... 419
PENA, J. E., R. M. BARANOWSKI, AND R. E. LITz-Life History, Behavior, and
Natural Enemies of Philephedra tuberculosa (Homoptera: Coccidae) ...... 423
ZALOM, F. G., E. T. NATWICK-Developmental Time of Sweetpotato Wi,' ., ri
(Homoptera: Aleyrodidae) in Small Field Cages on Cotton Plants ......... 427
FUNDERBURK, J. E., and T. P. MACK-Abundance and Dispersion of Georcoris
spp. (Hemiptera: Lygaeidae) in Alabama and Florida Soybean Fields .... 432
TLE-The Straw Itch Mite, Pyemotes tritici (Acari: Pyemotidae), as a
Biological Control Agent of Red Imported Fire Ants, Solenopsis invicta
(Hymenoptera: Formicidae) ........................................................ 439
ALVARADO-RODRIGUEZ, B.-Parasites and Disease Associated with Larvae of
Beet Armyworm, Spodoptera exigua (Lepidoptera: Noctuidae), Infesting
Processing Tomatoes in Sinaloa, Mexico ........................................... 444
OPP, S. B., AND R. J. PROKOPY-Seasonal Chipr,, in Resightings of Marked,
Wild Rhagoletis pomonella (Diptera: Tephritidae) Flies in Nature ......... 449
HAAG, K. H., J. C. JOYCE, W. M. HETRICK, AND J. C. JORDAN-Predation on
Waterhyacinth Weevils and Other Aquatic Insects by TI7 , Wetland Birds
in F lorida .................................... .......................... ...... ...... 457
CALLAHAN, J. L., AND C. D. MORRIS-Survey of 13 Polk County, Florida Lakes
for Mosquito (Diptera: Culicidae) and Midge (Diptera: ( I,, ...."...,i. 1, 1.. Pro-
du action .................... . ......................... ....... .. ............................. 471
FRANK, J. H., R. E. WOODRUFF, AND C. A. NUNEZ-Scapteriscus didactylus
(Orthoptera: Gryllotalpidae) in the Dominican Republic ....................... 478
BLAND, R. G.-Mating Behavior of the Grasshoper Melanoplus tequestae (Or-
thoptera: A crididae) ............................... .................................... 483
VICK, K. W., J. A. COFFELT, AND B. A. WEAVER-Presence of Four Species of
Stored-Product Moths in Storage and Field Situations in North-Central
Florida as Determined with Sex Pheromone-Baited Traps ...................... 488
O'BRIEN, L. B.-A Synopsis of New World Lophopidae (Homoptrera: Ful-
goroidea) .................. .............. .................................................. 493
SHELLEY, R. M.- The Scolopendromorph Centipedes of North Carolina, With a
Taxomonic Assessment of Scolopocryptops gracilis peregrinator (Crabill)
(Cl',,I i. i : Scolopendrom orpha) .................................................... 498
Liu, T. -X., AND M. KOSZTARAB-Two New Species of Chionaspis (Homoptera:
Coccoidea: Diaspididae) from North America ................. ................ 512

Continued on Back Cover

Published by The Florida Entomological Society


President ......................................... ................ .................. J. L. Taylor
President-Elect ........................... ................... R. S. Patterson
Vice-President ............................ ............................... ................... J. E Eger
Secretary .............................. .............................. ................... E. R. M itchell
Treasurer ..................................................... .. ................. A. C. Knapp

D. J. Schuster
C. O. Calkins
Other Members of the Executive Committee .............L. S. Osborne
J. H. Epler III
G. S. Wheeler
P. G. Koehler
J. R. McLaughlin


Editor .......................................... ................ J. R. McLaughlin

Associate Editors
Arshad Ali Carl S. Barfield Ronald H. Cherry
John B. Heppner Michael D. Hubbard Lance S. Osborne
Omelio Sosa, Jr. Howard V. Weems, Jr. William W. Wirth

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 December 22, 1987


The Florida Entomological Society will hold its 71st Annual Meeting 2-5 August,
1988 at the Sheraton Sand Key Hotel, Clearwater Beach, Florida, 1160 Gulf Boulevard,
Clearwater Beach, Florida 33515. Details of room rates, registration fees, registration
forms, and call for papers will be mailed by the Program Chairman and will appear in
the Newsletter. Room rates are $69 single or double; meeting rates are in effect from
August 1-7.
Members are reminded to begin thinking about persons they may wish to nominate
for recognition by the Society.
Requests for information should be addressed to:

J. E. Eger, Chairman
Program Committee, FES
Dow Chemical U. S. A.
5100 W. Kennedy Boulevard, Suite 450
Tampa, FL 33609

Phone: 813-287-8300

Foster and Cherry: Fall Armyworm Cold Tolerance 419


University of Florida
Institute of Food and Agricultural Sciences
Everglades Research and Education Center
P. O. Drawer A
Belle Glade, FL 33430


All life stages of seasonally acclimatized fall armyworms, Spodopterafrugiperda (J.
E. Smith), were subjected to temperatures of 0, -2.5, -5, -7.5, and -100C for 3 h in a
temperature cabinet. The adult stage was most susceptible to cold temperatures, with
only 26% of the moths surviving at -50C and no survival at the colder temperatures.
The egg was the most tolerant stage, with 30% survival at -10'C. Fall armyworm
pheromone traps located near Belle Glade, Fla. were checked weekly during the winters
of 1985-86 and 1986-87. Seven nights of subfreezing temperatures during those two
winters resulted in little or no adult mortality. Our study, which included lethal temper-
ature data, microclimate data, and field population data, shows that extreme low tem-
peratures are not important mortality factors on fall armyworm populations in southern


Todas las etapas de vida del aclimatado a la estacion del gusano cogollero, Spodoptera
frudiperda (J. E. Smith), fueron expuestas a temperatures de 0, -2.5, -5, -7.5, y -10C
por 3 h en un gabinete refrigerado. La etapa adulta fue mas susceptible a temperatures
frias, con solamente 26% de polillas supervivientes a -5 C y ninguna supervivencia en
las temperatures mas frias. El huevo fue la etapa mas tolerante con 30% de superviven-
cia a -100C. Las trampas de feromonas del gusano cogollero localizadas cerca de Belle
Glade, Fla. fueron verificadas semanalmente durante el invierno de 1985-86 y 1986-87.
Siete noche de heladas temperatures durante esos inviernos dieron por resultado pocas
o ninguna mortalidad en los adults. En un studio el cual incluye datos de temperatures
letales, datos de microclima, y datos de poblacion en los campos, muestran que tem-
peraturas extemadamentes bajas no afectaron lo suficiente como para causar una gran
mortalidad en la poblacion de gusano cogollero, en el sur de la Florida.

The fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith), is a polyphagous
pest whose annual damage and cost of control in the U.S. has been reported to be as
high as $300 million (Entomol. Soc. Amer. 1979). FAW adults will typically emigrate
from their overwintering range during early spring to infest more northern states
(Luginbill 1928). In mild winters FAW can survive in parts of Florida, Louisiana, and
Texas but during normal or severe winters the overwintering range is limited to south-
ern Florida and southern Texas, as well as more tropical locations (Chittenden 1900,
Hinds & Dew 1915, Luginbill 1928, Vickery 1929, Snow & Copeland 1969, Tingle &
Mitchell 1977). Several researchers have noted FAW overwintering survival in Florida.

Florida Entomologist 70(4)

Tingle & Mitchell (1977) reported that FAW normally overwinter in a salubrious area
around Hastings, Fla. (29" 43'N). Wood et al. (1979) investigated direct effects of ex-
tended periods of cold temperature on survival of pupae in Florida. Waddill et al. (1982)
found that FAW could overwinter in south Florida (250 30'N) even in severe winters
and could sometimes overwinter as far north as Gainesville, Fla. (29038'N). Pair &
Sparks (1986) estimated the overwintering range of FAW to be south of ca. 280 N
latitude in south Florida. Since southern Florida is an important overwintering site for
FAW, our objective was to more fully determine the effect of extreme cold tempera-
tures on survival of all FAW life stages under southern Florida field conditions.


During fall 1986, a FAW colony was established in an outdoor screen house using
field collected insects. Most FAW were free to move about within the screen house and
fed on hybrid sweet corn growing in pots, while other FAW in the screen house were
maintained in 1 oz plastic cups and fed artificial diet (Shorey & Hale 1965). FAW were
chosen from both sources for all tests. All FAW used in our temperature tests were
maintained outdoors in this screenhouse so that the insects would be seasonally
acclimatized. Seasonal acclimatization can be an important factor in the survival of
insects subjected to extreme temperatures (Salt 1961, Bursell 1974). The test insects
were collected from the colony and exposed to experimental temperatures within 1 h
after collection in a glass jar (0.95 liter) with a thermometer through the lid to monitor
the temperature within the jar. Approximately 10 individuals of each developmental
stage were put into the jar for each test. Several corn leaves were placed in the jar to
provide a resting site for adults and large larvae (4th-6th instar). Eggs, small larvae
(lst-3rd instar), and corn leaves were placed in two screen-topped 37 mm diameter metal
cans so that the eggs and small larvae would not get lost or crushed in the jar. The
tests were conducted in an environmental chamber and the temperature was checked
every 30 min to ensure maintenance of a constant temperature ( 0.5C) within the
jar. After reaching the test temperature, the duration of exposure to each temperature
was 3 h because thermograph records collected over several years from a weather
station at the Everglades Research and Education Center at Belle Glade, Fla. (260
40'N) showed that extreme low temperatures on winter mornings were almost always
of short duration, usually ca. 3 h. Live insects were counted during a 2 h observation
period after the temperature exposure because insects recorded as live during that
period remained alive and insects that appeared dead still appeared dead the following
day. The FAW eggs were stored on moist filter paper in petri dishes at room temper-
ature (ca. 250C) either until hatch or until it was obvious that hatching would not occur
(ca. 10 days). Each jar containing all FAW stages exposed to a specific temperature
was considered a replicate and only one replicate was tested on any one day. Four
replicates were made at each of the 5 test temperatures (0, -2.5, -5, -7.5, and -10C);
all tests were conducted between 3 Dec. 1986 and 20 Mar. 1987.
Bursell (1974) stated that when studying the effects of temperature on insect survi-
val, it is important to measure the temperature of the microenvironment, rather than
just the general environment. Microclimate temperatures experienced by FAW in
southern Florida cornfields were measured with a portable thermometer with a surface
temperature probe (4.5 mm diameter). Air temperatures were measured 0.5 m above
ground. Leaf temperatures were taken from corn leaves 0.3 m aboveground by taping
the probe to the undersurface of the leaf, where an egg mass would usually be ovipo-
sited. Whorl temperatures were measured by gently sliding the probe down into the
whorl of a corn plant, where larvae would typically be found. Soil temperatures were


December, 1987

Foster and Cherry: Fall Armyworm Cold Tolerance 421

determined by placing the probe 2 cm below the soil surface and gently packing soil
around it. FAW larvae drop to the ground and pupate 2-7 cm below the soil surface
(Luginbill 1928). Each of these temperatures was measured every 10 min at different
sites in a cornfield between 0500 and 0600 hours (EST) on 19 Mar. 1987. This hour was
chosen because extreme cold normally occurs during this hour in southern Florida.
Four cone traps (Hartstack et al. 1979) baited with a 4-component FAW pheromone
supplied by Terochem Laboratories Ltd. were checked weekly during December-March
of 1985-86 and 1986-87. Three of the traps were located on the 325 ha Everglades
Research and Education Center farm and the fourth trap was located on a local farm
approximately 16 km away. Temperature data collected at the weather station at our
research center at Belle Glade were compared with the moth catches in the pheromone
traps to determine how the 7 nights of sub-freezing weather experienced during those
two winters affected FAW moth catches.
Mean survival of various stages were compared at each test temperature using
Duncan's (1955) multiple range test.


Table 1 summarizes the results of the exposure of the FAW life stages to various
cold temperatures for 3 h. The egg stage was the most tolerant and the adult stage was
the least tolerant to cold temperatures. There was no significant reduction in egg survi-
val until -10.0C and even at this coldest test temperature, 30.3% of the eggs survived.
Small larvae also did not experience a significant loss in survivability until exposed to
-100C, but no survival was observed at this temperature. Large larvae, pupae, and male
and female adults had significant reductions in survival at -5.0C. No adults survived
at -7.5 or -10.0C. At -10.0C, 5% of the pupae survived.
The temperatures measured on the undersurface of a corn leaf, in the whorl of a
corn plant, and 2 cm below the soil surface were virtually the same as the ambient air
temperatures. Therefore, it can be assumed that air temperature measured 0.5 m
aboveground is a good estimator of the temperatures experienced by FAW. On 26 Dec.
1985, the air temperature at the research center dropped to -1.1C. The mean FAW
moth catch in the four pheromone traps during the week prior to the freeze was 49
moths per trap. The mean catch during the week of the freeze was 29 and the mean
catch the following week was 56 moths per trap. A more severe freeze was experienced
on 28 and 29 Jan. 1986, when the low temperatures were -3.3 and -2.8C, respectively.
The mean moth catches in this instance increased from 45 to 46 during the week of the


Temperature (C)
Stage 0 -2.5 -5.0 -7.5 -10.0

Eggs 91.0 a 96.0 a 88.0 a 70.0 a 30.3 b
Small larvae 95.0 a 95.0 a 87.0 a 66.8 a 0.Ob
Large larvae2 100.0 a 95.0 a 40.8 b 2.5 c 0.0 c
Pupae 100.0 a 97.0 a 49.5 b 20.8 c 5.0 c
Adult males 88.3 a 95.0 a 26.0 b 0.0 c 0.0 c
Adult females 88.3 a 80.5 a 25.8 b 0.0 c 0.0 c

'Means within a row followed by the same letter are not significantly different (P s 0.05), Duncan's (1955) multiple
range test.
'Small larvae = instars 1-3; large larvae = instars 4-6.

Florida Entomologist 70(4)

freeze and dropped to 36 the following week. Low temperatures of -1.7 and 0C were
measured on 2 and 3 Mar. 1986. The mean moth catches were 47, 28, and 19 moths per
trap for the weeks before, during, and after the freezes, respectively. Finally, during
two consecutive trapping periods in Jan. 1987, the low temperature reached -0.6C on
24 and 28 Jan. This time the mean trap catches increased from 24 moths per trap during
the week prior to the first freeze to 184 and 158 moths per trap during the next two
weeks. Many factors may affect pheromone trap catch such as population size, wind,
and temperature. However, overall our trap catch data show that subfreezing temper-
atures of short duration during the winters of 1985-86 and 1986-87 had little effect in
reducing FAW field populations, which is consistent with our laboratory data.


We thank Alvin Wilson for his assistance in this project. Florida Agricultural Exper-
iment Station Journal Series No. 8230.


BURSELL, E. 1974. Environmental aspects-temperature, pp. 2-43. In M. Rockstein
(ed.), The physiology of insecta, 2nd ed. Academic Press, New York and London.
CHITTENDEN, F. H. 1900. The fall armyworm in 1889. USDA, Div. Ent. Bul. 23:
DUNCAN, D. B. 1955. Multiple range and multiple F test. Biometrics 11: 1-42.
ENTOMOL. SOC. AMER. 1979. Southeastern Branch of the Entomol. Soc. of America
insect detection, evaluation, and prediction report. Vol. 3, 49 pp.
HARTSTACK, A. W., JR., J. A. WITZ, AND D. R. BUCK. 1979. Moth traps for the
tobacco budworm. J. Econ. Entomol. 97: 1077-1089.
HINDS, W. E., AND J. W. DEW. 1915. The grass worm or fall armyworm. Alabama
Agr. Exp. Stn. Bull. 186: 61-92.
LUGINBILL, P. 1928. The fall armyworm. USDA Tech. Bull. No. 34. 92 pp.
PAIR, S. D., AND A. N. SPARKS 1986. Evidence of annual long distance migration by
the fall armyworm. In A. N. Sparks (ed.), Long-Range Migration of Moths of
Agronomic Importance to the United States and Canada: Specific Examples of
Occurrence and Synoptic Weather Patterns Conducive to Migration (ESA Sym-
posium, 1982). USDA Misc. Publ.
SALT, R. W. 1961. Principles of insect cold-hardiness. Ann. Rev. Entomol. 6: 55-74.
SHOREY, H. H., AND R. L. Hale. 1965. Mass rearing of the larvae of noctuid species
on a simple artificial medium. J. Econ. Entomol: 58: 522-524.
SNOW, J. W., AND W. W. COPELAND. 1969. Fall armyworm: Use of virgin female
traps to detect males and to determine seasonal distribution. USDA Prod. Res.
Rep. No. 110, 9 pp.
TINGLE, F. C., AND E. R. MITCHELL. 1977. Seasonal populations of armyworms and
loopers at Hastings, Florida. Florida Entomol. 60: 115-122.
VICKERY, R. A. 1929. Studies of the fall armyworm in the Gulf Coast district of
Texas. USDA Tech. Bull. No. 138. 64 pp.
1982. Seasonal abundance of the fall armyworm and velvetbean caterpillar
(Lepidoptera: Noctuidae) at four locations in Florida. Florida Entomol. 65: 350-
WOOD, J. R., S. L. POE, AND N. C. LEPPLA. 1979. Winter survival of fall armyworm
pupae in Florida. Environ. Entomol. 8: 249-252.


December, 1987

Peia et al.: Scale Biology


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


Philephedra tuberculosa Nakahara and Gill is a pest of papaya (Carica papaya L.),
sugar apple (Annona squamosa L.), and various ornamentals. The life history of the
scale and its natural enemies are discussed. Life span is 24 2 days for males and 59 1
days for females. Important natural enemies include Coccophagous lycimnia Walker
(Hymenoptera:Aphelinidae), Trichomastus portoricensis Crawford (Hymenoptera:En-
cyrtidae), Diomus austrinus Gordon, Hyperaspis ornatella Gordon, Olla sp. (probl.
abdominalis (Say)), Cryptolaemus montrouzieri Muls., Azya sp., (Coleoptera:Coccinel-
lidae), Diadiplosis pulvinariae (Felt) (Diptera:Cecidomyiidae), Ocyptamus sp. (Dipt-
era:Syrphidae), Laetilia sp. (Lepidoptera:Pyralidae), Verticillium lecanii (Zimmenn.)
Viegas (Fungi Imperfecti).


Philephedra tuberculosa Nakahara y Gill es una de la papaya (Carica papaya L.),
del an6n (Annona squamosa L.), y de varias plants ornamentals. Se discuten la
biologfa, es dafo y los enemigos naturales de la escala. El ciclo de vida fluctua entire
24-2 dias para los machos y 591 dias para las hembras. Los enemigos naturales mis
importance incluyen Coccophagous lycimnia Walker (Hymenoptera: Aphelinidae),
Trichomastus portoricensis Crawford (Hymenoptera: Encyrtidae), Diomus austrinus
Gordon, Hyperaspis ornatella Gordon, Olla sp. (probablemente abdominalis (Say)),
Cryptolaemus montrouzieri Muls., Azya sp., (Coleoptera: Coccinelidae) Diadiplosis
pulvinariae (Felt) (Diptera: Cecidomyiidae), Ocyptamus sp. (Diptera: Syrphidae),
Laetilia sp. (Lepidoptera: Pyralidae), y Verticillium lecanii (Zimmenn.) Viegas (hongo

Philephedra tuberculosa Nakahara and Gill occurs in Central and northern South
America (Nakahara & Gill 1985). It was first found in Florida infesting gumbo limbo
(Bursera simaruba Sarg.) leaves and stems in Miami in 1981 (R. M. Baranowski, pers.
comm.). Infestations of the scale have been found during the past 2 years on more than
50 plant species (Nakahara & Gill 1985). Infestations of P. tuberculosa have on occasion
resulted in considerable losses to nurserymen in South Florida and can severely damage
papaya and Annona sp. (Pena et al. 1984). Scale infestations result in 3 types of damage
to papaya plants. First, flower and leaf drop occur from severely infested young plants.
Secondly, when the infestation on seedlings or on young plants is localized near the
apex, distortion of apical leaves is induced. Thirdly, females attached to the fruit cause
cosmetic damage that makes the fruit unmarketable. Moreover, because of its wide
range of hosts, including Citrus spp., it has the potential to become a very serious pest.
The goal of this study was to determine the life history and behavior of P. tuberculosa
and to identify its principal natural enemies.

Florida Entomologist 70(4)


Rearing and Life History. The scale was reared on 2-month-old papaya plants and
on green papaya fruits under laboratory conditions (T 27 2C; RH 75 3%) using a
modification of the technique reported by Tashiro (1966) for rearing the California red
scale. Fully formed ovisacs from P. tuberculosa were placed on 9.6 cm-diameter papaya
fruits (n = 15) positioned on 200 ml water-filled jars. Alternatively, 15 cm tall potted
papaya plants (n=20) were used. Ovisacs were placed on the apical bud and were
removed when eggs hatched. A maximum of 30 crawlers per fruit or per plant were
allowed to develop. The length of time that a crawler moves on a fruit was determined
by marking with a number the initial feeding site and recording the new sites at 2, 8,
72, 96, and 120 h. Observations were then made on alternate days, and body length and
thorax width of each scale were measured. Observations continued until all individuals
died, or reached adulthood.
Distribution of P.tuberculosa on 3-month-old papaya plants was determined by
counting the occurrence of male pupal tests and third instar females on the stem and
Parasite and Predator Survey. Sampling for parasites and predators of P. tuber-
culosa consisted of bimonthly of collections of infested leaves of C. papaya, Annona
spp., Bauhinia spp. and Bucida spp. The abundance of natural enemies was observed
from February 1983 until May 1984. Infested samples were placed in plastic bags and
transported to the laboratory, where they were placed on petri dishes and daily ob-
served for parasite emergence or for predator activity.


Life History and Behavior

Crawlers. The active nymph is pale yellow and oval, 0.40+ 0.04 mm (n= 115) in
length and 0.19 0.02 mm (n= 115) at the widest part of the thorax. The duration of
the 1st stage is 2-8 days. In our laboratory studies gravid females (n= 4) produced an
average of 87 crawlers/day over a 7 day period, but consistently produced the greatest
number on the 5th day. During the first 24 h after hatching the crawlers moved to the
less illuminated side of the fruit. One hundred percent of the crawlers moved during
the first 2 hours; 8 hours later 92% settled and remained settled for 72 h. In the
laboratory, one day later, 40% moved again; 2 days later, 72% had relocated. By the
6th day they had settled or dropped from the fruit. Crawlers dispersed as far as 12 cm
away from the parent female on the fruit. Thirty percent of the crawlers (n = 89) settled
around the female parent, whereas about 33% settled at 5-12 cm from the parent. As
we observed in the field, crawlers can emerge from beneath the ovisac or can be released
through a lateral ovisac rupture which is caused by external physical disturbance, wind,
Second instar male. Males can be differentiated during the second instar by the
presence of waxy filaments radiating from the prosoma and from the anal region. Eight
to twelve days after hatching, the male body is oval, flat to moderately convex, yellow-
green and possesses 2 anal filaments. No movement occurs during this period. The
number of filaments increases daily. Filaments (0.1-0.35 mm) are located around the
adfrontal areas. Ten days after molt, the filaments cover the whole body, and long
filaments can reach 0.5 cm. Duration of the second male stage is 4-6 days.
Second instar female. Females molt 10-16 days after egg hatch. The body is yellow-
green and oval-round with a wider abdomen than thorax. Body length ranges from 0.9
to 1.9 mm. Honeydew production is first noted. Dorsum is 0.3 mm high. The second


December, 1987

Pera et al.: Scale Biology

instar active females are active and frequently move from leaves to more succulent
tissues such as petioles or young stems. Duration of the stage is 8 days.
Male prepupa. After the second molt, the light green body shrinks, leaving space
between the prepupa and the test. The prepupa has red eyes and the antennae are
clearly visible. Two anal "spines" or caudal filaments are present. The abdomen width
is 0.6 mm and thorax 0.5 mm. Duration of the prepupal stage is 7-8 days.
Male pupa. The male pupae show clearly formed wings. Antennae and legs are
visible. Style length is 2.1 mm when completely formed. Average pupal length is
1.30.3 mm without style. The duration of this stadium is 5.5 days.
Third instar female. The female body is yellowish-green, and ranges from 3.2-5 mm
in length. The width of the thorax is 1.6 mm, the abdomen 2.2 mm. The number of
marginal setae is 152. Duration of this stadium is 6-8 days.
Adult male. Only winged adult males were observed. Color yellow to orange; eyes
black; scutellum dark orange; head black; average length with style 3.20 mm, without
style 2.1 mm. Emergence occurred during late afternoon. Adult male (n=9) life span
ranged 10 to 22 hours. Males were able to copulate as soon as they emerged. Copula
was maintained for ca. 1 hr. Mating occurred at any time of the day.
Adult female. The dorsum is differentiated from the rest of the body. Dorsum is
elevated 2.5-3 mm high. The female profile is highly convex. Waxy material secreted
through tubular ducts and pores of the prosoma and abdomen indicates ovisac formation.
When the female has the prosoma covered by the waxy material the ovisac is considered
formed. Ovisac length 0.8 cm (0.1-2.5). Ovisacs can be produced without egg formation,
although these females are probably not fertilized. Ovisac formation begins 8-10 days
after molting. Ovisac development is completed in ca. 30 days. A newly formed ovisac
is elongate and quadricarinate and contains 37-52 eggs; fully formed ovisac (n = 40)
may contain 360-900 eggs. The color of the ovisac changes from white to cream when
the eggs mature and changes to brown when the female dies.
Eggs. Eggs are oval and pale green when first deposited, but change to yellow and
finally to light orange before hatching. Egg length (n = 100) is 0.247+0.06 mm, and
width is 0.12 0.012 mm. The incubation period at 27C ranged from 12-17 days.
Distribution. On young papaya plants (48 cm ht), females were found in the upper
third of the stem and petioles. Females were twice as abundant on the leaf underside
as on the upperside. Males were twice (13.0+4.3) as abundant on leaves as females
(7.5 1.04), but were found less frequently than females on the petioles. This difference
probably occurs because of the longer female life span and the need for a more perma-
nent and succulent substrate for survival.

Parasite and Predator Survey

Two parasites and 9 predators were identified as natural enemies of P. tuberculosa
in southern Florida (Table 1). Coccophagous lycimnia Walker and Trichomastus por-
toricensis Crawford (Encyrtidae) were the only parasites emerging from T. tuberculosa.
Parasitism by C. lycimnia ranged from 7-41% for 2nd instar females and 14-41% for
2nd instar males T. portoricensis was recovered from 2nd instar males and females.
Moreover, virtually all recoveries were made from November 1983 through May 1984
when the 2nd instar stages were most abundant.
Trichomastus portoricensis is a previously unrecorded parasite in the continental
U.S.A. This parasite was recovered from scale infested papayas plants in the fall of
1983 and was found at relatively low frequencies at the sample locations. Parasitism
ranged from 0.5 to 5.3%.
Nine predators were found feeding on eggs and early stages of P. tuberculosa. The
most common predator was the lady beetle Diomus austrinus Gordon. Larvae and

426 Florida Entomologist 70(4) December, 1987


Order Family Species Function

Hymenoptera Aphelinidae Coccophagus lycimnia Walker parasite
Encyrtidae Trichomastus portoricensis Crawford "
Coleoptera Coccinellidae Diomus austrinus Gordon predator
Olla sp. (prob. abdominalis (Say))
Hyperaspis ornatella Gordon
Cryptolaemus mountrouzieri Muls.
Azya sp.
Diptera Cecidomyiidae Diadiplosis pulvinariae Felt)
Syrphidae Ocyptamus sp.
Lepidoptera Pyralidae poss. Laetilia sp.
Neuroptera Chrysopidae Chrysopa sp.

adults were observed consuming eggs on sugar apple, papaya, and Bauhinia trees.
Hyperaspis ornatella Gordon, Olla sp. (prob. abdominalis (Say)); Azya sp., Cryp-
tolaeumus montrouzieri Muls. as well as Laberus sp. were observed feeding on scale
eggs found on black olive trees (Bucida buceras L.). Chrysopa sp. larvae were observed
feeding on crawlers and early instars. Pyralid larvae posss. Laetilia sp., were observed
during March 1983 consuming scale eggs. From 1 infested fruit (17 female scales/fruit)
all scales had 1 pyralid larva consuming eggs. Larvae of a flower fly (Ocyptamus sp.)
were found on female ovisacs during October 1983. According to H. V. Weems, Jr.
(personal communication) Ocyptamus sp. appears to be new to Florida. The
cecidomyiid, Diadiplosis pulvinariae (Felt), a common scale insect egg predator, was
also found in infested sugar apple seedlings.
The fungus Verticillium lecanii was previously reported by Pena and McMillan
(1986) as one of the most promising biotic mortality factors for P. tuberculosa in south-
ern Florida. This fungus affects immature and adult stages of the scale. The fungus was
most common during July-October 1983. Further studies will be required to investigate
the potential for P. tuberculosa natural enemies.


We are indebted to the following individuals for their kindness in identifying speci-
mens: Dr. A. B. Hamon, Dr. R. E. Woodruff, Dr. H. V. Weems, Jr., Florida Depart-
ment of Agriculture and Consumer Services; Dr. R. D. Gordon, Dr. R. J. Gagne, Dr.
M. E. Schauff, BIII, USDA.
We are further indebted to Ms. S. Brown, Mrs. H. Glenn, and W. Dankers for
assistance in the present study. Special thanks are given to Mrs. C. Sullivan for kindly
typing this manuscript.
This research was supported in part by USDA Cooperative Agreement No. 5B-
7B30-9-116. Florida Agricultural Experiment Station Journal Series No. 7162.


NAKAHARA, S., AND R. J. GILL. 1985. Revision of Philephedra, including a review
of Lichtensia in North America and description of a new genus, Metapulvinaria
(Homoptera:Coccidae). Entomography 3: 1-42.
PENA, J. E., H. GLENN, AND R. M. BARANOWSKI. 1984. Important insect pests of
Annona spp. in Florida. Proc. Florida State Hort. Soc. 97: 337-340.

Zalom & Natwick: Whitefly Development

PENA, J. E., AND R. T. MCMILLAN, JR. 1986. Verticillium lecanii, a new fungal
parasite of the scale Philephedra tuberculosa (Homoptera:Coccidae) in Florida.
Florida Entomol. 69: 416-417.
TASHIRO, H. 1966. Improved laboratory techniques for rearing California Red Scale
on lemons. J. Econ. Entomol. 59: 604-607.

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


IPM Implementation Group
University of California
Davis, CA 95616


University of California
Cooperative Extension
El Centro, CA 92243


Immature development of the sweetpotato whitefly, Bemisia tabaci (Gennadius)
was studied on cotton plants in the field using small cages to confine insects on the leaf.
Eggs, nymphal instars, and empty pupal cases were counted every 2 to 4 days, and
physiological time was calculated using developmental thresholds of 100C (minimum)
and 320C (maximum). Developmental times for the various life stages were similar for
both years of the study (1984 and 1985). Heat units for eggs, nymphs, and pupae
averaged 27.0, 268.5, and 73.0 degree-days, respectively. The average generation time
was 369.5 degree-days.

Se estudi6 en el campo el desarrollo de las etapas inmaduras de la mosca blanca de
la papa dulce, Bemisia tabaci (Gennadis), en plants de algod6n usando jaulas pequefias
para confinar los insects en la hoja. Se contaron los huevos, estadios ninfales, y los
sacos vacios de pupas cada 2 a 4 dias, y se calcul6 el tiempo fisiol6gico usando los
umbrales de desarrollo de 10C (minimo) y 32'C (maximo). El tiempo de desarrollo de
las varias etapaps biol6gicas fueron similares en ambos afios del studio (1984 y1985).
Unidades de calor para los huevos, ninfas, y pupas, promediaron 27.0, 268.5, y 73.0
dia-grado respectivamente. El promedio de tiempo de una generation fue de 369.5

The sweetpotato whitefly, Bemisia tabaci (Gennadius) is a serious pest of various
crops in many parts of the world, particularly in tropical and subtropical areas (Mound
1962). The world literature on this pest was recently reviewed by Cock (1986). It infests
a wide range of host plants. In southwestern U.S. and in northern Mexico, lettuce,


Florida Entomologist 70(4)

melon, squash, and cotton crops were severely damaged in recent years (Natwick 1983).
Damage inflicted by this insect results in reduced yield (Mound 1965), contamination
due to honeydew and sooty molds (Gerling et al. 1980), and virus transmission (Dickson
et al. 1954). High B. tabaci populations during the 1986 season in California's Imperial
Valley prompted chemical defoliation of the cotton crop earlier than normal to prevent
excessive contamination of lint with honeydew and sooty molds (E. Natwick, Pers.
B. tabaci populations vary significantly from season to season, and control becomes
difficult when whitefly populations are excessive. The ability to forecast B. tabaci pop-
ulation cycles and abundance would give growers the ability to initiate cultural or chem-
ical control actions before whitefly populations become damaging and before migration
to other host crops occurs (Natwick and Zalom 1984). Research conducted in the desert
valleys by Butler et al. (1983) and by Natwick and Zalom (1985) resulted in the estima-
tion of developmental thresholds and generation times for B. tabaci. Field collections
of adult whiteflies in the Imperial Valley in 1982 and 1983 also provided a basis upon
which population abundance could be predicted (Zalom et al. 1985). Field observations
by Natwick and Zalom (1985) during the 1984 cotton production season confirmed that
the generation time for B. tabaci as measured by population peaks from their D-vac
samples (333.8D) was comparable to that previously proposed (323.3D). In practice,
however, it is difficult to use this information as it does not provide information on the
developmental time required for eggs and the nymphal instars. The purpose of this
study was to determine developmental times of the immature stages of B. tabaci under
field conditions.


These studies were conducted on cotton planted at the University of California's
Imperial Valley Agricultural Center (IVAC) located 5 km west of El Centro, Imperial
Co. Climatic data were collected from a California Irrigation Management Information
System automatic meteorological station near the site. The plants were not treated with
Cotton leaves were artificially infested with adult female B. tabaci by placing them
in small cages on the under surface of the leaves. The cages were 19 mm in diameter,
covered with fine mesh, and attached to the leaves with wire clips so that the uncovered
cage bottom fit snugly against the leaf. The females were allowed 3 days for oviposition
in 1984 and 4 days in 1985, and the adults were then removed. The cages were then
replaced over the arena on each leaf to exclude predators, parasites, and potential
oviposition by native female whiteflies. A total of 62 cages containing female B. tabaci
was placed on cotton leaves on 11 June, 1984. The experiment was repeated on 22 July,
In 1984, eight leaves with cages were removed every 2 to 4 days from 14 June
through 29 June, and 7 leaves with cages were removed on 2 July and 4 July. In 1985,
6 leaves with cages were removed on 22 July and 25 July. Thereafter, 5 leaves with
cages were removed every 3 days from 28 July through 18 August. The leaves were
immediately cooled and transported to a laboratory where the circumferences of the
cages were marked, and the cages removed. The marked area was then examined using
a binocular dissecting microscope for the number of eggs, 1st (crawlers), 2nd, 3rd, and
4th instar nymphs, 4th instar pupae, and empty pupal cases (representing emerged
adults). Only eggs and nymphs that appeared to be viable were counted. Empty or
shriveled eggs were not included in the totals. Fourth instar nymphs and pupae are
equivalent to "early" 4th instar substage and transitional plus phorate adult substages
of the 4th instar as described by Nechols and Tauber (1977).


December, 1987

Zalom & Natwick: Whitefly Development

The number of eggs and nymphs of each instar observed on each sampling date were
examined in relation to physiological time expressed as degree-days (D) between a
minimum developmental threshold of 10OC and a maximum threshold of 32'C (Zalom
et al. 1985) using Allen's (1976) double sine wave method. B. tabaci development was
then compared between the 2 years for the various life stages studied. Average 'D for
each life stage was calculated as:

x, + Y2 x1 + Y1
D [1]
2 2

D1 = Average accumulated D for a life stage.
x, = Accumulated D from initial infestation to peak of life stage in 1984.
x2 = Accumulated D from initial infestation to peak of life stage in 1985.
y1 = Accumulated D from initial infestation to peak of subsequent life stage in 1984.
Y2 = Accumulated D from initial infestation to peak of subsequent life stage in 1985.


Our data on the immature development of B. tabaci was very consistent in the 2
years of our study (Figure 1). This permits us to draw some generalities about the
developmental dynamics of this insect on desert cotton.
Eggs-Following infestation, eggs were observed for 210D in 1984 and 2090D in
1985. The peak abundance of eggs occurred at 92 after infestation in 1984 and at 111D
in 1985. The average duration of the egg stage for these 2 years was 270D.
Nymphs-The first nymphs were observed at 920D after infestation in 1984 and at
490D in 1985. Peak abundance of 1st through 4th instar nymphs occurred at 146, 146,
283, and 344D after infestation in 1984, and at 111, 149, 295, and 344D after infestation
in 1985. The average duration of each of the 4 nymphal instars in the 2 years was 190D,
140.5D, 55D, and 54D, respectively, or 268.50D total. The shortest duration was on
average 206D. First instar nymphs both appeared initially and reached peak abundance
earlier in 1985 than in 1984. This was likely due to longer adult exposure to the leaf
arena in 1985 than in 1984. Our data were insufficient to calculate the longest duration
of the range of nymphal development.
Pupae-Pupae were first observed at 344'D following infestation in 1984 and at
2090D following infestation in 1985. Peak abundance of this stage occurred at 396D in
1984 and at 400D in 1985. The average duration of this stage for 2 years was 73D. The
shortest duration averaged 43D. Again, insufficient data precluded our ability to calcu-
late the longest duration of the range.
Adult emergence-The first empty pupal cases observed after infestation were noted
at 344D in 1984 and at 295D in 1985. These empty pupal cases correspond to adult
emergence. Peak adult emergence following infestation occurred at 4540D in 1984 and
at 4880D in 1985. The average peak egg to peak adult emergence interval for the 2 years
was 369.50D. This compares favorably to the mean generation times of 334.1D (1982)
and 333.80D (1983) that was calculated between peaks of adult populations sampled from
Imperial Valley cotton fields using a D-vac suction machine (Natwick and Zalom 1985).
Implications-B. tabaci can have as many as 7 to 9 generations per year in the
southwestern U.S. and northern Mexico. The population dynamics of this insect in
individual fields becomes more difficult to discern as the season progresses. Because
there is a range of heat units required for development among individuals of a popula-
tion, the extremes of this range become compounded in successive generations resulting

Florida Entomologist 70(4)

December, 1987

Bemisao tobao
500 -1 984
.. 1985



200 -

100 -

0 100 200 300 400 500
Accumulated Degree-Days Between IOC and 32C

Bemis,4 ltobc



75 -

50 -


0 100 200 300 400 500
Accumulated Degree-Days Between IOC and 32C

Accumulated Degree-Days Between IO'C and 32C

Accumulated Degree-Days Between 10 C and 32C

Bemisrtolo9 ct
-- 1984
...--.- 1985




100 200 300 400 500
Accumulated Degree-Days Between IOC and 32C

Bem sa toboct
S....... 19 85



100 200 300 400 500
Accumulated Degree-Days Between 10C and 32C

Fig. 1. Accumulated degree-days between 10'C and 320C from initial infestation
with adult females for Bemisia tabaci developmental stages.

in considerable overlap of generations. This has been a factor making control efforts

difficult against later B. tabaci generations. In spite of generation overlap it is possible

to discern adult peaks in most generations in individual fields (Zalom et al. 1985).

Sampling adult peaks and using the information on nymphal development presented
in this study would prove useful in targeting control actions for early instar B. tabaci

nymphs. This would be especially true in earlier generations when systemic granular

insecticides or directed spray mixtures of foliar insecticides could applied.


ALLEN, J. C. 1976. Modified sine wave method for calculating degree-days, Environ.
Entomol. 5: 388-396.


..* .. ........

* \

Zalom & Natwick: Whitefly Development 431

BUTLER, G. D., T. J. HENNEBERRY, T. E. CLAYTON. 1983. Bemisia tabaci
(Homoptera: Aleyrodidae): Development, oviposition, and longevity in relation
to temperature. Ann. Entomol. Soc. America 76(2): 310-313.
COCK, M. J. W. 1986. Bemisia tabaci an important literature survey of the cotton
whitefly with an annotated bibliography. Commonwealth Agric. Inst. of Biol.
Control. Ascot Berks, England.
DICKSON, R. C., M. MCD. JOHNSON, AND E. F. LAIRD, JR. 1954. Leaf crumple, a
virus disease of cotton. Phytopathology 44: 479-480.
GERLING, D., U. MOTRO, AND R. HOROWITZ. 1980. Dynamics of Bemisia tabaci
(Gennadius) (Homoptera: Aleyrodidae) attacking cotton in the coastal plain of
Israel. Bull. Entomol. Res. 70: 213-219.
MOUND, L. A. 1962. Studies on the olfactory and colour sensitivity of Bemisia tabaci
(Genn.) Entomol. Expl. Appl. 5: 99-104.
MOUND, L. A. 1965. Effect of whitefly (Bemisia tabaci) on cotton in the Sudan Gezira.
Empire Cotton Growing Rev. 42: 290-294.
NATWICK, E. T. 1983. Beneficial insect monitoring as part of a pest abatement district
requiring applications of gossyplure to cotton in the Imperial Valley. Proc.
Beltwide Cotton Prod. Res. Conf. 36: 193-196.
NATWICK, E. T., AND F. G. ZALOM. 1984. Surveying sweetpotato whitefly in the
Imperial Valley. California Agric. 38(3&4): 11.
NATWICK, E. T., AND F. G. ZALOM. 1985. Verification of the cotton whitefly popula-
tion model, Bemisia tabaci Gennadius (Homoptera: Aleyrodidae). Proc. Beltwide
Cotton Prod. Conf. 38: 174-175.
NECHOLS, J. R., AND M. J. TAUBER. 1977. Age-specific interaction between the
greenhouse whitefly and Encarsia formosa: influence of the parasite on host
development. Environ. Entcmol. 6: 207-210.
ZALOM, F. G., E. T. NATWICK, AND N. C. TOSCANO. 1985. Temperature regulation
of Bemisia tabaci (Homoptera: Aleyrodidae) populations in Imperial Valley Cot-
ton. J. Econ. Entomol. 78(1): 61-64.

432 Florida Entomologist 70(4) December, 1987


Department of Entomology and Nematology,
University of Florida,
North Florida Research and Education Center,
Route 3 Box 4370,
Quincy, Florida 32351


Department of Entomology
301 Funchess Hall
Auburn University, Alabama 36849-4201


Seasonal abundance and dispersion characteristics of adult and nymphal Geocoris
spp. (bigeyed bug) populations were determined for soybean fields located in Alabama
and Florida. Three fields were sampled in 1985 and two fields were sampled in 1986.
Populations were present in each field from middle or late vegetative stages of crop
growth until crop senescence. Numbers increased through the growing season and
usually were greatest on the last or near the last sample dates. At least three complete
and additional partial generations were typical, with populations of different generations
broadly overlapping. Population dynamics were similar to previously reported data in
other states in the southern U.S., indicating a similar phenology for bigeyed bugs in
soybean fields throughout the region. Dispersion statistics of variance/mean ratio and
Taylor's power law were calculated for adult and nymphal sample estimates. Populations
of nymphs were always aggregated. Adults usually were randomly distributed, but
sometimes aggregated.


Se determine la abundancia estacional y las caracteristicas de dispersion de las pob-
lacioines adults y ninfales de Geocoris spp. en campos de frijoles de soya en Alabama
y la Florida. Se mustrearon tres campos en 1985 y dos en 1986. Poblaciones estuvieron
present en cada campo desde la etapa vegetativa media o tarde del crecimiento del
cultivo, hasta la senectud del cultivo. Los nameros aumentaron durante el period de
crecimiento y usualmente fueron mayor en la fltima o cerca de la iltima fecha in que
se muestre6. Fueron tipicas tres generaciones completes y una parcial adicional, con
poblaciones de diferentes generaciones solapando ampliamente. El dinamismo de la
poblaci6n fue similar a los datos reportados previamente en otros estados del sur de los
Estados Unidos, indicando una fenologia, similar de Geocoris spp. en campos de frijoles
de soya en toda la region. Se calcul6 la estadistica de dispersion de la proporci6n de
variacion/promedio y de la ley de Taylor, para los estimados de muestras de adults y
de ninfas. Poblaciones de ninfas siempre fueron agregadas. Los adults usualmente
estaban distribuidos al azar, aunque algunas veces se agregaban.


Funderburk & Mack: Geocoris in Soybeans

Geocoris spp. (bigeyed bugs) are common, polyphagous insect predators in many
crops. Populations of Geocoris punctipes (Say) occur in large numbers in soybean fields
in the southern U.S. (Tugwell et al. 1973, Turnipseed 1972, Shepard et al. 1974a). The
diet of arthropod prey is supplemented with some feeding on plants, which improves
survival and decreases developmental time (Naranjo & Stimac 1985). The predator is
an important biological control agent of Anticarsia gemmatilis Hubner (Elvin et al.
1983) and Nezara viridula (L.) (Crocker & Whitcomb 1980), which are major pests of
soybean. Bigeyed bugs may be important agents in the suppression of populations of
occasional pests of soybean, such as Heliothis zea Boddie (Whitcomb & Bell 1964), H.
virescens, (F.) (McDaniel & Sterling 1979), Pseudoplusia includes (Walker) (Richman
et al. 1980), and many others.
The population dynamics of bigeyed bugs in soybean fields has been found to vary
according to geographical location. Numbers were greatest in late August to early
September in Kentucky (Raney & Yeargan 1977), early August or mid-September in
North Carolina (Deitz et al. 1976), late September in South Carolina (Shepard et al.
1974a), and between late June and early August in Mississippi (Pitre et al. 1978). More
than one generation occurred in each case, as populations were present during most of
the growing season. No published information exists on the population dynamics of
bigeyed bugs in soybean in more southerly areas of the southern U.S. growing region.
The dispersion characteristics of bigeyed bug populations in South Carolina soybean
fields were investigated by Waddill et al. (1974). Populations were randomly distributed
on most sample dates, fitting the Poisson distribution. Pieters & Sterling (1973) re-
ported that their populations in cotton were clumped on most occasions, fitting the
negative binomial. Dispersion characteristics of insect populations have been shown to
vary with population density and sample-unit size (Wilson & Room 1982, 1983). There-
fore, additional information taken over a range of population densities and at a different
sample-unit size from the Waddill et al. (1974) study would provide valuable information
about the dispersion characteristics of their populations in soybean.
The primary purpose of the present study was to determine the temporal abundances
of bigeyed bug populations in soybean in Alabama and Florida. Their dispersion charac-
teristics over a range of population densities also were quantified. Such information will
allow for implementation of pest management practices that conserve these natural
control agents in soybean fields.


Bigeyed bug populations were monitored over two cropping seasons in soybean
fields located in Florida and Alabama. Each was a production field that was grown in
a manner typical for the agroecosystem, but none were treated with an insecticide that
would directly affect bigeyed bug populations. (Commercial preparations of Bacillus
thuringiensis were applied in some fields late in the growing season to suppress very
large outbreak populations of the A. gemmatilis.) Fields were disked before planting,
with no subsoiling. Planting dates ranged from late May to late June. Bigeyed bug
nymphal and adult populations were monitored in three fields in 1985. All fields were
in different counties of Alabama, which were Dallas, Elmore, and Henry counties (Fig.
1). Nymphal and adult populations were monitored in two fields in 1986. One field was
located in Dallas Co., Alabama and the other in Gadsden Co., Florida (Fig. 2). All fields
were at least 2 ha in area. Soybean varieties were 'Tracey M' for the Alabama fields
and 'Braxton' for the Florida field.
Nymphal and adult population density was estimated ca. weekly. Sampling began
at the early vegetative crop-growth stages and continued until late seed stages of crop

Florida Entomologist 70(4)

growth. An exception was the Dallas Co., Alabama field sampled in 1986. There, ex-
treme drought resulted in retardation of plant growth, and sampling was discontinued
in mid-August. Each field was divided into 12 parts of equal area, and four random
samples taken within each.
Sampling procedures were according to those established in previous studies as
being most appropriate for estimating nymphal and adult bigeyed bug populations in
soybean (Rudd & Jensen 1977, Shepard et al. 1974b, Turnipseed 1974). Descriptions of
methods for sampling their populations in soybean are contained in Irwin & Shepard
(1980). The ground cloth method was employed. Forty-eight, 1.8-m samples were taken
in each field on each sample date when their populations were in detectable numbers.
This sample-unit size differed from the 1.2-m size in the Waddill et al. (1974) study. For
each sample, the ground cloth was laid between two rows, the soybean plants from both
sides (0.9 m on each side) beaten onto the cloth, and the number of nymphs and adults
counted. Also, bases of the plants and adjacent soil were examined for bigeyed bugs.
The variance/mean ratio was calculated for adult and nymphal sample counts on each
sample date in each field (Southwood 1978). These analyses were performed by using
Myers' (1978) FORTRAN program. Log-transformed means and variances of adult and
nymphal sample counts then were calculated for each sample date in each field, and
Taylor's (1984) power law relationship for each field determined. A Taylor's power law
relationship also was determined for each year by combining data from all fields sampled
in that year. All Taylor's relationships were determined by using SAS programs (SAS
Institute 1982a,b).


The mean numbers ( SE) of adult and nymphal bigeyed bugs on each sample date
in each soybean field sampled in 1985 and 1986 are shown in Figs. 1 and 2, respectively.

7 -........ ELMORE CO., AL
W 6 -----DALLAS CO., AL .
u 5-...

S... .. ... ... '

a- 4-
a- 0 ."-"-", -- ..... -T- "- ---

0 -..----------- --I-----

6/18 6/287/8 7/18 7/28 8/7 8/17 8/27 9/6 9/16 9/26

Fig. 1. Mean numbers (+ SE) of adult and nymphal Geocoris spp. populations in
three soybean fields sampled in 1985 (Dallas, Elmore, and Henry Cos., Alabama).

December, 1987

Funderburk & Mack: Geocoris in Soybeans


0 ---c DALLAS CO., AL


soybean fields sampled in 1986 (Dallas Co., Alabama and Gadsden Co., Florida).

Overall precision of the sampling program, as determined by the SE values, was rarely
small. numbers ADULTS(i.e., < 5% of total).
O 3 ; .... .1 4 1.

O "". .. .. ...... t *-.-

numbers during the rest of the growing season in all but one of the fields. Adult and
o I I I,' ',--
6/13 7/3 7/23 8/12 9/1 9/21 10/1I


Fig. 2. Mean numbers (+ SE) of adult and nymphal Geocoris spp. populations in two
soybean fields sampled in 1986 (Dallas Co., Alabama and Gadsden Co., Florida).

Overall precision of the sampling program, as determined by the SE values, was rarely
over 25% of the mean except when populations were very small. Populations ofbigeyed
bugs were present in all soybean fields during most of the growing season. Nearly all
of the bigeyed bugs in all fields were G. punctipes, but G. uliginosis was noted in very
small numbers (i.e., < 5% of total).
Sparse populations of adults were first detected in the fields during middle to late
vegetative soybean growth stages. Usually, nymphal populations appeared about a
week or two later. Both adult and nymphal populations then were present in substantial
numbers during the rest of the growing season in all but one of the fields. Adult and
nymphal populations did not occur in large numbers in that field until late in the growing
season. In nearly all fields, generational cycles were indicated by peaks in nymphal
populations ca. every 30 days. There was considerable overlap between the generations
in all fields, because the number ofbigeyed bugs rarely declined greatly in magnitude.
Population size was small in the 1985 Dallas Co. and 1986 Dallas Co., Alabama fields,
compared to the other fields sampled in this study. The drought may be the reason for
the lower populations in 1986 Dallas Co., Alabama field.
Population size varied among fields, but population trends were similar. At least 3
complete generations occurred, with additional partial generations also present. The
number of bigeyed bugs increased through the growing season, with numbers usually
greatest on the last or near the last sample date.
For each sample date in the 1985 and 1986 soybean fields, the variance/mean ratio
of adult and nymphal populations are presented in Table 1. Variance/mean ratios of <1,
1, and >1 represent uniform, random, and clumped distributions, respectively (South-
wood 1978). According to this measure of dispersion, nymphal populations were aggre-
gated in 55.8% of the field/date data sets when data were sufficient for analysis. Adult
populations were aggregated 47.1% of the time. Values were >1, but not significantly
greater than 1, on most other dates for both nymphs and adults.

Florida Entomologist 70(4)

Co., FL.

Sample S Sample S
date Nymphs Adult date Nymphs Adults

1985 Dallas Co., AL
22 Aug. 1.02
5 Sept. 1.69*
13 Sept. 1.34*
17 Sept.

1985 Elmore Co., AL
21 June
28 June
3 July 1.32*
11 July 1.24
19 July 1.17
24 July 1.43*
1 Aug. 1.26
8 Aug. 1.53*
16 Aug. 1.76*
19 Aug. 1.83*
30 Aug. 2.01*
9 Sept. 2.90*
16 Sept. 3.65*

1985 Henry Co., AL
24 June 1.87*
1 July
8 July
15 July 1.28
22 July 1.34*
29 July 1.06
6 Aug. 1.12
13 Aug. 1.25
20 Aug. 1.67*
27 Aug. 3.14
10 Sept. 1.11

1986 Gadsden Co., FL
16 June
24 June
30 June
7 July 1.06
14 July 1.11
21 July 2.04*
28 July 2.97*
4 Aug. 1.96*
11 Aug. 1.63*
18 Aug. 1.41*
25 Aug. 1.36*
2 Sept. 1.23
8 Sept. 1.37*
17 Sept. 4.08*
24 Sept. 2.07*
3 Oct. 0.99

1986 Dallas Co., AL
27 June
4 July
11 July 0.97
18 July 1.16
25 July 1.20
1 Aug. 1.23
8 Aug. 1.34
15 Aug. 1.09







*Significantly different than 1 (P < 0.05) by X2 test.

Taylor's power law relates variance (s2), to mean density (m) by the relationship, s2
= amb. Taylor et al. (1978) considered the slope (b) to be a constant for a species (with
values of b<1, b=1, and b>l indicating uniform, random, and clumped distributions,
respectively) and the intercept (a) to be reflected by sample-unit size. Taylor's power
law is a particularly useful dispersion index, because it allows for a description of a
species distribution pattern as changing with density.
Regression statistics of Taylor's power law relationships for bigeyed bug nymphal
and adult sample estimates for each soybean field and for each year are shown in Table
2. For all relationships involving nymphal populations, b was statistically > 1 (P < 0.05)
according to students t tests. These values of b and the relatively high r2-values (mean
= 0.94 and 0.94 for 1985 and 1986, respectively) indicate that bigeyed bug nymphal

December, 1987

Funderburk & Mack: Geocoris in Soybeans


Nymphs Adults
Field and Year Intercept Slope r2 Intercept Slope r2

Dallas Co., AL 1985 -0.02 1.59* 0.95 -0.01 1.63 0.74
Elmore Co., AL 1985 -0.05 1.40* 0.98 0.05 0.95 0.59
Henry Co., AL 1985 -0.02 1.25* 0.89 -0.01 1.22* 0.94
All Fields 1985 -0.02 1.33* 0.96 0.01 1.16* 0.96
Dallas Co., AL 1986 0.00 1.16* 0.99 0.01 0.98 0.79
Gadsden Co., FL 1986 -0.07 1.43* 0.88 0.09 0.89 0.68
All Fields 1986 -0.04 1.40* 0.93 0.03 1.07 0.86

*Significantly > 1 (P < 0.05) according to a t test.

populations were aggregated over a wide range of population densities. Results were
less definitive for adult populations, with b statistically > 1 for 1 field. For 3 fields, b
was very near 1, strongly indicating a random distribution. The precision of the regres-
sion relationships was good, with mean r-values of 0.76 and 0.74 for 1985 and 1986,


Bigeyed bugs typically were common throughout most of the growing season in the
soybean fields sampled in this study, with greatest population densities usually occur-
ring near the end of the growing season. These results from Alabama and Florida are
similar to previously published studies conducted in other states in the southern U.S.
(Raney & Yeargan 1977, Dietz et al. 1976, Shepard et al. 1974a, Pitre et al. 1978).
Bigeyed bugs apparently have a similar phenology in soybean throughout the region.
Bigeyed bugs obviously are important, indigenous natural enemies of soybean pests in
the southern U.S. Enhancement and conservation of their populations should be a major
priority in soybean IPM programs.
Nymphal populations were clumped in the fields sampled in this study, which con-
trasts with the conclusion drawn by Waddill et al. (1974) that populations were random
and best described by the poisson distribution. Differences between the studies may
indicate that dispersion characteristics of bigeyed bug nymphs change as a function of
population density and/or sample-unit size. However, our results from a wide-range of
densities in a number of fields over two growing seasons would be a strong indication
that clumping is typical for nymphal populations. Adult populations usually were ran-
domly distributed, but sometimes clumped. These results corroborate the conclusions
drawn by Waddill et al. (1974) that their populations are best described as randomly


We thank Andrew Brown and Dale Spurgeon for assistance in sampling. Assistance
of Carole Backman in data collation and analysis also is greatly apperciated. Paper of
the Fla. Agric. Exp. Sta. Journal Series No. 8242.


Florida Entomologist 70(4)


CROCKER, R. L., AND W. H. WHITCOMB. 1980. Feeding niches of the big-eyed bugs
Geocoris bullatus, G. punctipes, and G. uliginosus (Hemiptera: Lygaeidae:
Geocorinae). Environ. Entomol. 9: 508:513.
AND R. E. STINNER. 1976. A guide to the identification and biology of soybean
arthropods in North Carolina. North Carolina Agric. Exp. Sta. Bull. 238. 264 pp.
ELVIN, M. K., J. L. STIMAC, AND W. H. WHITCOMB. 1983. Estimating rates of
arthropod predation on velvetbean caterpillar larvae in soybeans. Florida En-
tomol. 66: 320-330.
IRWIN, M. E., AND M. SHEPARD. 1980. Sampling predaceous Hemiptera on soybean,
pp. 505-531. In: M. Kogan and D. C. Herzog (ed.) Sampling Methods in Soybean
Entomology. Springer-Verlag, Inc. New York.
MCDANIEL, S. G., ANDW. L. STERLING. 1979. Predator determination and efficiency
on Heliothis virescens eggs in cotton using 2P. Environ. Entomol. 8: 1083-1087.
MYERS, J. H. 1978. Selecting a measure of dispersion. Environ. Entomol. 7: 619-621.
NARAJO, S. E., AND J. L. STIMAC. 1985. Development, survival, and reproduction
of Geocoris punctipes (Hemiptera: Lygaidae): Effects of plant feeding on soybean
and associated weeds. Environ. Entomol. 14: 523-530.
PIETERS, E. P., AND W. L. STERLING. 1973. Inferences on the dispersion of cotton
arthropods in Texas. Environ. Entomol. 2: 863-867.
Beneficial arthropods on soybean and cotton in different ecosystems in Missis-
sippi. Miss. Agric. For. Exp. Sta. Tech. Bull. 90. 9 pp.
RANEY, H. G., AND K. V. YEARGAN. 1977. Seasonal abundance of common
phytophagous and predaceous insects in Kentucky soybeans. Trans. Kentucky
Acad. Sci. 38: 83-87.
RICHMAN, D. B., R. C. HEMENWAY, AND W. H. WHITCOMB. 1980. Field cage
evaluation of predators of the soybean looper, Psuedoplusia includes (Lepidopt-
era:Noctuidae). Environ. Entomol. 9: 315-317.
RUDD, W. G., AND R. L. JENSEN. 1977. Sweep net and ground cloth sampling for
insects in soybean. J. Econ. Entomol. 70: 301-304.
SAS INSTITUTE. 1982a. SAS user's guide: statistics. SAS Institute, Cary, N. C.
1982b SAS user's guide: basics. SAS Institute, Cary, N. C.
SHEPARD, M., G. R. GARNER, AND S. G. TURNIPSEED. 1974a. Seasonal abundance
of predaceous arthropods in soybeans. Environ. Entomol. 3: 985-988.
SHEPARD, M., G. R. GARNER, AND S. G. TURNIPSEED. 1974b. A comparison of
three sampling methods for arthropods in soybeans. Environ. Entomol. 3: 227-
SOUTHWOOD, T. R. E. 1978. Ecological methods, 2nd ed. Wiley, New York.
TAYLOR, L. R., I. P. WOIWOOD, AND J. N. PERRY. 1978. The density dependence
of spatial behavior and the rarity of randomness. J. Anim. Ecol. 47: 383-406.
TAYLOR L. R. 1984. Assessing and interpreting the spatial distributions of insect
populations. Annu. Rev. Entomol. 29: 321-357.
TUGWELL, P., E. P. ROUSE, AND R. G. THOMPSON. 1973. Insects in soybeans and
a weed host (Desmodium sp.). Arkansas Agric. Exp. Stn. Rep. Ser. 214. 18 pp.
TURNIPSEED, S. G. 1972. Management of insect pests of soybean. Proc. Tall Timbers
Conf. Econ. Anim. Control Habitat Manage. 1: 189-203.
TURNIPSEED, S. G. 1974. Sampling soybean insects by various D-Vac, sweep, and
ground cloth methods. Florida Entomol. 57: 219-223.
Sequential sampling plans for Nabis spp. and Geocoris spp. in soybeans. Environ.
Entomol. 3: 415-419.
WHITCOMB, W. H., AND K. BELL. 1964. Predaceous insects, spiders and mites of
Arkansas cotton fields. Arkansas Agric. Exp. Stn. Bull. 690: 1-84.

December, 1987

Thorvilson et al.: Pyemotes vs. Fire Ants 439

WILSON, T. P., AND P. M. ROOM. 1982. The relative efficiency and reliability of three
methods for sampling arthropods in Australian cotton fields. J. Australian En-
tomol. Soc. 21: 175-181.
1983. Clumping patterns of fruit and arthropods in cotton with implications
for binomial sampling. Environ. Entomol. 12: 50-54.


Department of Entomology, Texas Tech University
Lubbock, Texas 79409

American Farm Bureau Federation,
Natural and Environmental Resources
225 Tougy Ave., Park Ridge, Illinois 60058


Texas Department of Agriculture, P.O. Box 12847
Austin, Texas 78711


The straw itch mite, Pyemotes tritici (Lagreze-Fossat and Montan6), is a parasitic
pyemotid mite that has generated considerable interest as a biological control agent of
the red imported fire ant, Solenopsis invicta Buren. To evaluate the efficacy of this
mite against S. invicta under the conditions of central Texas, a large-scale field trial
was conducted during 1985. A total of 244 S. invicta colonies was mapped and colony
size and activity ratings recorded. During May, two applications of mites cultured on
Angoumois grain moth-infested wheat kernels were introduced into 103 colonies, and
141 control colonies were treated with Angoumois grain moth-infested wheat kernels
not infested with mites. The colonies were also twice treated in October. Throughout
the test period, colonies were periodically inspected and rated. Analysis of the data
using t-tests did not reveal significant differences in ratings between mite-treated and
control colonies. Therefore, the straw itch mite, P. tritici, was not found to be an
effective biological control agent of the red imported fire ant.

El acaro de la picaz6n de la paja,Pyemotes tritici (Lagreze-Fossat and Montane), es
un acaro pyem6tido parasitico que ha generado considerable interns como agent de
control biol6gico para la hormiga de fuego roja importada, Solenopsis invicta Buren.
Para evaular la efficiencia del acaro sobre S. invicta, bajo las condiciones del centro de
Texas, un ensayo a larga escala se condujo en el campo durante 1985. Un total de 244

Florida Entomologist 70(4)

colonies de S. invicta se tazaron, tanto on en tomafio como en actividad, y se registraron
en un mapa. Durante el mes de mayo dos aplicaciones del acaro fueron introducidas en
cultivos de granos de trigo infestados con la alevilla de grano de Angoumois. Unas 103
colonies fueron tratadas, mientras que las otras 141 sirvieron de testigo. Las colonies
fueron tratadas dos veces en octubre. Durante el period de prueba las colonies fueron
inspeccionadas y tazadas periodicamente. El analisis de los datos utilizando la prueba t
no revel6 diferincias significativas entire colonies tratadas y colonies que sirvieron de
control. Por lo tanto, el acaro de la picaz6n de la paja, P. tritici, no se encontr6 que sea
un agent de control biol6gico efectivo para la hormifa de fuego roja importada.

The straw itch mite, Pyemotes tritici (Lagreze-Fossat and Montan6), is a parasitic
pyemotid mite that has generated considerable interest as a biological control agent of
stored grain pests (Moser 1975, Bruce & LeCato 1979, Bruce 1983). In addition, this
mite was found to parasitize all life stages of the red imported fire ant, Solenopsis
invicta Buren (Bruce & LeCato 1979, 1980, Bruce 1983). Field trials in Georgia and
Florida using P. tritici as a treatment "inactivated" more than 50% of the S. invicta
mounds (Bruce & LeCato 1980, Bruce 1983). However, a relatively small number of
colonies were treated in their study and some of the mounds received 10 (predeter-
mined) mite applications.
In Brazil, Flechtmann (1981) obtained P. tritici from the cowpea weevil, Callosob-
ruchus maculatus (F.), reared on chickpea, Cicer arietinum. When a large number of
gravid female mites was present in the culture, all the material (grain, insects and
mites) was introduced into the nest entrance of a two-year old colony of the leaf-cutting
ant, Atta sexdens rubropilosa Forel. After two or three days, all ant nest activity
ceased, and remained inactive for 16 days. However, after this period the colony recup-
erated and resumed activity (Flechtmann 1981).
Three rates of P. tritici (ca. 5.0-7.5, 10.0-15.0, and 20.0-30.0 mites/ant) were applied
to S. invicta mounds at each of three sites in Florida during May and June (Jouvenaz
& Lofgren 1986). Single applications of mites caused no significant reduction of the S.
invicta population index, although the highest rate may have initiated colony movement.
Despite the lack of efficacy in some tests for this mite to control S. invicta in limited
laboratory and field trials, and in spite of human dermatitis associated with P. tritici
(Moser 1975), the straw itch mite has been touted as a biological control agent for S.
invicta. In response to continuing promotion of P. tritici ("Fire Mites"; Biofac, Inc.,
Mathis, Texas), a large-scale field study was initiated in cooperation with the Texas
Department of Agriculture. The objective of this study was to determine the effect of
P. tritici on S. invicta colonies.


Three pasture sites (separated by several km) in Bandera Co., Texas, were chosen
for the study. At each site, randomly selected S. invicta colonies were mapped and
labeled with wooden stakes. The top of each mound was opened with a shovel, and a
pretreatment, weighted numerical rating based on worker numbers and brood charac-
teristics (Lofgren & Williams 1982) was recorded for each mound. All colonies in the
study were queen-right (worker brood present). Randomly selected colonies were
treated 16-17 May 1985, at two of the three sites, with wheat kernels infested by
Angoumois grain moth larvae that were parasitized by straw itch mites. Site A con-
tained 34 mite-treated and 39 control colonies; site B contained 61 mite-treated and 83
control colonies; site C contained 8 mite-treated and 19 control colonies. Control treat-
ments consisted of unparasitized Angoumois grain moth larvae on wheat kernels. All
materials were transported to the study sites in insulated styrofoam containers and


December, 1987

Thorvilson et al.: Pyemotes vs. Fire Ants

mite activity was determined by visual inspection. Mite-infested and control materials
were supplied by Biofac, Inc., Mathis, Texas and treatments were made within 24 h of
receipt from the company.
Introduction of the test and control materials into colonies was done according to
instructions from Biofac, Inc. and was as follows: six shallow holes (ca. 7.5 cm deep and
5.0 cm dia.) were punched into the top of each mound. Sixty-eight ml of wheat kernels
or wheat kernels with P. tritici were poured into the holes of each mound and covered
with soil. Each 68 ml of mite-infested wheat contained ca. 60,000 mites. Two wk after
first treatment (30 May), all colonies were again individually rated (Lofgren & Williams
1982) and the treatment repeated. Air temperature and mound soil temperature during
applications ranged from 220 to 33 C and from 260 to 300 C, respectively, and relative
humidity was greater than 50%. These conditions were within the limits established by
Bruce (1984) for laboratory survival and reproduction of P. tritici adult females reared
on cigarette beetle pupae. Treatments were applied between 0830 and 1215 C.D.S.T.
The condition of the colonies was also evaluated 4, 6, and 12 wk after first treatment
(summer trial: 13, 27 June and 8 Aug.). Because S. invicta colonies often move in
response to repeated disturbance (Williams & Lofgren 1983, Jouvenaz & Lofgren 1986),
a new colony that appeared within ca. 2 m of an abandoned, treated mound was assumed
to have relocated and subsequent observations were directed toward the new mound.
However, colony movement was infrequently observed during the present study.
These same colonies were rated and retreated in an identical manner at 20 and 30
wk after first treatment (fall trial: 4 and 18 Oct.). In addition, these colonies were rated
33 and 35 wk after first treatment (8 and 21 Nov.) The third location (site C), consisting
of 8 mite-treated and 19 control colonies, was rated and treated on 4 and 18 Oct. Final
readings for this site were also taken on 8 and 21 November. All 244 colonies in the
three sites (103 mite-treated and 141 control colonies) were evaluated through 2 Nov.
1985. Mean colony ratings between and within treatments at each site were compared
through time and analyzed by t-tests.
In addition, worker ants were collected 29 May from 28 colonies at field sites A and
B by placing a white, 7.7 x 12.7 cm card onto each mound and then brushing the ants
into a labeled vial containing alcohol. Later, a random sample of 25 ants was removed
from each vial and the maximum head capsule width of each ant was measured with an
ocular micrometer of a dissecting microscope calibrated to 0.024 mm. The mean head
capsule width and the standard error were calculated for each sample. Head capsule
data were compared with those of Greenberg et al. (1985) for monogynous and polygyn-
ous S. invicta colonies.


Few significant differences (P<0.05) were detected posttreatment between mite-
treated and control colonies (Table 1). The differences that were detected indicated that
mite-treated colonies had a higher mean colony rating than did control colonies (site A:
12 wk: summer trial; site B: 35 wk: fall trial). No significant differences were detected
in colony ratings at site C.
As expected, mean colony ratings declined from May through November as a result
of normal phenological fluctuations resulting from changing seasons. Mean ratings of
mite-treated colonies at site A were significantly reduced (P<0.05) from 0 wk (initial
application) to 12 wk posttreatment (summer trial) and from 20 wk (subsequent applica-
tion) to 35 wk posttreatment (fall trial); control colonies significantly (P<0.05) declined
only during the former time period (summer trial). In contrast, the control colonies at
site B significantly (P<0.05) declined from 20 wk (subsequent application) to 35 wk
posttreatment (fall trial). At both sites A and B, mean colony ratings at 35 wk for

442 Florida Entomologist 70(4) December, 1987

M M 0 C

M 04 00 I n

z ao o-4-

C6 2 C;
SP 0 a

-4 rI ut COO cc t 0-

.- o oo l tCQ
Mo E
)W CC 'rI


O IO -'

a a 2

4-t- II -O

to o-o

t -ia>C ^f S 3 -^CS 'S oo i a S- <

^S c -2S a
^ 6 a s

Thorvilson et al.: Pyemotes vs. Fire Ants

mite-treated and control colonies were significantly (P<0.05) less than those at 0 wk.
No significant differences were detected through time within either treatment (test or
control) at site C.
Percent of colonies having reproductive brood at both the beginning and end of the
summer trial (13 June-8 Aug.) for site A was 62% for the control colonies and 82% for
the mite-treated colonies. Also, at site B, a higher percent of mite-treated colonies had
reproductive brood (92%) at both the beginning and end of this trial, compared with
reproductive brood in the control colonies (82%). However, neither mite-treated nor
control colonies at any of the three sites were found to have reproductive brood at both
the beginning and end of the fall trial (4 Oct.-21 Nov.). This lack of reproductive brood
would normally be expected during the fall.
Mean head capsule width measurements of worker ants from 28 colonies ranged
from 0.64 mm to 1.02 mm. Although worker head capsule widths have been used as
predictors of monogyny and polygyny, the large variances in mean widths in the present
study prevented the classification of colonies based on the criteria of Greenberg et al.
In summary, the methods and procedures of our study did not detect the straw itch
mite, P. tritici, as an effective control agent for red imported fire ants, S. invicta, in
central Texas. In fact, based on colony ratings and the presence or absence of reproduc-
tive brood, mite-treated colonies appeared more viable than control colonies. In addi-
tion, several researchers experienced severe dermatitis from the straw itch mite during
the course of this study. Although Bruce & LeCato (1980) and Bruce (1983) obtained
good control of S. invicta with P. tritici, results could have been indicative of S. invicta
colony movement. Williams & Lofgren (1983) state that relocation resulting from re-
peated examinations and ineffectual treatment is a common phenomenon in S. invicta
studies. Therefore, since repeated applications of P. tritici did not control S. invicta in
this study, we concur with Jouvenaz & Lofgren (1986) that the straw itch mite (P.
tritici) is not efficacious in the control of the red imported fire ant.


We thank Leland Chandler, Robert W. Sites, and James C. Cokendolpher for their
helpful comments on the manuscript and Gerardo Camilo for the Spanish abstract. We
also thank Jane Allert, Willie Kalka, and Raymond Hicks of Bandera, Texas, for allow-
ing us to conduct this study on their land. Personnel from the Texas Tech University
Center at Junction are gratefully acknowledged for making those facilities available to
project members. Contribution No. T-10-177, College of Agricultural Sciences, Texas
Tech University, Lubbock, Texas. Supported by Interagency Agreement # 1753 (84-
85), Texas Department of Agriculture.


BRUCE, W. A. 1983. Mites as biological control agents of stored product pests, pp.
74-78. In Hoy, M. A., G. L. Cunningham, and L. Knutson (eds.), Biological
Control of Pests by Mites. Proceedings of a conference held April 5-7, 1982 at
the Univ. of California, Berkeley. Special Publ. 3304. University of California.
185 pp.
BRUCE, W. A. 1984. Temperature and humidity: Effects on survival and fecundity of
Pyemotes tritici (Acari: Pyemotidae). Internat. J. Acarol. 10: 135-138.
BRUCE, W. A. AND G. L. LECATO. 1979. Pyemotes tritici: Potential biological control
agent of stored-product insects. pp. 213-220. In Rodriquez, J. G., (ed.), Recent
Advances in Acarology, Vol. 1. Academic Press, New York. 631 pp.

Florida Entomologist 70(4)

December, 1987

BRUCE, W. A. AND G. L. LECATO. 1980. Pyemotes tritici: A potential new agent for
biological control of the red imported fire ant, Solenopsis invicta (Acari:
Pyemotidae). Internat. J. Acarol. 6: 271-274.
FLECHTMANN, C. H. W. 1981. Um possivel agent de control biologico da sauva.
Resumos, VII Congresso Brasileiro do Entomol. (Fortaleza, Ceara): 1.
GREENBURG, L., D. J. C. FLETCHER, AND S. B. VINSON. 1985. Differences in
worker size and mound distribution in monogynous and polygynous colonies of
the fire ant, Solenopsis invicta Buren. J. Kansas Entomol. Soc. 58: 9-18.
JOUVENAZ, D. P. AND C. S. LOFGREN. 1986. An evaluation of the straw itch mite,
Pyemotes tritici (Acari: Pyemotidae), for control of the red imported fire ant,
Solenopsis invicta (Hymenoptera: Formicidae). Florida Entomol. 69: 761-763.
LOFGREN, C. S. AND D. F. WILLIAMS. 1982. Avermectin Bla: A highly potent in-
hibitor of reproduction by queens of the red imported fire ant (Hymenoptera:
Formicidae). J. Econ. Entomol. 75: 798-803.
MOSER, J. C. 1975. Biosystematics of the straw itch mite with special reference to
nomenclature and dermatology. Trans. R. Ent. Soc. London. 127: 187-191.
WILLIAMS, D. F. AND C. S. LOFGREN. 1983. Imported fire ant (Hymenoptera: For-
micidae) control: Evaluation of several chemicals for individual mound treat-
ments. J. Econ. Entomol. 76: 1201-1205.

--c- --^- -*- -*- --- -- >- ^


Department of Agriculture
Apartado Postal 185
Guasave, Sinaloa, Mexico 81000


The impact of six species of parasites and a polyhedrosis virus on larvae of beet
arymworm, Spodoptera exigua (Hiibner), was investigated in processing tomato in the
Guasave and Del Fuerte Valleys of Sinaloa, Mexico. The Polyhedrosis virus was the
most important mortality factor; several epizootics occurred during the course of the
study. Cotesia marginiventris (Cresson) was the most common parasite, followed by
Meteorus laphygmae Viereck and Pristomerus spinator (F.). Disease and parasitization
reduced beet armyworm to non-injurious levels during the seedling stage of the crop
and thus should be considered as important components of the recently implemented
IPM program on processing tomatoes in this area.


Se investig6 el impact de seis species de parasitos y de un virus poliedrosis en
larvas del gusano soldado, Spodoptera exigua (Hubner), en tomate industrial en los
valles de Guasave y Del Fuerte en Sinaloa, Mexico. El poliedrosis virus fue el factor de
mortalidad mds important; various factors epiz6ticos ocurrieron durante el curso de


Alvarado-Rodriguez: Beet Armyworm Natural Enemies

este studio. Cotesia marginiventris (Cresson) fue el parasito mas comin, seguido por
Meteorus laphygmae Viereck y Pristomerus spinator (F.). Enfermedades y parasitos
reducieron a niveles no-daninos los gusanos soldado durante el estado de plants de
semillero del cultivo, por lo tanto deben de ser considerados como components impor-
tante del recientemente implementado program de Manejo Integral de Plagas en tomato
industrial en esta area.

Beet armyworm, Spodoptera exigua (Hiibner) (Lepidoptera: Noctuidae), is a
polyphagous species that inflicts economic damage on several crops including corn,
wheat, soybean, sorghum and tomato. This insect is among the most important pests
of processing tomato in the Guasave and Del Fuerte Valleys of Sinaloa, M6xico (Anon-
ymous 1983). Larvae usually feed on foliage. However, they may also damage fruit.
The tolerance for direct damage to processing fruit is only 4%. Broad-spectrum insect-
icides are currently the primary means for reducing beet armyworm damage to tomato
fruits, as well as for control of other lepidopterous pests. In view of this, Campbell's
de Mexico at Sinalopasta-Guasave initiated the development of an integrated pest man-
agement (IPM) program in processing tomatoes. One aspect of the program was to
determine which parasites and pathogens attack S. exigua larvae during the tomato
growing season. The results of these investigations are reported here.


The study was conducted on five commercial fields located at Teresita, Palotal, Ruiz
Cortines, Chicorato and Chino in the Guasave and Del Fuerte Valleys of Sinaloa,
Mexico. The fields were planted on different dates during January 1986; they would be
considered late-season plantings for the region. The fields in Palotal and Chicorato were
transplanted with seedlings of UC-82B in single rows at a density of 3 plants per meter.
For the others, the same cultivar was direct seeded (6 plants per meter) in double rows
0.3 m apart on raised beds 1.1 m. wide. All fields were produced according to recom-
mended horticultural practices for the area (Anonymous 1983).
The field in Chicorato had a regular-based insecticide spray program (total of 10
applications). Applications of broad-spectrum insecticides were made biweekly from
transplanting until harvest. Insecticides used included methomyl (Lannate 90 WP, 0.36
kg/ha), fenvalerate (Pydrin 30 EC, 0.12 It/ha) and metamidophos (Monitor 60 EC, 0.6
It/ha). As for the other fields, these insecticides were applied only as needed (approxi-
mately five sprays), according to a monitoring system that was developed in California
(Lange et al. 1985, Weakly et al. 1986) and recently implemented through an IPM
program in this region. First application on these fields were made at the fruit set.
S. exigua (mostly 2nd to 4th instars) were collected monthly from seedling to fruit
maturation. The randomly selected plants for sampling fruitworm eggs [sample consists
in selecting the leaf below the highest open flower on the tomato plant, while crossing
the fields in a diagonal pattern, as described by Lange et al. (1985) and Weakly et al.
(1986)], were also used as sites for sampling S. exigua larvae. A total of 30 plants for
each field at each sampling date were entirely examined for presence of S. exigua
larvae. The insects found were placed in a paper bag previously filled with tomato
leaves as a way to reduce larval mortality due to handling and taken to laboratory for
rearing. Each larva was individually placed in a 30 cm3 plastic cup containing pinto bean
diet and held in a rearing room at 26 3C (Shorey and Hale 1965). Number of larvae
collected varied between sampling dates (10 to 128) depending on field infestation, as
the insect population ranged from 0.4 to 4.2 larvae per plant. Data on emergence of
moths, parasites and larvae infected with pathogens were recorded.

Florida Entomologist 70(4)


Six parasites and a nucleopolyhedrosis virus (NPV) were identified as biotic factors
affecting S. exigua larvae. The parasites were: Cotesia marginiventris (Cresson)
(Braconidae), Meteorus laphygmae Viereck (Braconidae) and Chelonus insularis Cres-
son (Braconidae); Pristomerus spinator (F.) (Ichneumonidae) and Campoletis
flavicincta (Ashmead) (Ichneumonidae) and a Lespesia sp. (Tachinidae).
Host mortality at the various sites are presented in Table 1. Overall, there was an
increase in mortality as the growing season progressed. The mean percentage of mortal-
ity from all sites for each month was lowest in January (27%) intermediate in February
(72%) and March (66%), and highest in April (74%). These findings are in accordance
with those reported by Pacheco-Mendivil (1985) on cotton in Sonora, Mexico and Oatman
et al. (1983) on summer plantings of fresh market tomatoes in California. The former
author suggested that efficiency of the biocontrol agents is dependent upon increases
in temperature as the season progresses.
Infection with NPV was detected at all sites and samples dates (Table 1) and was
the major mortality factor affecting S. exigua larvae. A mean of 30.2% of collected
larvae died due to infection with NPV. Moreover, four epizootics of the virus (i.e., 75
to 100% larval mortality), were documented: Ruiz Cortinez (February 13), Chicorato,
(February 28 and April 12) and Chino (February 26). Epizootics are reported to be
especially common late in the season in northern California (Bisabri-Ershadi & Ehler
1981); however, in the present study they occurred at different times during the season.
Relative abundance of parasites reared from S. exigua larvae is summarized in Table
2. C. marginiventris was the most prevalent parasite and accounted for highest overall
mean percentage of parasitization (17.9%). It was followed by P. spinator and M.
laphygmae averaging 9.2 and 9.0% parasitization, respectively. C. insularis and C.


Percent Mortality
Larvae Parasiti- Poly- Unknown
Site Date collected zation hedrosis cause Total

Teresita 21 Jan. 37 16.2 10.8 32.4 27.0
9 Feb. 40 52.5 15.0 15.0 67.5
10 Mar. 60 53.2 16.6 13.3 69.8
Palotal 6 Feb. 19 20.9 47.3 10.5 68.2
6 Mar. 21 19.0 33.3 14.2 52.3
16 Apr. 71 26.6 42.2 25.3 68.8
Ruiz Cortinez 13 Feb. 18 0 77.7 5.5 77.7
8 Mar. 60 49.9 20.0 13.3 69.9
11 Apr. 92 35.6 19.5 26.0 55.1
Chicorato 20 Feb. 35 34.2 11.4 22.8 45.6
28 Feb. 10 0 90.0 10.0 90.0
25 Mar. 128 63.0 14.0 17.1 77.0
12 Apr. 12 16.6 83.4 0 100.0
Chino de 26 Feb. 20 5.0 80.0 15.0 85.0
los Lopez 20 Mar. 18 27.7 33.3 5.5 61.0
14 Apr. 31 48.5 9.6 41.9 58.1
21 Apr. 14 35.5 50.0 14.5 85.5

December, 1987

Alvarado-Rodriguez: Beet Armyworm Natural Enemies 447


Cotesia Pris- Meteorus Chelonus Cam-
margin- tomerus laphy- in- poletis Lespesia
Site Date iventris spinator gmae sularis flavicincta sp.

Teresita 21 Jan. 16.2 0 0 0 0 0
9 Feb. 30.0 10.0 12.5 0 0 0
10 Mar. 26.9 10.0 13.3 0 6.6 0
Paalotal 6 Feb. 15.7 0 5.2 0 0 0
6 Mar. 9.5 0 9.5 0 0 0
16 Apr. 2.8 1.4 14.0 7.0 0 1.4
Ruiz Cortinez 13 Feb. 0 0 0 0 0 0
8 Mar. 11.6 20.0 10.0 3.8 5.8 0
11 Apr. 3.2 14.1 3.2 7.6 4.3 3.2
Chicorato 20 Feb. 31.4 0 2.8 0 0 0
28 Feb. 0 0 0 0 0 0
25 Mar. 48.4 1.5 8.5 3.1 1.5 0
12 Apr. 8.3 8.3 0 0 0 0
Chino de 26 Feb. 5.0 0 0 0 0 0
los Lopez 20 Mar. 22.2 0 5.5 0 0 0
14 Apr. 19.8 3.2 9.6 6.4 9.6 0
21 Apr. 0 14.2 14.2 7.1 0 0

flavicincta accounted for 5.8 and 5.6% parasitization, respectively. Lespesia sp. was
least common; it was recovered from only two samples in this study. As season progress-
ed, parasitization of S. exigua larvae increased. In general, higher total percent parasiti-
zation was correlated with increased number of parasite species recovered.
Interaction between parasites and NPV may produce increased mortality response
in host insects. In this study, a higher mortality of S. exigua larvae was generally
observed due to the complimentary action of parasites and NPV in an additive way
rather than a synergistic one. Such an interaction can also be beneficial when female
parasites reared from NPV-infected hosts serve to spread the pathogen (Irabagon &
Brooks, 1974). However, these authors also reported that interactions can be highly
detrimental to the parasite especially when parasites could not complete development
in diseased hosts. Therefore, all possible outcomes resulting from the interaction be-
tween the parasites and NPV with S. exigua larvae in the processing tomatoes should
be investigated so that more rational use of the agents can be made.
Natural biological control of S. exigua was especially significant early in the season.
The directed seeded fields of Chino, Palotal, Ruiz Cortinex and Teresit, which were
under on-demand spray program, were severely infested with S. exigua. Although as
many as 12 larvae per meter were counted at the seedling stage (before thinning), the
mortality factors reported here, in concert with timely fertilization, cultivation, and
irrigation cultural practices (promoting a more rapid vegetative growth after thinning),
reduced the larval density to levels that did not adversely affect the plant stands. The
effects of abiotic factors as well as of predators on larval populations might be involved,
but they were not measured in this study. A reduced feeding characterized by small
holes on leaflets by parasitized larvae was generally observed. These observations are
consistent with those of Grant & Shepart (1984) and Cobb et al. (1985).
Parasites are less effective control agents of insect pests usually when insecticides
are improperly used (Oatman & Kennedy 1976, Johnson et al. 1980 a, b). However, in

Florida Entomologist 70(4)

the field of Chicorato where ten sprays were applied, parasitization ranged from 16.6
to 63%. It is possible to enhance parasitization through a more proper use of insecticides.
Parasites may be fairly abundant even when moderate numbers of insecticide sprays
must be used as indicated by IPM monitoring programs.
Future research should be focused on life table studies in order to fully assess
generation or real mortality (Bisabri-Ershadi & Ehler 1981, Ehler 1977) rather than
partial or apparent mortality as reported here. In addition, the role of natural enemies
in reducing direct damage to fruit needs to be evaluated. This will pose special problems
because even parasitized larvae feed on fruit and cocoons of parasites can result in
reductions in fruit quality.


I am indebted to the following individuals for identifying parasites: R. W. Carlson
and P. W. Marsh, Biosystematics and Beneficial Insects Institute, USDA, Beltsville,
Maryland; and J. C. Hall and J. Lukman, Dept. of Entomology, University of California,
Reiverside. I also thank the staff of Agriculture Dept. SINALOPASTA, at Guasave,
Sin., for their assistance in the field collection of insect samples. Finally, I am grateful
to Lester E. Ehler for his valuable suggestions to improve the manuscript.


ANONYMOUS. 1983. Guia para la asistencia tecnica agricola Valle Del Fuerte. Sec-
retaria de Agricultura y Recursos Hidraulicos. INIA-CIAPAN-CAEVF. 234 pp.
BISABRI-ERSHADI, B., AND L. E. EHLER. 1981. Natural biological control of west-
ern yellow-striped armyworm, Spodoptera praefica (Grote), in hay alfalfa in
northern California. Hilgardia. 49: 1-23.
COBB, C. H., J. E. GRANT, AND M. SHEPARD. 1985. Effect of parasitism by Micro-
plitis demolitor (Hymenoptera: Braconidae) on foliage consumption by Heliothis
zea (Lepidoptera: Noctuidae) larvae. Florida Ent. 68: 490-3.
EHLER, L. E. 1977. Natural enemies of cabbage looper on cotton in the San Joaquin
Valley Hilgardia. 45: 73-106.
GRANT, J. F., AND M. SHEPARD. 1984. Laboratory biology ofMeteorus autographae
(Hymenoptera: Braconidae), an indigenous parasitoid of soybean looper
(Lepidoptera: Noctuidae) larvae. Environ. Ent. 13: 838-842.
IRABAGON, T. A., AND W. M. BROOKS. 1974. Interaction of Campoletis sonorensis
and a nuclear polyhedrosis virus in larvae of Heliothis virescens. J. Econ. En-
tomol. 67: 229-231.
JOHNSON, M. W., E. R. OATMAN, AND J. A. WYMAN. 1980 a. Effects of insecticides
on populations of the vegetable leafminer and associated parasites on summer
pole tomatoes. J. Econ. Entomol. 73: 61-6.
1980 b. Effects of insecticides on populations of the vegetable leafminer and
associated parasites on fall pole tomatoes. J. Econ. Entomol. 73: 67-71.
G. ZALOM. 1985. Integrated pest management for tomatoes. Univ. California
Div. Agric. Nat. Res. Publ. 3274: 33-37.
OATMAN, E. R., AND G. C. KENNEDY. 1976. Methomyl induced outbreak by
Liriomyza sativa on tomato. J. Econ. Entomol. 69: 667-8.
JOHNSON, AND H. W. BROWING. 1983. Parasitization of lepidopterous pests
on fresh market tomatoes in southern California. J. Econ. Entomol. 75: 452-5.
PACHECO-MENDIVIL, F. 1985. Plagas de los cultivos agricolas en Sonora y Baja
California. Secretaria de Agricultura y Recursos Hidraulicos. INIA-CIANO. 414

December, 1987

Opp & Prokopy: Seasonal Activity of Apple Maggot Fly 449

SHOREY, H. J., AND R. L. HALE. 1965. Mass rearing of larvae of nine noctuid species
on a simple artificial media. J. Econ. Entomol. 58: 522-524.
WEAKLEY C. V., F. G. ZALOM, AND L. T. WILSON. 1986. Worm sampling methods
for processing tomatoes. California Tomato Growers. 29: 4-8.


Department of Entomology
Fernald Hall
University of Massachusetts
Amherst, MA 01003


We observed individually marked, wild male and female apple maggot flies (AMF)
on a host apple tree for 24 days. Although we resighted a fairly high proportion (30.6%)
of 599 marked flies, we observed most flies (74.9%) only during the first week after
marking and release. Following the onset of reproductive maturity, as evidenced by
mating and oviposition, we saw males over more consecutive days than females. This
presumably occurred due to male arrestment following contact with female marking
pheromone deposited on fruit. Peak diel time of mating by unmarked flies corresponded
more closely with peak time of observation of marked males rather than marked fe-
males. We estimate some adult AMF may live up to 4 weeks in nature.


Nosotros observamos por 24 dias a machos y hembras salvajes de la manzana (AMF)
marcadas individualmente en un arbol hospedero de manzanas. Aunque vimos de nuevo
una proporci6n bastante alta (30.6%) de 599 moscas marcadas, observamos la mayoria
de las moscas (74.9%) solamente durante la primera semana despues de marcadas y
sueltas. Seguido al comienzo de madurez reproductive, demonstrado por el
apareamiento y la puesta de huevos, nosotros vimos los machos mas dias consecutivos
que las hembras. Esto corri6 presumiblemente debido a la cohibici6n de los machos
despues del contact con feromonas depositadas por hembras en la fruta. El auge del
tiempo de apareamiento de las moscas no marcadas correspondi6 mis al auge del tiempo
de observaci6n de machos marcados que a hembras marcadas. Nosotros estimamos que
algunos adults de la mosca de la manzana pudieran vivir hasta 4 semanas en la

The apple maggot fly (AMF), Rhagoletis pomonella (Walsh), is a well-known pest
of apples in northeastern North America and, in recent years, has been detected in
many western regions, including California (Joos et al. 1984). This fruit-parasitic te-
phritid fly has attained its pest status primarily due to expansion of its host range from

Florida Entomologist 70(4)

December, 1987

the native host, hawthorn (Crataegus spp.), to fruits more desirable for human con-
sumption, such as apple, pear, and sour cherry (Boller & Prokopy 1976).
Scientific interest in this fly extends beyond the realm of immediate pest control to
include empirical studies of physiology, behavior, and ecology (Dean & Chapman 1973,
Boller & Prokopy 1976, Prokopy & Roitberg 1984). The AMF has proven to be an
excellent subject for studies of foraging behavior (Prokopy & Roitberg 1987), visual
ecology (Owens & Prokopy 1986), resource utilization (Reissig 1979, Averill & Prokopy
1987), and sexual selection (eg. Prokopy & Bush 1973, Opp & Prokopy 1986). Neverthe-
less, large gaps in our knowledge of the behavior and ecology of this fly in its natural
environment still exist. For instance, we have yet to determine details of dispersal in
relation to food, oviposition site, and mate foraging behaviors. In addition, we know
little about individual variation in fly behavior over the host fruiting season in nature.
We undertook this study of marking and releasing wild AMF in nature to attempt
to answer such basic questions as: How long will an individual fly remain on the same
host tree? Does this residence duration differ between the sexes and change over the
fruiting season of the host? Does the onset of reproductive maturity following eclosion
affect residence duration?

Site.-In early June, 1984, we chose a small, Early Macintosh variety apple tree in
an unsprayed apple orchard naturally infested with AMF on the campus of the Univer-
sity of Massachusetts, Amherst (Fig. 1). We pruned the tree, thinning the leaves so
that all branches were clearly visible to an observer standing either on the ground or
on a 2.3m ladder. By mid-July, 1502 apples were ripening on this tree, whose canopy
was ca 5m tall X 5m diam. The two Early Macintosh variety apple trees in closest
proximity to the pruned tree (canopies within 2m) bore few or no fruit that season. In


0 0000


000 0000 0000

Fig. 1. Arrow denotes location of observation tree (0) in relation to other Early
Macintosh (E), Macintosh (M), and Cortland (C) variety apple trees at Orchard Hill,
University of Massachusetts, Amherst.


Opp & Prokopy: Seasonal Activity of Apple Maggot Fly 451

addition, trees of a later fruiting variety (Macintosh) in the adjacent row (canopies ca
5m away) bore few or no fruit that season. The closest fruiting trees (Cortland) that
season were located two rows away (ca 12m).
Marking Individuals.-The observation tree was checked daily until the first newly
closed adult AMF was sighted on June 24. Then, using mouth aspirators, we collected
flies daily from the tree for 12 days (until July 5). These flies were brought to the
laboratory for sex determination, measurement, and marking. Size was determined by
measuring the length of the dorsal mesothorax using an eyepiece micrometer on a
dissecting microscope. Each fly was immobilized briefly on ice and was marked individu-
ally with dots of one or two colors of Liquid PaperT on the dorsum of the thorax. A
symbol was then written on the Liquid Paper with a waterproof black felt pen (see
Walker & Wineriter 1981). Preliminary laboratory studies had indicated that marks
applied in this manner were non-toxic to the flies, yet were waterproof and durable.
By using four colors singly and in two color combinations along with 49 different sym-
bols, we were able to develop over 300 unique marks. Although we marked and released
327 female and 272 male AMF, not all flies seen on the observation tree over the course
of the experiment were marked, either because they eluded capture or because they
emerged or flew to the observation tree following the 12-day period of collection and
marking. We released all marked flies on leaves of the observation tree at dusk on the
day of collection.
Observations.-For 24 days after the first day on which flies were captured and
released (i.e. until July 18), we censused the tree for marked flies. Censuses were
conducted at one hour intervals between 0900 and 1700 hours when ambient tempera-
ture was above 21C and below 33C (the approximate activity thresholds of the flies)
(Prokopy et al. 1972, Johnson 1983), except during periods of heavy rain. During the
census periods, we also recorded the numbers of pairs of unmarked AMF in copula on
the observation tree. We accumulated 148 census-hours over the 24-day period for an
average of 6.2 census-hours per day.
To ensure that all portions of the observation tree were evenly censused for flies,
we divided the tree into 8 approximately equal-area sections based on the natural limb
structure of the tree. Leaves, fruit, and branches were examined for 5 min per section.
With this method, we were confident that all areas of the tree were inspected each hour
except the top sides of leaves located in the top 10% of the canopy.
Statistics.-To test for differences in resighting frequencies between the sexes and
over the season, we used G-tests with Yate's correction for continuity on frequencies
(see Sokal & Rohlf 1981). We used t-tests for unequal variances to assess both the
differences in total numbers of days in which files of each sex were sighted and the
influence of fly size on mating and resighting.


Of the 599 marked AMF which were released, we saw 183 (30.6%) at least once
during the 24 days of census. The great majority of these flies (137 of 183; 74.9%) were
seen only during the first week of observation. The remainder (46 of 183; 25.1%) were
seen during the first week but then were absent for an intervening period of 1-2 weeks
before resighting. We did not see equal proportions of marked male and female flies;
significantly more marked males were seen (100 of 272; 36.8%) then marked females (83
of 327; 25.4%) (G=8.52 with Yate's correction; p<0.001). Multiple sightings of males
over time were also more common than of females; whereas only 4% of females were
seen on more than two consecutive days, 24% of males were seen on more than two
consecutive days. Thus, on average, individual males were seen over more days

452 Florida Entomologist 70(4) December, 1987

(mean=2.18 days; S.E. =0.27) than females (mean=1.37 days; S.E.=0.09) (t=2.85;
p<0.05; df= 181.0). The maximum number of consecutive days over which we saw an
individual male or female was 14 and 7, respectively.
The oviposition and mating behaviors of male and female flies changed over the
season. Early in the census season, before July 7, we did not observe either marked or
unmarked females ovipositing into apples in the orchard. The apples were sufficiently
ripe to allow oviposition because when apples from our observation tree were brought
into the laboratory, our wild, laboratory-maintained AMF readily attempted oviposition
(D. R. Papaj, Dept. Entomology, University of Massachusetts, Amherst, personal com-
munication). Thus, we hypothesize that the flies observed in the field prior to July 7
were not ovipositing because they were not reproductively mature. This contention is
supported by the fact that no flies were observed mating prior to July 7.
Prior to July 7, we detected no significant difference between the number of male
or female flies observed on only one day versus the number of flies observed on more
than one day (G= 1.13; p>0.05). After July 7, the pattern of sightings of males and
females differed (Fig. 2), though not significantly (G= 1.00; p>0.05), probably due to
low sample sizes (n = 17 females; n= 33 males). The primary difference in sighting fre-
quency between males and females resulted not from a change in the frequency of
seeing females (both before and after July 7, most marked females were seen on only
one day; G=0.95; p>0.05), but was due to a change in the pattern of male sightings.
Following the onset of oviposition, males were more likely to be seen for many days
(G=5.80; p<0.05). The maximum time span over which a male was periodically re-
sighted was 22 days, and the maximum time span for a female was 24 days.
The peak time of day in which marked flies were seen also differed between males
and females. During the 1500 h census, we saw slightly more marked females than at
any other time (mean = 1.08 females/census hour), whereas the greatest mean number
of males were seen during the 1600 h census period (mean=2.89 males/census hour)


0.8 (61)
J 0.6
o (IT)
z 0.4

o (5)
20 --

SIGHTED --- 0 1

Fig. 2. Proportion of marked female and male AMF seen once or more than once in
relation to the onset of reproductive maturity on July 7. (Numbers of individuals.)

__ __ ~_^I\

q ....

Opp & Prokopy: Seasonal Activity of Apple Maggot Fly 453

(Fig. 3). In addition, the latter census period, during which we saw the greatest num-
bers of males, was one of the periods in which the fewest females were seen. For each
census period, a greater mean number of males than females were seen. The pattern
of sightings of unmarked mating pairs corresponded more closely with the pattern of
sightings of marked males than of marked females; most were seen at 1600 h, with a
considerable decrease during the 1700 h census period (Fig. 3).
We observed very few marked flies in copula. Only 12 marked males (12% of all
sighted marked males) and only 6 marked females (7.2% of all sighted marked females)
were observed copulating, in every case with an unmarked partner. Only 1 marked
female was seen to mate more than once (2 matings); 4 marked males (33% of all marked
males observed mating) mated multiply during census periods. One marked male mated
6 times and the other 3 marked males mated twice.
Sighted, marked females did not differ in size from females which were not seen
(t=-0.44, p>0.05, df=175.1). Marked males which were sighted were significantly
larger than marked males which were not sighted (t =-2.00, p<0.05, df =224.9). Marked
males and females which we observed mating did not differ in size from flies which were
not observed to mate (males: t=-0.98, p>0.05, df=19.4; females: t=-0.84, p>0.05.
df= 6.6).


Dispersal prior to reproduction is fairly common in adult insects and is sometimes
accompanied by the loss of flight ability once reproduction begins (Harrison 1980). In



S "' V \ PAIRS



0900 1000 1100 1200 1300 1400 1500 1600 1700


Fig. 3. Relationship between hour of day and average numbers of marked individual
male and female and unmarked mating pairs of AMF seen on the observation tree over
the 24 day observation period.

Florida Entomologist 70(4)

AMF, many pre-reproductive adults dispersed away from the site of emergence. Ap-
proximately 75% of the newly emerged AMF we marked left the host tree after being
seen within the first week and were not seen again. The remaining 25% apparently left
the host tree shortly after emergence but returned when reproductively mature, 1-2
weeks later. Using radiolabelled AMF, Neilson (1971) also found that many flies which
dispersed outside of a naturally infested orchard early in the season later returned.
Similar dispersal behaviors of the immature adults of a close relative of the AMF,
Dacus tryoni (Froggatt), the Queensland fruit fly, have been reported (Fletcher 1973,
1974). Using mark-recapture methods in a naturally infested orchard, Fletcher (1973,
1974) found that 75% of D. tryoni left the orchard in their first week and did not return.
In later weeks, as flies became mature, many re-entered the orchard. Although D.
tryoni are larger and capable of longer dispersal flights than AMF (Neilson 1971,
Fletcher 1974), the same general pattern of dispersal away from hosts prior to reproduc-
tion followed by return when reproductively mature occurs as we observed in the AMF.
This pattern likely corresponds to the change from primarily food foraging behavior
when reproductively immature to mate and host foraging behaviors when mature (Har-
rison 1980).
We detected distinct differences between wild male and female AMF in the tendency
to remain on a host tree, with those differences magnified following the onset of fly
reproductive maturity. In general, we resighted many more males than females, but
the most striking differences between the sexes occurred after oviposition began. Fol-
lowing oviposition, female AMF deposit a marking pheromone on the surface of fruit
that deters further egglaying (Prokopy 1972). Previous studies using field-caged flies
showed this marking pheromone elicits female emigration from host trees (Roitberg et
al. 1982, 1984). Just the opposite behavior, arrestment of activity, occurs in male AMF
when they contact marking pheromone on fruit (Prokopy & Bush 1972).
Both before and after the onset of oviposition and deposition of marking pheromone,
we found that female AMF were not likely to remain on a single host apple tree for
more than one day. Although this effect may have been heightened by our thinning of
tree leaves, we could not detect any increase in female emigration from the host tree
which might have been due to contact with marking pheromone. The lack of fruit on
immediately adjacent host trees may have caused females to remain on our observation
tree longer than if suitable host fruit were available nearby, or may have resulted in
longer dispersal flights by females to find new oviposition sites (see also Neilson 1971).
Fletcher (1973) found that the length of time D. tryoni remained in an orchard was in
great part determined by the quantity of fruit available for oviposition. Hendrichs &
Reyes (1987), however, felt that the length of time D. longistylus (Wied.) females spent
on a host was influenced by encounters with patrolling males which were continually
attempting forced copulations.
Male AMF tended to remain on the tree longer and were seen for more consecutive
days in the latter than in the earlier part of the season. We hypothesize that once
females had commenced oviposition, males frequently were contacting female marking
pheromone on fruit. Contact with marking pheromone would arrest male activity on
fruit (Prokopy & Bush 1972). Similarly, Johnson (1983) found that male AMF responded
more strongly to the mating-oviposition stimulus of a red sphere trap than to the feeding
stimulus of a yellow panel trap. By remaining on fruit, males increase the probability
of encountering females arriving on the fruit to oviposit, and thus increase their oppor-
tunities to mate, since over 90% of matings occur on fruit (Prokopy et al. 1987) and
most occur when females are in some phase of oviposition behavior (Prokopy & Bush
1973, Smith & Prokopy 1980). This observation is consistent with the hypothesis of
Thornhill & Alcock (1983) that when females of a species multiply mate, males would
be expected to search for mates near sites of female oviposition. Although we were not

December, 1987

Opp & Prokopy: Seasonal Activity of Apple Maggot Fly 455

able to document multiple female mating in this study, we expect multiple mating to
occur in nature because laboratory studies have shown that female AMF benefit from
multiple copulations in terms of increased fecundity and fertility (Opp & Prokopy 1986).
The peak time of mating by unmarked AMF corresponded more closely with time
of observation of peak male presence than peak female presence (see also Prokopy et
al. 1972). Since most matings on fruit are male rather than female initiated (Smith &
Prokopy 1980), peak time of male abundance on fruit might be one of the primary factors
governing diel mating patterns.
Using the maximum time span over which marked male and female AMF were
sighted (24 days for females and 22 days for males), we conservatively estimate that,
as adults, some flies may live up to 4 weeks in nature. Neilson & Wood (1966) estimated
from field and laboratory cage studies that AMF adults may live up to 1 month when
supplied with aphid honeydew. Although female size appeared to have no influence on
longevity in our study, body size may have affected longevity of males because more
large males were resighted over time than small males. We detected no interaction
between body size and mating success for either sex, although body size is known to
influence mating success in other dipteran species (Borgia 1981, Burk & Webb 1983,
Sivinski 1984).
This study provides information on individual fly activities in nature but raises many
questions concerning AMF behavior. For example, although we know that AMF are
not likely to remain on the host tree early in the season for more than one day, we do
not know where these pre-reproductive individuals go. Furthermore, we do not have
comprehensive information concerning the natural food of these flies and their food
foraging behavior, although we know that protein is necessary to attain reproductive
maturity (Webster et al. 1979). Finally, many questions remain concerning male-female
interactions, especially the average numbers of times individuals mate on host plants.
We know that most matings occur on fruit and are male initiated but we have no
estimate of variance in individual mating success. We plan to address these and many
more questions in future studies.


We thank M. Aluja, S. Cooley, J. Hendrichs, and P. Powers for helpful comments
on this manuscript. This work was supported by NSF Doctoral Dissertation grant
BSR-8514524 and by Massachusetts Agricultural Experiment Station Project 604.


AVERILL, A. L., AND R. J. PROKOPY. 1987. Chapter 4.5. Host-marking pheromones.
In: A. S. Robinson and G. H. S. Hooper, eds. Fruit flies: their biology, natural
enemies, and control. Elsevier, Amsterdam. (in press)
BATEMAN, M. A., AND F. J. SONLEITNER. 1967. The ecology of a natural population
of the Queensland fruit fly, Dacus tryoni. I. The parameters of the pupal and
adult populations during a single season. Australian J. Zool. 15: 303-335.
BOLLER, E. F., AND R. J. PROKOPY. 1976. Bionomics and management of Rhagoletis.
Annu. Rev. Entomol. 21: 223-246.
BORGIA, G. 1981. Sexual competition in Scatophaga stercoraria: size and density-re-
lated changes in male ability to capture females. Behaviour 78: 185-205.
BURK, T., AND J. C. WEBB. 1983. Effect of male size on calling propensity, song
parameters and mating success in Caribbean fruit flies, Anastrepha suspense
(Loew). Ann. Entomol. Soc. America 76: 678-682.
DEAN, R. W., AND P. J. CHAPMAN. 1973. Bionomics of the apple maggot in eastern
New York. Search Agric. Entomol. (Geneva, N.Y.) 3.

Florida Entomologist 70(4)

FLETCHER, B. S. 1973. The ecology of a natural population of the Queensland fruit
fly, Dacus tryoni. IV. The immigration and emigration of adults. Australian J.
Zool. 21: 541-565.
FLETCHER, B. S. 1974. The ecology of a natural population of the Queensland fruit
fly, Dacus tryoni. V. The dispersal of adults. Australian J. Zool. 22: 189-202.
HARRISON, R. G. 1980. Dispersal polymorphisms in insects. Annu. Rev. Ecol. Syst.
11: 95-118.
HENDRICHS J., AND J. REYES. 1987. Reproductive behaviour and post-mating fe-
male guarding in the monophagous multivoltine Dacus longistylus (Wied.) (Dipt-
era: Tephritidae) in southern Egypt. In: A. P. Economopoulos, ed. II. Inter-
natiopnal symposium on fruit flies. Elsevier, Amsterdam.
JOHNSON, P. C. 1983. Response of adult apple maggot (Diptera: Tephritidae) to
Pherocon A.M. traps and red spheres in a non-orchard habitat. J. Econ. Entomol.
76: 1279-1284.
Joos, J. L., W. W. ALLEN, AND R. A. VAN STEENWYK. 1984. Apple maggot: A
threat to California's apple industry. California Agric. 38: 9-11.
NEILSON, W. T. A. 1971. Dispersal studies of a natural population of apple maggot
adults. J. Econ. Entomol. 64: 648-653.
NEILSON, W. T. A., AND F. A. WOOD. 1966. Natural food source of the apple maggot.
J. Econ. Entomol. 59: 997-998.
OPP, S. B., AND R. J. PROKOPY. 1986. Variation in laboratory oviposition by
Rhagoletis pomonella (Diptera: Tephritidae) in relation to mating status. Ann.
Entomol. Soc. America 79: 705-710.
OWENS, E. D., AND R. J. PROKOPY. 1986. Relationship between reflectance spectra
of host plant surfaces and visual detection of host fruit by Rhagoletis pomonella
flies. Physiol. Entomol. 11: 297-307.
PROKOPY, R. J. 1972. Evidence for a pheromone deterring repeated oviposition in
apple maggot flies. Environ. Entomol. 1: 326-332.
PROKOPY, R. J., AND G. L. BUSH. 1972. Mating behavior in Rhagoletis pomonella
(Diptera: Tephritidae). III. Male aggregation in response to an arrestant. Cana-
dian Entomol. 104: 275-283.
PROKOPY, R. J., AND G. L. BUSH. 1973. Mating behavior of Rhagoletis pomonella
(Diptera: Tephritidae). IV. Courtship. Canadian Entomol. 105: 873-891.
PROKOPY, R. J., AND B. D. ROITBERG. 1984. Foraging behavior of true fruit flies.
American Sci. 72: 41-49.
PROKOPY, R. J., AND B. D. ROITBERG. 1987. Chapter 5.1. Fruit fly foraging be-
havior. In: A. S. Robinson and G. H. S. Hooper. eds. Fruit flies: their biology,
natural enemies, and control. Elsevier, Amsterdam. (in press)
PROKOPY, R. J., E. W. BENNETT, AND G. L. BUSH. 1972. Mating behavior in
Rhagoletis pomonella (Diptera: Tephritidae). II. Temporal organization. Cana-
dian Entomol. 104: 97-104.
PROKOPY, R. J., S. S. COOLEY, AND S. B. OPP. 1987. Prior experience influences
fruit residence of male apple maggot flies. J. Insect Behav. (in press)
REISSIG, W. H. 1979. Survival of apple maggot larvae, Rhagoletis pomonella (Dip-
tera: Tephritidae), in picked and unpicked apples. Canadian Entomol. 111: 181-
J. PROKOPY. 1982. Foraging behaviour of Rhagoletis pomonella, a parasite of
hawthorn (Crataegus viridis), in nature. J. Anim. Ecol. 51: 307-325.
ROITBERG, B. D., R. CAIRL, AND R. J. PROKOPY. 1984. Oviposition deterring pher-
omone influences dispersal distance in tephritid fruit flies. Entomol. Exp. Appl.
35: 217-220
SIVINSKI, J. 1984. Sexual conflict and choice in a phoretic fly, Borborillusfrigipennis
(Sphaeroceridae). Ann. Entomol. Soc. America 77: 232-235.
SMITH, D. C., AND R. J. PROKOPY. 1980. Mating behavior of Rhagoletis pomonella
(Diptera: Tephritidae). VI. Site of early-season encounters. Canadian Entomol.
112: 585-590.

December, 1987


Haag et al.: Bird Predation

SOKAL, R. R., AND F. J. ROHLF. 1981. Biometry. W. H. Freeman, San Francisco.
859 p.
THORNHILL, R., AND J. ALCOCK. 1983. The evolution of insect mating systems.
Harvard University Press. Cambridge, Mass. 547 p.
WALKER, T. J., AND S. A. WINERITER. 1981. Marking techniques for recognizing
individual insects. Florida Entomol. 64: 18-29.
WEBSTER, R. P., J. G. STOFFOLANO, JR., AND R. J. PROKOPY. 1979. Long-term
intake of protein and sucrose in relation to reproductive behavior of wild and
laboratory cultured Rhagoletis pomonella. Ann. Entomol. Soc. America 72: 41-


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

Center for Aquatic Plants, IFAS, University of Florida
Gainesville, FL 32611

1614 N.W. 12th Road, Gainesville, FL 32601


Department of Food and Resource Economics, University of Florida
Gainesville, FL 32611


Common moorhens (Gallinula chloropus), boat-tailed grackles (Quiscalus major)
and red-winged blackbirds (Agelaius phoeniceus) were collected monthly from Sep-
tember 1984 to August 1985 at a waterhyacinth (Eichhornnia crassipes) infested lake
in northcentral Florida. Contents of esophagi and gizzards were examined to assess
predation on waterhyacinth weevils, Neochetina eichhorniae and N. bruchi, and other
aquatic insects. Waterhyacinth weevils were rarely found in stomach contents of com-
mon moorhens, whose herbivorous diet consisted largely of coontail, hydrilla and
duckweed. Although diets of the other two bird species varied seasonally, both con-
sumed significant proportions of aquatic insects. Adults of Donacia leaf beetles were
found to be very abundant in stomach contents of grackles and blackbirds (aggregate
percent: R= 17.2, x= 26.9, respectively). Neochetina weevils were frequently found in
gizzards of boat-tailed grackles and red-winged blackbirds (percent occurrence: R = 52.5,
R= 41.5, respectively), but accounted for a relatively small proportion of the total diet
(aggregate percent: x= 1.0, = 6.3, respectively). Implications for biological control of
waterhyacinth are discussed.


458 Florida Entomologist 70(4) December, 1987


La gallineta de agua (Gallinula chloropus), el tordo mayor (Quiscalus major), y el
tordo alas-rojas (Agelaius phoeniceus) fueron colectados mensualmente desde Sep-
tiembre de 1984 hasta Agosto de 1985 en un lago infestado por lirio de "waterhyacinth"
(Eichhornia crassipes) en el Centro Norte de la Florida. Los contenidos de es6fagos y
mollejas fueron examinados para determinar la depredaci6n sobre los gorgojos del lirio
de agua, Neochetina eichhorniae y Neochetina bruchi, y ostros insects acuaticos.
Estos gorgojos del lirio de agua fueron raramente encontrados en los contenidos
estomacales de las gallinetas de agua cuya dieta hebivora consisti6 principalmente en
"coontail", "hydrilla" y "duckweed". A pesar de que las dietras de las otras dos aves
variaron estacionalmente, ambas consumieron proporciones significativas de insects
acuaticos. Los adults del escarabajo Donacia, fueron incontrados abundantemente en
el contenido estomacal de los tordos mayores y de los tordos de alas-rojas (porcentaja
agregado: = 17.2, R =2.69, respectivamente). Los gorgojos, Neochetina, fueron encon-
trados frecuentemente en las mollejas de los tordos mayores y de los tordos de alas
rojas (porcentaje de ocurrencia: R = 52.5, = 41.5, respectivamente), pero representaron
una proporci6n relativamente pequefia de la dieta total (porcentaje agregado: x= 1.0,
= 6.3, respectivamente). Se discuten las implicaciones para el control biol6gico del lirio
de agua.

This study was undertaken in the context of a project designed to examine the
factors influencing biological control of waterhyacinth, Eichhornia crassipes (Mart.)
Solms, in Florida (Haag 1986a,b). Waterhyacinth is a widespread aquatic weed in the
southeastern United States and rapidly is becoming a problem in parts of Texas and
California. Recent efforts to control this weed have focused on biological control. Two
phytophagous weevils, Neochetina eichhorniae Warner and N. bruchi Hustache (Col-
eoptera: Curculionidae) were imported from Argentina and released in the U. S. in the
early 1970's (Perkins 1973). Adult weevils feed on the surface of leaves and petioles,
whereas larvae feed within the petioles and crown of the plant. These insects have
dispersed widely in Florida from the original sites of introduction and are now found
wherever waterhyacinth infestations occur. Although waterhyacinth still poses an aqua-
tic weed problem at some sites in Florida, it is generally accepted that infestations have
been reduced in number and severity as a result of stress from feeding by larval and
adult Neochetina (Center 1982).
Since their introduction to Florida, waterhyacinth weevils have become a commonly
encountered member of the aquatic and semiaquatic insect community found on
waterhyacinth weed mats (Lynch et al. 1947, O'Hara 1961, 1968, Katz 1967, Wahlquist
1969, Hansen et al. 1971). In a review of the literature on native insects found on
waterhyacinths, Balciunas (1977) noted that few studies had classified organisms to the
generic level, and that comparison of faunas from various sites was therefore difficult.
In addition to these studies, much of the other literature concerning insects associated
with waterhyacinths deals with potential biological control agents (Bennett 1968, 1972,
Gordon & Coulson 1969, Coulson 1971, Perkins 1974, Spencer 1974).
Wetland birds are frequently found roosting and feeding on waterhyacinth mats.
These floating mats, which form an extension of the littoral emergent plant community
characteristic of many shallow southeastern lakes, provide shelter and foraging areas
for waterfowl. However, the feeding habits of birds in this habitat have rarely been
examined. Among the birds found on aquatic vegetation in Florida are the common
moorhen (Rallidae: Gallinula chloropus (L.)), the boat-tailed grackle (Icteridae: Quis-
calus major Vieillot), and the red-winged blackbird (Icteridae: Agelaius phoeniceus
(L.)). They are often the most commonly encountered bird species on waterhyacinth

Haag et al.: Bird Predation 459

plants in many lakes in north central Florida and were therefore of particular interest
in this study.
Although there are several reports on the diets of these three species in either
aquatic or terrestrial habitats (Beal 1900, Allen 1914, Wetmore 1916, Bent 1926, Howell
1932, Simpson 1939, Beal et al. 1941, Bird & Smith 1964, Snelling 1968, Orians 1973,
Voights 1973, Reagan 1977, Dolbeer et al. 1978, Mulholland & Percival 1983), no studies
are available which compare diets of red-winged blackbirds, boat-tailed grackles and
common moorhens in wetland habitats.
The objective of this study was to examine the diets of three species of wetland birds
frequently found on floating aquatic vegetation and determine the extent to which they
feed on waterhyacinth weevils and other periodically abundant aquatic insects on
waterhyacinth mats.


The study area was located in the northeastern sector of Orange Lake, a large (4,921
ha), shallow (mean depth = 2.9 m) nutrient-enriched lake in northcentral Florida (Shire-
man & Haller 1979), approximately 32 km southeast of Gainesville. The shore region or
littoral zone is occupied by a heterogenous community of floating, emergent and sub-
mersed vegetation, which varies both seasonally and in response to fluctuating water
levels. Among the plants most commonly encountered in the study area were spatter-
dock (Nuphar luteum ssp. macrophyllum (Small) E. O. Beal), American frogbit (Lim-
nobium spongia (Bosc) Steud.), pickerelweed (Pontederia cordata var. lancifolia
(Muhl.) Torr.), mild smartweed (Polygonum hydropiperoides Michx.), water fern (Sal-
vinia minima Baker), common duckweed (Lemna minor L), Carolina mosquitofern
(Azolla caroliniana Willd.), hydrilla (Hydrilla verticillata L.f.Royle), coontail
(Ceratophyllum demersum L.), and waterhyacinth. A detailed listing of the aquatic
flora of Orange Lake was published by Shireman et al. (1983).
Quarterly (Spring = March, April, May; Summer = June, July, August; Fall =
September, October, November; Winter = December, January, February) assessments
of waterhyacinth weevil densities were made at Orange Lake during the study. A
sample consisted of 50 plants collected from a single weed mat. Plants were taken to
the laboratory and adult weevils were collected and counted. Weevils were preserved
in 70% isopropyl alcohol and later enumerated by species and sex.
Although Orange Lake supports a diverse avian fauna (Sprunt 1954), our preliminary
observations indicated that the three most common birds found on mats of
waterhyacinth were the common moorhen, the boat-tailed grackle, and the red-winged
blackbird. All three species were collected monthly from October 25, 1984 through
September 25, 1985. Birds were shot from an airboat during the morning, since frequent
storms made afternoon sampling unsafe. Whenever possible, birds were collected while
foraging on waterhyacinth mats. In all other cases, birds were collected in the vicinity
of waterhyacinths. Three individuals of each species were collected each month, with
two exceptions. Extremely cold weather decimated the waterhyacinth mat and pre-
cluded sampling in January 1985. Intense herbicide control programs again diminished
the waterhyacinth population so severely in June 1985 that no birds were collected
during this month.
Birds were kept on ice and taken to the laboratory. Esophagi and gizzards were
then removed and kept frozen until analyses could be completed. It is generally accepted
that esophageal tracts should be examined in dietary studies since food items in this
organ have not yet been subjected to physical and chemical breakdown (Korschgen
1980). However, esophagi were often empty in the birds which we collected. Therefore,
both esophagi and gizzards were opened in a petri dish and their contents combined.



I 0








N i









u 0







g. ,- II

g 1

OJD 0)

December, 1987

00 Cq


0 0 M 0C -
0 -

te -4
-I 00 I

$"1 o o

CM- P 0C

. ^
" "S,

Florida Entomologist 70(4)


0000 C 00 O M 0000
Z 0 0 C
0O0 0 0 0 O 0 l0

0 C -4l Z CCl CM

m 3 a .
ll li i
g S^-S *BS StB' a i ^ o



b CC


O ^ II



Haag et al.: Bird Predation

I10 -1 C N
oI 0 F I-A 0 _
o o *0 Cl0000l l l0


1 10 0
o o

C Cl 0

-00 0
t 0



0 ~000060000~000

001 | ^0C^
,-40 Iz-O
Lf~ oo

.S I

d dia t ta^ 3 e o0
< Q -
p_ o
Sd ., '

) 0
1=2 ^F
i-o tt C


a 0)


0 )i

.0) 0

+01 0) 0)^

0) 0) 0)

li 0

E 0 h
0) 8
0) )i0)i0

Florida EioiiulgiJs 70(4)

Food items were identified to the lowest possible taxonomic category. Invertebrates
were identified using Merritt & Cummins (1978) and Pennak (1978). Plant material was
identified using Knobel & Faust (1977), Godfrey & Wooten (1979, 1981) and Tarver et
al. (1986). Seeds were identified using Martin & Barclay (1973). Items which could not
be identified were submitted to local experts.
The contribution of each food type was measured using volumetric displacement.
Individual analyses were pooled and reported as the mean seasonal (Spring, Summer,
Winter, Fall) composition of food items for each bird species. Data analyses of food
habits included calculation of percent occurrence (number of birds consuming each food
item divided by number of birds in sample), aggregate volume (volume of each food
item from all birds in sample divided by the total food volume from all birds in the
sample), and aggregate percent (the proportion of the jth food item in the ith bird,
averaged over all birds in the sample) (Martin et al. 1946, Swanson et al. 1974). A
Spearman rank correlation coefficient (Daniel 1978) was used to compare the seasonal
ranks given to foods by aggregate volume and aggregate percent to determine if these
values were significantly different from one another.


No efforts were made to collect equal proportions of male and female birds in this
study. Of the birds collected, females comprised 86% of the moorhens, 67% of the
boat-tailed grackles and 3% of the red-winged blackbirds. Due to their coloration, male
red-winged blackbirds are more readily observed than females and this probably ac-
counts for their preponderance in our collections. As a consequence of the lack of unifor-
mity in sex ratios of birds collected, no efforts were made to correlate bird diet with sex.
Spearman rank coefficients, calculated to compare values for aggregate volume and
aggregate percent of food items in each species, were not significantly different
(p<0.01). Therefore discussion will refer to values for aggregate percent since this
parameter allows equal influence by each bird, regardless of total individual gizzard
volumes (Martin et al. 1946).
The diet of common moorhens collected in this study was predominantly herbivorous.
Identifiable plant material accounted for 32.1% to 49.6% of total food consumed, depend-
ing on the season in which birds were collected (Table 1). Leaves and stems of coontail
(Ceratophyllum demersum) and hydrilla (Hydrilla verticillata) predominated, with
hydrilla contributing almost 50% of total food volume in the summer months. Seeds of
smartweed (Polygonum sp.) were consumed extensively in the fall by this species. The
diet was most diverse in the winter months, when seeds of 9 different plant species
were found in gizzard contents.
Animal material comprised only 0.2% to 4.6% of total food volume of common
moorhens. Among those species consumed were water bugs (Belostoma spp.), plant
hoppers (Fulgoridae), diving beetles (Dytiscidae), crayfish (Procambarusfallax Hagen)
and shell snails (Planorbella sp.) O'Meara et al. (1982) reported a similar diet for com-
mon moorhens in natural wetlands. Waterhyacinth weevils were not found in stomach
contents during the summer months in our study. These weevils were occasionally
found in the other three seasons but never contributed more than a trace (< 0.01%) to
total food volume.
Much of the material in moorhen gizzards was grit. In addition, there was a consid-
erable volume of material which could not be identified due to advanced stages of
digestion. This material ranged from 45.8% to 65.2% of total food volume. Mulholland
& Percival (1983) reported that the most common food items in gizzards of common
moorhens on Orange Lake were seeds of smartweed (Polygonum sp.), leaves and stems


December, 1987

Haag et al.: Bird Predation

of Hydrilla, and snails (Planorbella sp.). Food volume averaged 93% plant material and
7% animal material. Data collected in the present study parallel these figures, although
a higher proportion of food materials were unidentifiable in our study. Among insects
that Mulholland & Percival (1983) commonly found were damselflies and dragonflies
(Odonata), leaf beetles (Coleoptera: Chrysomelidae) and members of the weevil family
(Curculionidae). The weevils were not identified beyond the family level in their study.
Unfortunately, these specimens were not available for further examination and it is not
known whether any Neochetina were present.
Proportions of plant and animal food in the diet of boat-tailed grackles collected from
Orange Lake varied greatly with season (Table 2). In fall and winter, plant food com-
prised 32.8% and 35.7%, respectively, of food volume. Seeds of sorghum (Sorghum
vulgare), corn (Zea mays) and water shield (Brasenia schreberi) were consumed in
largest amounts in these months. In spring, plant material consisted solely of corn (10%
of total diet), whereas in summer no plant foods were identifiable in grackle stomach
Animal material was abundant in esophageal tracts and gizzards of boat-tailed grack-
les in all seasons. Chrysomelid beetles in the genus Donacia were the single most
abundant organisms. Although not found in stomach contents in winter, they comprised
30.7% of total food in spring, 27.4% in summer and 10.9% in fall. In the fall, other
abundant insects included dragonfly nymphs (Libellulidae), noctuid caterpillars, and
rat-tailed maggots (Eristalis sp.). In winter, spiders were frequently consumed along
with pyralid and tipulid larvae. In spring, dragonfly nymphs were abundant in stomach
contents, as well as gastropod snails and weevils (Curculionidae) other than Neochetina.
In the summer months, spiders, diving beetles (Dytiscidae), water beetles (Hydro-
philidae), scarab beetles (Scarabaeidae), noctuid caterpillars and soldierfly larvae
(Stratiomyidae) were frequently consumed. Waterhyacinth weevils contributed 3.1% to
the food volume in the fall but were present in very small numbers for the remainder
of the year.
In his samples, Beal (1900) found that boat-tailed grackles consumed 60% vegetable
material (mostly grain) and 40% animal material. Insects comprised about 25% of the
food eaten, with grasshoppers and carabid beetles among the most abundant animals
eaten. Among the other beetles consumed, curculionids were particularly numerous
(22% of annual food in June; 4% of annual food year round). In another terrestrial study,
Beal et al. (1941), noted that the diet of boat-tailed grackles was composed of 47%
animal material. Among the items eaten were crustaceans and insects, particularly
grasshoppers in the late summer. This suggests opportunistic feeding on a seasonally
abundant species, as we found with Donacia. Although beetles were identified by Beal
et al. (1941) no one family was especially abundant.
Red-winged blackbirds collected from Orange Lake also had large amounts of animal
material in their stomachs (Table 3). Plant material ranged from 2.9% to 20.4% of total
food volume. In the fall, seeds of smartweed were the predominant plant food, whereas
during the winter months seeds of water willow (Decodon verticillatus (L.) Ell.),
smartweed and corn were the most abundant plant food items. In the spring, seeds of
water shield and several grasses were found in stomach contents. In summer months,
the only plant foods found were grass seeds (Panicum).
Red-winged blackbirds ate a wide variety of insects at Orange Lake. In the fall,
abundant food items included arachinids (5.1%), chrysomelid beetles (Donacia) (18.4%)
and noctuid caterpillars (5.1%). In the winter months, arachnids (7.3%), damselfly
nymphs (6.0%), various caterpillars (10.5%) and ants (5.2%) were abundant. Arachnids
were also frequent food items in spring (13.2%), while Donacia leaf beetles contributed
up to 50% of total food volume during those months. In the summer, Donacia sp. were
also very abundant (38.8%). These data contrast with reports that species of Donacia











o S

N '^

3 5



o 5
0 <'
0 K

Florida Entomologist 70(4)


--4 Cl.

~. II



S "



C.) 5



2 1

a 0

, CK
w II

CO LO 0 i

00I CO L-- 00


"0 "~~ ^

^ S5 E-; |-
,^ .8owg?

(I2 ? -
QC W "^
'y~J S SS
H h S^ & H Ssi^^ W
MI 03e Wllh-

December, 1987

0 t-CO -(oo t-o
u~c~9d~~cr3 -4 C


I -

00 0 00000

0 0 CM I

O *-1 -1-0 1-0 CO

C) P

0Cae~ a) MI.
C 0

S)= -a.) U C C 2 .

; o o t t M 1U 0) S
eCh-< ZiCQ CL

oo0oo 0 oo00oM0co0 m

C) 000 C00 M I cc 0 M M 10

0000 l

Haag et al.: Bird Predation

t-1- 00 00

S -d td i- I I I Itod

( I co K ao

0 D00000000 L- L- 0 000 t-

0000 0 m w ooo

' I I I I I| I

Noo0 00 0 00 t- 000 000 O 0-
" C00o c0 1-4 N -



0000 0000

C C)CC 00

177OO OOOC-|

Soo0 o L-


o a a 'o -'4
oa w cz a5 d
ce e W l'Bq e
cus 0)O~ "L. 1.O 1W ~ pei
0 4 146
~ 4 -
0)cs sz' *~& c e~

Z. u 6~ E)- 4


0 -

'0 >~

.o .5
o,. 0
r '~.E
01 CdE

-B Cd Cd

.0 0












December, 1987



0) v. II
a o
g 1





0 ) II


ScSS c



a s


o000o4 CO N O -4
-I COC01" -4-l

-o o

l ~ ~ ^ f r *
00 3

0 0

'0 4

Florida Entomologist 70(4)

1 1I I II IH



I I K- I I I S Io
o Tc ol I-

-0 = t- 000 - 0 00
"1 1 1-0 14 M 1

n C-4

0 -t-oo000-

0 o t- T-
I 000I

0 M'-cl'-0l ._0
CO C -

a o 0

I I M l I I


C 00. 00
--i II00i
-^5 ia M I'-''

Haag et al.: Bird Predation

Ijd I I I cod

C I- I

0soooooo00000 OOO I t-00-O

0 |'0

1-o00 ~ 00i0 1 0 I 0
r-LO N 1 -4 -


L-O( L-0-t 40C'- (= .-O= 4 -M--L- - 4 0 L- 0

00 IO Q- I -4 1 d C 1 0 l

Co= o C -I ,- ,
| o-d -1 | o |

i-IO O '^ i-IO r-I CO I SO '- '-I O '-I '-l -IO

?, %
o 3
0 3 0w -
ce m8 ce ce a
cd 0 r
-c? c
., c S Ei S ^ c s >-
~ g a) a)
$E aaI n- 'C.I ^aS ^'
ce E pia C"
Ao 0 0 c

~~C U 0 r ,~ a~
u I' l wX ^ ilacl oE i .&l ^E !l ^F Fe Eua s1


od o- I:-lll



I oI

u3 CO0


o o
Ia I
0z 0 t-i

. .

P -~

-S .6

-o V
3 .3

0 .
C C)0
r3*0 ^
E-S i

.0.0 0

*C-s C
'S E*S
1.1 1

is'n-5 f
S^ g E
E" C

(0 L- Ci Ic 00 m = -IIMOS
CI C) cc CD C> Coeo C| m C| mO o

Florida Entomologist 70(4)

are unpalatable to some species of avian predators (Otto 1985). Neochetina adults were
often found in stomach contents in fall (9.1%), winter (6.0%) and summer (10.1%) but
were not found in spring.
Howell (1972) studied the feeding habits of red-winged blackbirds in Florida and
found that 27% of food eaten was animal material. This included beetles, grasshoppers,
caterpillars, wasps, ants, flies and dragonflies. Selling (1968) compared diets of red-
winged blackbirds and common grackles in a Wisconsin cattail marsh. Animal material,
chiefly insects, comprised 92% of the blackbird diet. Geometrid and noctuid larvae
predominated in diets of both birds, and odonates were also important food items.
Dolbeer et al. (1978) examined diets of red-winged blackbirds in winter in an agricul-
tural area in Tennessee, where insects comprised s 1% of stomach contents. Corn and
weed seeds were the most abundant food items. Bird & Smith (1964) compared diets of
red-winged blackbirds from agricultural and marsh districts. In the agricultural areas,
diets were 90% vegetable material in spring, whereas in summer animal material,
mostly insects, comprised up to 70% of blackbird diets. In marsh areas, foods were
almost exclusively animal material (81-100%). Differences in food items eaten in the two
areas were attributed to differences in local abundance. Voights (1973) examined diets
of red-winged blackbirds, and he too found evidence of opportunistic feeding. Diets of
red-winged blackbirds in early June were composed largely of noctuid caterpillars,
whereas in late June damselflies and larval soldierflies were abundant. Birds foraged
more in areas where food availability, due to insect emergence patterns, increased.
Although the data collected in our study establish that predation on waterhyacinth
weevils and other aquatic insects does occur, and varies seasonally with abundance of
these food items it is difficult to estimate the effects of this predation pressure on weevil
populations. Inherent in the problem is the difficulty in assessing weevil populations
themselves. Distribution of the insects is quite patchy at Orange Lake (K. H. Haag,
unpublished data), and at most other sites examined. During the course of our sampling
at Orange Lake, weevil densities ranged from 0 weevils per 50 plants to as high as 39
weevils per 50 plants (mean density = 17.6 10.3 weevils per 50 plants). In addition,
no assessment was made of feeding rates of the individual birds, so we do not have a
clear idea of how many insects might be consumed in a 24-h period.
We have resolved several misconceptions, however. It was widely held by plant
management workers (Eddie Knight, U.S. Army Corps of Engineers, personal com-
munication) that moorhens were doing most of the feeding on waterhyacinth weevils at
Orange Lake. These birds are the largest and most visible of the three species, as well
as the most abundant. Clearly, they are primarily herbivorous and consume insects only
rarely during their feeding activities. Boat-tailed grackles consume weevils, although
the insects contribute only a small percentage to their total food volume. However,
these birds are more numerous than the red-winged blackbirds at Orange Lake. They
have also been reported to be numerous at other waterhyacinth field sites, including
monotypic stands of E. crassipes used in resource recovery systems (David Haselow,
Amasek Inc., personal communication). Due to their abundance and wide distribution,
these birds may exert a greater predation pressure on waterhyacinth weevils than the
other two species. Further studies are needed to fully evaluate the effects of this preda-
tion pressure on waterhyacinth weevils.


The authors wish to thank Dale Habeck for providing facilities and support needed
to conduct this study. Gary Buckingham, Rosi Mulholland, and Frank Slansky, Jr.,
reviewed the manuscript and provided many helpful suggestions. This work was sup-


December, 1987

Haag et al.: Bird Predation

ported in part by funding from the U.S. Department of Agriculture, ARS Cooperative
Agreement No. 58-7B39-3-570 and the Center for Aquatic Weeds, Institute of Food and
Agricultural Sciences. Published as Journal Series No. 7923 of the Florida Agricultural
Experiment Station.


ALLEN, A. A. 1914. The redwinged blackbird: a study in the ecology of a cattail
marsh. Proc. Linnean Soc. New York. 24-25: 43-123.
BALCIUNAS, J. K. 1977. Species abundance relationships of aquatic insects in
monotypic waterhyacinth communities in Florida, with special emphasis on fac-
tors affecting diversity. Unpublished Ph. D. dissertation. Univ. of Florida,
Gainesville. 166 pp.
BEAL, F. E. L. 1900. Food of the bobolink, blackbirds and grackles. USDA, Division
of Biological Survey, Washington, D.C. 77 pp.
BEAL, F. E. L., W. L. McATEE, E. R. KALMBACK. 1941. Common birds of the
southeastern United States in relation to agriculture. Conservation Bulletin 15.,
US Dept. of Interior, Washington, D.C. 43 pp.
BENNETT, F. D. 1968. Insects and mites as potential controlling agents of
waterhyacinth (Eichhornia crassipes) (Mart.) (Solms). Pro. 9th British Weed
Control Conf. 832-835.
BENNETT, F. D. 1972. Survey and assessment of the natural enemies of
waterhyacinth, Eichhornia crassipes. PANS 18(3): 310-311.
BENT, A. C. 1926. Life histories of North American marsh birds. U. S. Nat. Mus.
Bull. 1325: 1-490.
BIRD, R. D. AND L. B. SMITH. 1964. The food habits of the redwinged blackbird,
Agelaius phoeniceus in Manitoba. Canadian Field Nat. 78: 179-186.
CENTER, T. D. 1982. The waterhyacinth weevils. Aquatics 4(2): 8.
COULSON, J. R. 1971. Prognosis of control of waterhyacinth by arthropods. Haycinth
Control J. 9(1): 31-34.
DANIEL, W. W. 1978. Biostatistics: a foundation for analysis in the health sciences.
2nd ed. John Wiley & Sons, New York. 504 pp.
Agricultural impact of a winter population of blackbirds and starlings. Wilson
Bull. 90(1): 31-44.
GODFREY, R. K., and J. W. WOOTEN. 1979. Aquatic and wetlands plants of the
southeastern United States: Monocotyledons. Univ. Georgia Press, Athens,
Georgia. 712 pp.
GODFREY, R. K. AND J. W. WOOTEN. 1981. Aquatic and wetlands plants of the
southeastern United States: Dicotyledons. Univ. Georgia Press, Athens, Geor-
gia. 933 pp.
GORDON, R. D. AND J. R. COULSON. 1969. Report on field observations of arthropods
on waterhyacinth in Florida, Louisiana and Texas. Aquatic Plants Control Prog.,
U.S. Army Engineers, Vicksburg, Mississippi. Tech. Rep. 6: B3-B37.
HAAG, K. H. 1986a. Effective control of waterhyacinth using Neochetina and limited
herbicide application. J. Aquat. Plant Manage. 24: 70-75.
HAAG, K. H. 1986b. Effects of herbicide application on mortality and dispersive be-
havior of the waterhyacinth weevils, Neochetina eichhorniae and Neochetina
bruchi (Coleoptera:Curculionidae). Environ. Entomol. 15(6): 1192-1198.
HANSEN, K. L., E. G. RUBY, AND R. L. THOMPSON. 1971. Trophic relationships in
the waterhyacinth community. Quart. J. Florida Acad. Sci. 34(2): 107-113.
HOWELL, A. H. 1932. Florida bird life. Coward-McCann, Inc., New York. 579 pp.
KATZ, E. E. 1967. Effects of the chemical eradication of waterhyacinths on associated
aquatic fauna. M.S. Thesis, Stetson Univ., Deland, Florida. 52 pp.
KNOBEL, E. AND M. E. FAUST. 1977. Field guide to the grasses, sedges, and marshes
of the United States. Dover Publications Inc., Syracuse, New York.
KORSCHGEN, L. J. 1980. Procedures for food-habits analyses, pp. 113-127 in S. D.

470 Florida Entomologist 70(4) December, 1987

Schemnitz, ed. Wildlife Management Techniques Manual. The Wildl. Soc., Wash-
ington, D.C.
LYNCH, J. J., J. E. KING, T. K. CHAMBERLAIN, AND A. L. SMITH. 1947. Effects
of aquatic weed infestations on the fish and wildlife of the Gulf states. U.S. Dept.
Interior Fish/Wildlife Serv. Special Sci. Report, No. 39.
MARTIN, A. C. AND W. D. BARCLAY. 1973. Seed identification manual. 2nd edition.
University of California Press. Berkeley, California. 221 pp.
MARTIN, A. C., R. H. GENSCH, AND C. P. BROWN. 1946. Alternative methods in
upland gamebird food analysis. J. Wildl. Manage. 10: 8-12.
MERRITT, R. W. AND K. W. CUMMINS. 1978. An introduction to the aquatic insects
of North America. Kendall/Hunt Pub. Co., Dubuque, Iowa. 441 pp.
MULHOLLAND, R. 1983. Feeding ecology of the common moorhen and purple gallinule
on Orange Lake, Florida. M.S. thesis. Univ. Florida, Gainesville. 79 pp.
MULHOLLAND, R. AND H. F. PERCIVAL. 1983. Food habits of the common moorhen
and purple gallinule in northcentral Florida. Proc. S. E. Assoc. Fish and Wildl.
Agencies. 36: 527-536.
O'HARA, J. 1961. The invertebrate fauna associated with waterhyacinths in South
Florida. M.S. Thesis, Univ. Miami, Miami, Florida. 66 pp.
O'HARA, J. 1968. Invertebrates found in waterhyacinth mats. Quart. J. Florida. Acad.
Sci. 30(1): 73-80.
O'MEARA, T. E., W. R. MARION, O. B. MEYERS, AND W. M. HETRICK. 1982. Avian
food habits on phosphate-mine settling ponds and natural wetlands. Proc. S. E.
Assoc. Fish. Wildl. Agencies 36: 515-526.
ORIANS, G. H. 1973. The red-winged blackbird in tropical marshes. Condor. 75: 28-42.
OTTO, C. 1985. Effects of temporal and spatial variations in food availability on life
cycle and palatability of a chrysomelid beetle (Donacia cinerea). Aquatic Insects
7: 19-28.
PENNAK, R. W. 1978. Freshwater invertebrates of the United States 2nd ed. John
Wiley and Sons, New York. 803 pp.
PERKINS, B. D. 1973. Release in the United States of Neochetina eichhorniae
(Warner), an enemy of waterhyacinth. Proc. 26th Ann. Meet. S. Weed Sci. Soc.
26: 368 (abstract).
PERKINS, B. D. 1974. Arthropods that stress waterhyacinth. PANS 29(3): 304-314.
REAGAN, W. 1977. Resource partitioning of North American gallinules in southern
Texas. M.S. Thesis, Utah State Univ., Logan, Utah. 72 pp.
SHIREMAN, J. V. AND W. J. HALLER. 1979. The ecological impact of integrated
chemical and biological aquatic weed control. Ann. Rept. Aug. 1 1978-July 31
1979 to Environ. Protection Agency. Inst. Food and Agric. Sci. Aquatic Weed
Center, Univ. of Florida, Gainesville, Florida.
AND D. E. CANFIELD JR. 1983. Ecological impact of integrated chemical and
biological weed control. No. PB 83-264242 Nat. Tech. Inform. Serv. Springfield,
Virginia. 332 pp.
SIMPSON, T. W. 1939. The feeding habits of the coot, Florida gallinule, and least
bittern on Reelfoot Lake. J. Tennessee Acad. Sci. 14: 110-115.
SNELLING, J. C. 1968. Overlap in feeding habits of red-winged blackbirds and common
grackles nesting in a cattail marsh. Auk 85: 560-585.
SPENCER, N. R. 1974. Insect enemies of aquatic weeds. PANS 20(4): 444-450.
SPRUNT, A., JR. 1954. Florida bird life. Coward-McCann, New York. 527 pp.
1974. Advantages in mathematically weighting waterfowl food habits data. J.
Wildl. Manage. 38: 302-307.
TARVER, D. P., J. A. RODGERS, M. J. MAHLER, R. L. LAZOR. 1986. Aquatic and
wetland plants of Florida, 3rd edition. Florida Dept. of Natural Resources. Tal-
lahassee. Florida. 127 pp.
VOIGHTS, D. K. 1973. Food niche overlap of Iowa marsh icterids. Condor. 75: 392-329.
WAHLQUIST, H. 1969. Effect of waterhyacinths and fertilization on fish-food or-

Callahan & Morris: Survey of Aquatic Diptera

ganisms and production of blue-gill and redear sunfish in experimental ponds.
Proc. Southeastern Assoc. Game Fish Comm. 23: 373-384.
WETMORE, A. 1916. Birds of Puerto Rico. U.S. Dept., Agric., Bull. No. 326.
Sa 0 a -a a -a


Polk County Environmental Services
P.O. Box 39
Bartow, FL 33830


A survey was conducted at 30 sites in 13 lakes in Polk County, Florida to: (1)
determine the extent of mosquito and midge production in the lake sites, (2) define more
precisely some of the environmental factors that influence mosquito and midge densities,
and (3) determine if the physical appearance of a site would enable accurate estimates
of the mosquito and midge production potential. Although the emphasis of this survey
was on Coquillettidia perturbans (Walker), data were collected on the mosquitoes
Anopheles, Culex, Mansonia, and Uranotaenia and the chironomid midge Glyptoten-
dipes paripes Edwards. Results of previous studies in marshes indicate that higher
numbers of Coquillettidia larvae are found in sites with (1) dense stands of emergent
vegetation, (2) substrates characterized by a thick layer of detritus, (3) water depths
less than 1 m, and (4) dense stands of shoreline vegetation. Production in sites with
these characteristics was found to be greater than previously realized. Estimates of
Coquillettidia production made at the beginning of the study were found to be correct
in 20 of the 30 instances. Production potentials that were over- or under-estimated were
due primarily to changes in trap locations (due to fluctuating water levels) in which
dramatic changes in habitat conditions occurred. It appears that, with only cursory
observation, an accurate estimate of a site's potential for mosquito and midge production
can be made.


Se condujo un reconocimiento de 30 lugares en 13 lagos en el Condado de Polk,
Florida para: (1) determinar el grado de producci6n de mosquitos y de moscas de agua
en localidades en el lago, (2) definir con mas precision algunos de los factors ambientales
que influyen la densidad de los mosquitos y de las moscas de agua, y (3) determinar si
la apariencia fisica de un lugar pudiera facilitar estimadas del potential de producci6n
de mosquitos y de moscas de agua. Aunque el 6nfasis de este reconocimiento se le di6
a Coquillettidia perturbans (Walker), se colectaron datos de los mosquitos Anopheles,
Culex, Mansonia, y Uranotaenia y la quirin6mida mosca de agua Glytotendipes paripes
Edwards. Resultados de previous studios en pantanos indicaron que se encontraron
altos nimeros de larvas de Coquilletidia en lugares con (1) densa emergente vegetaci6n,
(2) substratos caracterizados por un capa densa de detritus, (3) profundidad de agua
menos de 1 m, y (4) densa vegetaci6n en la orilla. La producci6n en lugares con estas
caracteristicas fu6 mayor de lo que previamente se creia. Se encontr6 que estimados
de producci6n de Coquillettidia hechos al principio de este studio eran correctos en 20
de cada 30 veces. Los potenciales de producci6n que se sobre-estimaron o bajo-estima-
ron, fueron debidos principalmente a combios de lugar de las trampas (debido a los
combios del nivel del agua) en los cuales ocurrieron cambios dramaticos en las con-

Florida Entomologist 70(4)

diciones de habitaci6n. Aparentemente, con solo observaciones superficiales se puede
hacer un estimado del potential de producci6n de mosquitos y de moscas de agua de un

The marshy conditions in the littoral zones of many central Florida lakes appear
conducive to the production of large numbers of mosquitoes and midges (Beck & Beck
1969, Ali 1980, Morris et al. 1985, Callahan & Morris 1987). The nutrient rich waters,
thick layers of non-compacted organic matter, and dense stands of aquatic plants with
roots loosely penetrating the detrital layers are conditions that many mosquitoes and
chironomid midges find attractive (McNeel 1932, Bidlingmayer 1968, Beck & Beck 1969,
Guille 1976). Judging from the complaints from residents living adjacent to these lakes,
large numbers of pestiferous mosquitoes are periodically produced in these bodies of
water. While mosquito production in the lakes may not have as great a potential as that
of natural marshes or phosphate mined areas, the 37,665 ha covered by the 550 fresh-
water lakes in Polk County provide numerous habitats for mosquito larvae. This survey
was conducted in an urban area to (1) estimate mosquito and midge production in
selected lake sites, (2) define more precisely some of the environmental factors that
influence their densities, and (3) determine if the characterization of the physical appear-
ance of a site would enable accurate estimates of the mosquito and midge production


Thirty sampling sites, ranging in size from 0.1 to 1.4 ha, were selected in 13 lakes
with surface areas from 10 to 298 ha. Larval mosquitoes were collected with the wand
and pump system (Morris et al. 1985) approximately once every 4 weeks from April 23
to August 23, 1985. Adult mosquitoes and midges were collected with emergence traps
(Slaff et al. 1984) from April 9 to November 4, 1985. Trap tops were collected once
every 2 weeks. Emergence traps in several lakes were moved farther from shore than
originally planned when lack of rainfall caused lake levels to decline.
Results of previous studies in marshes indicated that the better sites for Coquillet-
tidia perturbans (Walker) larvae are those that have: (1) water depths less than 1 m,
(2) dense stands of shoreline vegetation (Morris & Callahan, unpublished), (3) dense
stands of emergent vegetation, and (4) substrates characterized by thick layers of de-
tritus (McNeel 1932, Bidlingmayer 1968, Batzer & Sjogren 1986). At each site, observa-
tions were made on the density and height of shoreline vegetation, species and density
of the aquatic vegetation, water depth, and bottom type. These observations were made
without prolonged or complicated procedures or highly sophisticated sampling devices
in an attempt to mimic the methods that are or could be used by typical mosquito
Sites with shorelines dominated by shrubs or trees less than 3 m high with little or
no understory, pastures, highways, or other exposed areas were classified as "open".
Sites with shorelines dominated by woods with trees taller than 3 m and dense under-
stories of herbs, shrubs, and vines were classed as "wooded".
Sampling sites were confined to monocultures or co-dominant stands of aquatic
plants (Table 1). Most of the sites contained only cattails (Typha spp.). Several sites
had cattails mixed with pickerelweed (Pontederia cordata Linn.), maidencane (Panicum
hemitomon Schult.), or water hyacinth (Eichhornia crassipes (Mart.) Solms.). The re-
maining sites contained monocultures of maidencane or sedge (Carex spp.) or were a
mixture of water hyacinth and sedge. Density of the aquatic plants was recorded as
"sparse" when the airboat easily moved through the site and "dense" when the plants
tended to impede the boat's progress.

December, 1987

Callahan & Morris: Survey of Aquatic Diptera


Site Expected
size Shore Site veg.a Bottom production
Lake-Site (ha) veg. /density type of Coq.

Lk Alfred-NE
Lk Alfred-SW
Lk Cannon-NW
Lk Hartridge-NE
Lk Hartridge-NW
Lk Hartridge-S
Lk Howard-SW
Lk Idylwild-N
Lk Idylwild-SE
Lk Idylwild-SW
Lk Jessie-N
Lk Jessie-SE
Lk Lulu-NW
Lk Lulu-SE
Lk Lulu-SW
Lk Mariam-ECT
Lk Mariam-EMX
Lk Mariam-N
Lk Mariam-NE
Lk Mariam-SW
Lk Mariam-W
Lk Marianna-N
Lk Marianna-SE
Lk Marianna-W
Lk Mirror-N
Lk Rochelle-NE
Lk Shipp-S
Lk Spring-E
Lk Spring-N
Lk Spring-NW



aCT= cattail, CX= sedge, MC maidencane, PW= pickerelweed, WH= water hyacinth

Bottom type was determined by using a pole to probe the substrate. Hard, barely
impenetrable substrates were labeled "sand", while soft, easily probed deposits were
classified as "muck" or "detritus", depending on grab or core samples (APHA 1981)
and/or visual observation.
Using the criteria listed above, 1 to 6 sampling sites were selected on each lake in
areas judged to have the greatest potential of producing Coquillettidia. All sites had
water depths less than 1 m. The other 3 habitat characteristics associated with high
numbers of Coquillettidia larvae (tall, dense shoreline vegetation; dense stands of aqua-
tic vegetation; thick layers of detritus) were used to classify the 30 sites as "expected
good" or "expected fair" for potential Coquillettidia production (Table 1). Sites with all
3 characteristics were labeled "expected good"; all other sites were classified as "ex-
pected fair". Even though a site may have been selected as the best or one of the best
available on a particular lake, it still may have received only an "expected fair" rating
in terms of Coquillettidia production. Since trap locations were changed due to fluctu-
ations in the water levels, site characteristics were re-evaluated at the conclusion of the
survey; expected production potentials were not revised.



474 Florida Entomologist 70(4) December, 1987

Although the emphasis in this study was on Coquillettidia, data were also collected
on the mosquitoes Anopheles crucians Wiedemann, An. quadrimaculatus Say, Culex
erraticus (Dyar and Knab), Cx. pilosus (Dyar and Knab), Cx.nigripalpus Theobald,
Cx. salinarius Coquillett, Mansonia titillans (Walker), Uranotaenia lowii Theobald,
and Ur. sapphirina (Osten Saken) and the chironomid midge Glyptotendipes paripes
A fungus problem on the emergence trap tops prevented identification to species in
many cases. As a result, analyses were performed using mosquito genera instead of
individual species. Adult mosquito and midge production is presented as the mean
number collected per day per hectare. Data were analyzed on an IBM PC XT using the
SPSS/PC + statistical program (Norusis 1986). The F-test was used to determine
homogeneity of variance. The Mann-Whitney U test and the Kruskal-Wallis one-way
analysis of variance were used to determine significant differences in habitat charac-
teristics and in mosquito and midge numbers between the sites. The Pearson product-
moment correlation was used to determine relationships between mosquito and midge
numbers at the sites (Steele & Torrie 1980, Norusis 1986).


Over 46% of the adults collected on the emergence trap tops were Culex (Table 2).
Uranotaenia were the next most frequently collected mosquitoes (24.6%); Anopheles
and Coquillettidia were collected in approximately equal numbers (14.2% and 12.9%,
respectively), while Mansonia were rarely collected (2.1%).
As shown in Table 3, Lake Alfred-SW consistently ranked high for all genera of
mosquitoes. Lake Howard-SW was a prime site for Culex and all genera except Coquil-
lettidia. Lake Mariam-NE and Lake Mariam-SW ranked high for Coquillettidia and all
genera except Culex. The top ranked sites for Coquillettidia (Lake Alfred-SW, Lake
Mariam-NE, Lake Mariam-SW, and Lake Hartridge-NW) also had high rankings for
total mosquitoes (ranks 1, 3, 6, and 4, respectively).
Although Anopheles, Culex, Mansonia, and Uranotaenia were found in higher num-
bers in the wooded sites, their numbers were not much higher than those found in sites
adjacent to open shorelines. Coquillettidia were the only mosquitoes found in signific-
antly greater numbers in sites adjacent to wooded shorelines. Mosquitoes in general
were found in higher numbers in those sites with detritus and/or in those with dense
stands of emergent aquatic vegetation.
Only 3 of the 10 sites expected to produce large numbers of Coquillettidia actually
had a high level of production (Table 3). Of the 7 that were found to be "fair" production
sites for Coquillettidia, the changes in emergence trap locations appear to have affected
production results. At the conclusion of the study, traps in 5 of these 7 sites (Lake
Hartridge-NE, Lake Idylwild-N, Lake Idylwild-SE, Lake Idylwild-W, and Lake Mir-
ror-N) had been moved to sand bottom locations. The other 2 traps in Lake Mariam-ECT
and Lake Mariam-EMX had been relocated to sparse stands of emergent vegetation.
Seventeen of the 20 sites originally labeled "expected fair" had moderate or poor
production of Coquillettidia (Table 3). The 3 exceptions were Lake Alfred-NE, Lake
Hartridge-NW, and Lake Mariam-N. The emergence trap at the Lake Mariam-N loca-
tion was moved from a muck bottom to a detrital bottom during the survey. Those at
Lake Alfred-NE and Lake Hartridge-NW were moved to locations with more dense
emergent vegetation or thicker layers of detritus. These changes appeared to have
made these sites more conducive to Coquillettidia than originally estimated.
The presence or absence of a thick layer of detritus was found to be a very important
factor when considering the production potential of a site for Coquillettidia. This impor-

Callahan & Morris: Survey of Aquatic Diptera


Lake-Site Anoph. Culex Mans. Coq. Uran. Total Midges n

"Expected good" sites
Lk Alfred-SW
Lk Hartridge-NE
Lk Idylwild-N
Lk Idylwild-SE
Lk Idylwild-SW
Lk Mariam-ECT
Lk Mariam-EMX
Lk Mariam-NE
Lk Mariam-SW
Lk Mirror-N
"Expected fair" sites
Lk Alfred-NE
Lk Cannon-NW
Lk Hartridge-NW
Lk Hartridge-S
Lk Howard-SW
Lk Jessie-N
Lk Jessie-SE
Lk Lulu-NW
Lk Lulu-SE
Lk Lulu-SW
Lk Mariam-N
Lk Mariam-W
Lk Marianna-N
Lk Marianna-SE
Lk Marianna-W
Lk Rochelle-NE
Lk Shipp-S
Lk Spring-E
Lk Spring-N
Lk Spring-NW





0 0 0 163 0
0 0 0 0 0
426 304 0 243 61
210 2031 0 0 0
272 10687 0 0 3608
0 348 116 0 348
0 174 0 116 58
127 127 0 0 127
149 0 0 0 74
316 519 0 0 104
524 262 0 327 131
327 131 65 65 196
122 122 0 61 0
159 317 0 0 79
0 0 0 0 0
121 182 0 61 242
0 0 69 0 69
0 484 0 0 0
0 0 0 0 0
0 0 0 0 0
138 784 13 52 255

tance has been reported in previous studies by McNeel (1932), Bidlingmayer (1968),
Guille (1976), and Batzer & Sjogren (1986) and should have a more dominant role than
the aquatic plant species when sites are delineated for study.


Only 126 mosquito larvae were collected in the 965 larval samples. Since the numbers
were so low, statistical analyses were not performed on this data.
Although the wand and pump system was developed for collecting Coquillettidia
and Mansonia larvae (Morris et al. 1985), over 60% of the larvae (76 larvae) collected






Florida Entomologist 70(4)


Lake-Site Anoph. Culex Mans. Coq. Uran. Total Midges

Lk Alfred-NE 18 20 6 5 18 21 25
Lk Alfred-SW 3 2 1 1 1 1 15
Lk Cannon-NW 18 20 6 11 18 26 2
Lk Hartridge-NE 10 19 6 10 18 16 6
Lk Hartridge-NW 6 4 6 4 17 4 19
Lk Hartridge-S 13 6 6 11 18 12 18
Lk Howard-SW 5 1 6 11 4 2 27
Lk Idylwild-N 18 20 6 11 18 26 12
Lk Idylwild-SE 18 16 6 11 18 22 7
Lk Idylwild-SW 4 5 6 11 12 8 13
Lk Jessie-N 18 7 4 11 7 13 29
Lk Jessie-SE 18 15 6 7 17 18 11
Lk Lulu-NW 17 16 6 11 16 17 10
Lk Lulu-SE 16 20 6 11 13 23 3
Lk Lulu-SW 9 3 6 11 10 7 21
Lk Mariam-ECT 11 10 6 10 9 11 23
Lk Mariam-EMX 14 8 6 9 2 5 30
Lk Mariam-N 7 13 6 6 8 10 26
Lk Mariam-NE 1 11 2 2 5 3 20
Lk Mariam-SW 2 16 5 3 3 6 14
Lk Mariam-W 8 14 3 8 6 9 24
Lk Marianna-N 15 16 6 10 18 19 16
Lk Marianna-SE 12 12 6 11 11 15 22
Lk Marianna-W 18 20 6 11 18 26 28
Lk Mirror-N 18 18 6 11 18 25 17
Lk Rochelle-NE 15 17 6 10 14 14 9
Lk Shipp-S 18 20 3 11 15 24 8
LkSpring-E 18 9 6 11 18 20 5
LkSpring-N 18 20 6 11 18 26 1
Lk Spring-NW 18 20 6 11 18 26 4

in this study were Culex. Approximately 25% (31) were Coquillettidia, 6% (8) were
Anopheles, 6% (8) were Mansonia, and 2% (3) were Uranotaenia.
In general, greater numbers of larvae were found at those sites associated with high
numbers of adult mosquitoes. Almost one-half of the Coquillettidia larvae (14) were
collected at Lake Alfred-SW. The Lake Mariam-SW site accounted for 15 of the 31
Coquillettidia collected. Most of the Culex(51 of the 78 larvae) were collected at Lake


All midges were identified as Glyptotendipes paripes Edwards, the predominant
nuisance species of chironomid midge in Florida (Beck & Beck 1969, Ali 1980).
At the conclusion of the study, midge numbers were only slightly higher in those
sites with open shorelines than in those with wooded shorelines. Patterson (1964, 1965)
noted that chironomid midge larvae are found predominantly in the peripheral sandy
bottom areas of Florida lakes, many having a larval density of 500 or more per square
foot. In this study, midge numbers were also found to be much higher in sites with sand

December, 1987

Callahan & Morris: Survey of Aquatic Diptera

bottoms. Midge numbers were only slightly higher at those sites with sparse stands of
emergent vegetation.
The mean number of midges collected from those sites labeled "expected fair" for
Coquillettidia production was typically much greater than the numbers collected from
the "expected good" Coquillettidia production sites (Tables 2 and 3). The most produc-
tive midge sites, Lake Spring-N and Lake Cannon-NW, were 2 of the 5 sites where no
adult mosquitoes of any species were collected, even though the production of Coquillet-
tidia in these sites was expected to be moderate.
Those sites with open shorelines, sparse stands of emergent vegetation, and sand
bottoms usually had lower mosquito and higher midge production (r = -0.1005, p <
0.01) (Table 3). Conversely, sites with higher rankings for Coquillettidia and total
mosquitoes usually had lower numbers of midges. There were only 2 sites (Lake Alfred-
NE and Lake Marianna-W) in which both midge and total mosquito production were
extremely low.


Results of this survey indicate that lakes are capable of producing large numbers of
both man-biting mosquitoes and midges. In inhabited areas where lakes are a dominant
part of the topography, this production can significantly increase the extent of the
mosquito and midge problems.
Of the 2 sampling methods used in this survey, the emergence trap was more effi-
cient and produced a better estimate of the production in a lake. This is not surprising
since the emergence trap samples are continuous while the wand and pump samples are
temporal. Emergence traps are inexpensive to construct and operate and provide reli-
able results over a lengthy period of time. With both mosquitoes and midges, frequent
collection of the trap tops provides an accurate picture of the species compositions and
Results of this lake study corroborate those of previous marsh studies (McNeel 1932,
Bidlingmayer 1968, Morris & Callahan, unpublished) in which the highest numbers of
Coquillettidia perturbans were found to be produced at sites (1) less than 1 m deep, (2)
where the lake bottom is covered by a thick layer of detritus, (3) with dense stands of
emergent aquatic vegetation, and (4) that are adjacent to shorelines dominated by tall,
dense vegetation. These conditions are also conducive to the production of larger num-
bers of other man-biting mosquitoes. This high level of production along the lakeshores
probably creates more of the mosquito problem in these areas than previously realized.
Chironomid midges appear to have a wider tolerance to environmental conditions
than mosquitoes (Ali 1980), but in this study the numbers were found to be generally
higher in sites (1) with predominantly sand bottoms, (2) with sparse stands of aquatic
vegetation, and (3) that are adjacent to sparsely vegetated shorelines.
Littoral zones are extremely diverse and it is usually not possible to classify an
entire lake as being a "good" or "fair" environment for mosquitoes or midges. However,
in the majority of cases, using simple methods to classify aquatic and shoreline vegeta-
tion, water depth, and bottom type, it is possible to easily and quickly examine clearly
defined sections along lake edges and produce accurate estimates of the mosquito and
midge production potential for control purposes.


ALI, A. 1980. Nuisance chironomids and their control: a review. Bull. Entomol. Soc.
Am. 26: 3-16.
AMERICAN PUBLIC HEALTH ASSOCIATION. 1981. Standard methods for the exami-

Florida Entomologist 70(4)

December, 1987

nation of water and wastewater, 15th ed. APHA-AWAA-WPCF, Washington,
BATZER, D. P. AND R. D. SJOGREN. 1986. Larval habitat characteristics of Coquil-
lettidia perturbans (Diptera: Culicidae) in Minnesota. The Canadian En-
tomologist. 118: 1193-1198.
BECK, E. C. AND W. M. BECK, JR. 1969. The Chironomidae of Florida. II. The
nuisance species. Florida Entomol. 52: 1-11.
BIDLINGMAYER, W. L. 1969. Larval development of Mansonia mosquitoes in central
Florida. Mosq. News 28: 51-57.
CALLAHAN, J. L. AND C. D. MORRIS. 1987. Habitat characteristics of Coquillettidia
perturbans in central Florida. J. Am. Mosq. Control Assoc. 3: 176-180.
GUILLE, G. 1976. Recherches eco-ethologiques sur Coquillettidia (Coquillettidia)
richiardii (Ficalbi), 1889 (Diptera-Culicidae) du littoral Mediterraneen Francais.
II. Milieu et compartment. Ann. Sci. Natur. Zool., Paris. 18: 5-112.
MCNEEL, T. E. 1932. Observations on the biology of Mansonia perturbans (Walk.)
Diptera, Culicidae. Proc. N.J. Exterm. Assoc. 19: 89-96.
MORRIS, C. D., J. L. CALLAHAN, AND R. H. LEWIS. 1985. Devices for sampling
and sorting immature Coquillettidia perturbans. J. Am. Mosq. Control Assoc.
1: 247-250.
NORUsIs, M. J. 1986. SPSS/PC + for the IBM/PC/XT/AT. SPSS Inc., Chicago, IL.
PATTERSON, R. S. 1964. Recent investigations on the use of BHC and EPN to control
chironomid midges in central Florida. Mosq. News 24: 294-299.
PATTERSON, R. S. 1965. Control of chironomid midges in Florida. Florida Anti-Mosq.
Assoc. Ann. Rept. 36: 35-39.
SLAFF, M., J. D. HAEFNER, R. E. PARSONS, AND F. WILSON. 1984. A modified
pyramidal emergence trap for collecting mosquitoes. Mosq. News 44: 197-199.
STEELE, R. G. D. AND J. H. TORRIE. 1980. Principles and procedures of statistics:
a biometrical approach, 2nd ed. McGraw-Hill Book Company, New York, NY.


Entomology and Nematology Department, 3103 McCarty Hall,
University of Florida, Gainesville, Florida 32611, USA

Division of Plant Industry, Florida Department of Agriculture and Consumer Services,
P.O. Box 1269, Gainesville, Florida 32602, USA


Departamento de Sanidad Vegetal, CESDA, San Crist6bal,


Scapteriscus didactylus (Latreille) is reported for the first time from the Dominican
Republic, where it is widespread in coastal localities and one inland locality sampled.
Materials identified from guts of adults and nymphs from the inland and 4 coastal


Frank et al.: Dominican Mole Cricket 479

localities, all sampled in October 1986, were mainly of arthropod origin. Guts from 4 of
5 specimens from one locality contained nematodes of the genus Chitwoodiella which is
not considered to be pathogenic. The date of introduction of S. didactylus into His-
paniola is unknown, but an historical account suggests its presence in Puerto Rico in


Scaoteriscus didactylus (Latreille) es reportado por primera vez en la Repdblica
Dominicana, donde se encuentra distribuido en zonas costeras y en una localidad inves-
tigada en el interior. Material identificado en el intestine de adults y ninfas del interior
y de 4 localidades del litoral, todas de muestras hechas en Octubre del 1986, fueron
principalmente de origen artr6podo. Intestinos de 4 de 5 especimenes de una de esas
localidades conteniAn nematodos del g6nero Chitwoodiella el cual no es considerado
patog6nico. La fecha de introducci6n del S. didactylus en La Espafola es desconocida,
pero una ficha hist6rica sugiere su presencia en Puerto Rico en el 1797.

Changa is a colloquial name used in Puerto Rico for the mole cricket Scapteriscus
didactylus (Latreille). The same name is used in the Dominican Republic where S.
didactylus is also a pest, though neither its presence nor the damage it causes seem to
have been recorded. Collections made in 1986 at 7 localities in the Dominican Republic
yielded information on distribution, habitat, food, damage, and a parasite of S. didac-
tylus, with some preliminary information on seasonality.


Mole crickets were collected by digging with a trowel or pocket knife and were
preserved in 70 isopropyl alcohol, at the localities listed below. Localities 3-7 yielded
nymphs and adults (m = male, f = female, n = nymph), and all specimens seen there
were collected, thus giving information on population structure. Pronotal lengths of
nymphs were measured with a microcaliper to provide an indication of size, because the
number and diagnostic characters of the nymphal instars have not been determined;
nymphs are listed in descending order of pronotal length in mm. Five specimens (arbit-
rarily numbered N.1-N.5) from each of sites 3-7 were dissected, and dissected nymphs
were classed for size as small (<5 mm), medium (5-7 mm), large (>7 mm). Digestive
tracts were examined for content. Most tracts contained unidentified, brown silty mate-
rials ('soil') which were not evidently cellular and could not be attributed to plant or
animal origin; identified materials are listed. The presence or absence of eggs in abdo-
mens of adult females was noted.


1. PROV. SAN CRISTOBAL, Playa Palenque, 13-III-1986, J. Cicero [2m].
2. PROV. MONSENOR NOUEL, Bonao, at light in restaurant at night, 19-VI-1986,
L. A. Stange and R. E. Woodruff [1f].
3. PROV. AZUA, 2 km S. of Hato Nuevo, in galleries in recently-irrigated field of
green peppers, during day, 23-X-1986, J. H. Frank and C. A. Nufiez [8m, 5f, 28n (7.8,
7.1, 6.4, 6.2, 6.1, 6.1, 6.1, 5.9, 5.8, 5.5, 5.2, 4.9, 4.9, 4.3, 4.1, 4.0, 3.9, 3.9, 3.8, 3.6, 3.5,
3.5, 3.4, 3.2, 3.1, 3.1, 2.4, 2.1)]. Dissections: 3.1 [female, not gravid, with sand grains
and a few insect fragments], 3.2 [male, with few plant fibers and sand grains and insect
fragments including an almost intact specimen of Thecturota sp. (Staphylinidae:
Aleocharinae)], 3.3 [medium nymph, with soil and a few plant fibers and insect frag-

Florida Entomologist 70(4)

December, 1987

ments], 3.4 [medium nymph, with insect fragments, and very few plant fibers], 3.5
[medium nymph, with many plant pieces which were neither green nor fibrous, probably
roots of pepper plants, without insect fragments].
4. PROV. BARAHONA, Playa Saladilla, in galleries in sand high on sea beach, mostly
under cow dung, during day, 25-X-1986, J. H. Frank [1m, 3f, 7n (8.9, 8.8, 8.3, 7.8, 7.6,
7.6, 5.9)]. Dissections: 4.1 [female, gravid, with soft, greenish-brown material, perhaps
cow dung, with sand grains and fragments of a minute beetle], 4.2 [large nymph, with
sand grains and insect fragments], 4.3 [large nymph, with sand grains and many frag-
ments of a spider (possibly a lycosid)], 4.4 [large nymph, with a few, small insect
fragments], 4.5 [female, not gravid, with soft, greenish-brown material, perhaps cow
dung, with a few insect fragments, perhaps a fly larva, and sand grains].
5. DISTRITO NACIONAL, Boca Chica, in galleries in sand at base of small buildings
for vending food, ca. 20 m from sea beach, at night, 27-X-1986, J. H. Frank [5f, 4n (7.9,
3.7, 3.6, 2.9)]. Dissections: 5.1 [female, not gravid, with sand grains, without arthropod
fragments but with small, semi-soft, white particles which might be of rice grains], 5.2
[female, gravid, with sand grains and parts of an ant (Solenopsis geminata {Fabricius})
and with semi-soft white particles which might be of rice grains], 5.3 [female, gravid,
with sand grains], 5.4 [small nymph, with sand grains and insect fragments, and with
semi-soft white particles which might be of rice grains], 5.5 [small nymph, with sand
grains and a few, small insect fragments, and with semi-soft white particles which might
be of rice grains].
6. PROV. LA ALTAGRACIA, Nisibon, in galleries in sand around small buildings ca.
20 m from sea beach, during daylight, 30-X-1986, J. H. Frank [2m, 3f, 2n (8.0, 5.2)].
Dissections: 6.1 [large nymph, with sand grains and a few fragments of an ant (Tet-
ramorium simillimum {F. Smith})], 6.2 [male, with sand grains and a few fragments
of an ant (T. simillimum)], 6.3 [female, not gravid, with sand grains and very few
fragments of an ant (Solenopsis {Diolorhoptrum} sp.)], 6.4 [female, not gravid, with
sand grains and a few fragments of an ant (Solenoosis {Diplorhoptrum} sp.)], 6.5 [male,
with sand grains and a few fragments of an ant (T. simillimum)].
7. PROV. PUERTO PLATA, Playa Cabarete, in galleries in small sand dunes over-
grown with Jacquemontia reclinata House (beach morning glory) vines, during day,
31-X-1986, J. H. Frank [1m, 4f, 4n (7.8, 6.6, 5.4, 5.1)]. Dissections: 7.1 [female, not
gravid, with sand grains and a few plant fibers and many nematodes (Chitwoodiella
sp.)], [7.2 large nymph, with a few sand grains, insect fragments, and nematodes (Chit-
woodiella sp.)], 7.3 [male, with insect fragments and sand grains and a few plant fibers,
without nematodes], 7.4 [female, gravid, with plant fibers and sand grains and nema-
todes (Chitwoodiella sp.)], 7.5 [medium nymph, with many nematodes (Chitwoodiella
sp.) and a few sand grains and plant fibers].


Damage to pepper plants was observed at locality 3, where most plants were 20-30
cm tall. Some of the plants had collapsed following damage to underground parts of the
stems and roots. Mole cricket galleries were numerous at the soil surface. Slight damage
was seen, with fewer galleries, in a nearby, irrigated field of egg plant. The digestive
tract of 1 of the 5 mole crickets dissected from this locality appeared to contain fresh
plant materials, and tracts of 3 others contained smaller quantities of plant fibers which
were more digested.
Size distribution is shown in Fig. 1. Locality 3 was a habitat disturbed by cultivation
and irrigation, and most nymphs there were small. The other habitats were relatively
undisturbed. Overall, 4 of the 10 females dissected were gravid with well-developed
eggs (representing localities 4, 5, and 7), suggesting ability of populations to expand in


Frank et al.: Dominican Mole Cricket

0 10-


<5 5-7 >7 f m

nymphs adults

Fig. 1. Size and age distribution of Scapteriscus didactylus specimens collected in
the Dominican Republic in 1986. Shaded parts of bars show numbers and stages of
specimens from localities 4-7 (see text) and unshaded parts show the same for locality
3. Stages of nymphs are based on pronotal lengths in mm.

October under conditions favorable for survival of the immature stages. Additional data
on size distribution of nymphs and content of ovarioles of females are necessary to
answer demographic questions which could help in control attempts.
Castner & Fowler (1984b) found S. didactylus widespread throughout coastal areas
of Puerto Rico, and abundant at scattered inland localities. The same appears to be true
of its distribution in the Dominican Republic: it was present but not abundant at the
coastal localities 1 and 4-7, and was abundant at the inland locality 3.
The nematode parasites (Chitwoodiella sp., Chitwoodiellidae, Oxyurida) found in
the hindgut of 4 of the 5 dissected mole crickets from locality 7 are not considered to
be pathogenic (K. B. Nguyen, pers. comm.).
In 23 of the 25 digestive tracts examined there were arthropod fragments or mate-
rials which might be attributed to living plants (excluding cow dung and fragments of
rice grains), or both. In 15 of the tracts there were arthropod fragments, in 4 a mixture
of arthropod fragments and plant materials, and in 4 plant materials only (the ratio is

482 Florida Entomologist 70(4) December, 1987

15:4:4). These findings contrast with those of Castner & Fowler (1984a) who found a
ratio of 1:1:30 in digestive tracts of S. didactylus from Puerto Rico. However, the
specimens from Puerto Rico were all taken from a single locality at a golf course, where
the abundance of grass was reflected in the fact that 29 of the hindguts contained grass
blades. Localities 3-7 in the Dominican Republic had little or no grass, and it is not
surprising that the digestive tracts of mole crickets collected there had no grass. Evi-
dently the diet of S. didactylus is dependent upon habitat, and arthropods may form a
substantial portion of the diet where suitable plant materials are not abundant. By
dissection of digestive tracts, Matheny (1981) found that S. vicinus Scudder in Florida
is largely herbivorous, whereas S. acletus Rehn & Hebard is largely carnivorous. We
cannot support the contrast proposed by Castner & Fowler (1984a), that S. didactylus
is largely herbivorous and S. imitatus Nickle & Castner is omnivorous, because the
samples on which that proposal was based came from only 2 localities: one with abundant
plant material for S. didactylus and a different one for S. imitatus. If diets of mole
crickets of these 2 species are to be contrasted, then specimens dissected should prefer-
ably come from habitats where both species occur. Barrett (1902) stated that the diet
of S. didactylus in Puerto Rico consists almost entirely of living plants, but he did not
describe his method of determination; if his collections came from localities where this
mole cricket was damaging plants, then his conclusion is not surprising.
Establishment of S. didactylus in Puerto Rico was suspected by Nickle & Castner
(1984) to have occurred in the mid 1800s. A work by Ledru (1957) records results of a
French expedition to Puerto Rico in 1797 and contains 2 pertinent statements: the first
(p. 152) recorded: "La cigarra 6 topo-grillo (Achaeta grillotalpa Fab.), como una cuarta
parte mas pequefa que la de Europa"; the second (p. 172) noted "Los objetos con que
la espedicion enriqueci6 el Museo de Paris, fueron los siguentes: 450 aves empajadas,
4000 mariposas 6 insects. ." Ledru's "Achaeta grillotalpa" was equated with "Scap-
teriscus vicinus" by Wolcott (1936), and Wolcott's "Scapteriscus vicinus" with S. didac-
tylus by Nickle & Castner (1984). Thus, S. didactylus may have occurred in Puerto
Rico by 1797, and voucher specimens may be present in collections of the Museum
National d'Histoire Naturelle, Paris. The date of introduction of S. didactylus into
Hispaniola is unknown. Nickle & Castner (1984) mentioned a record from Haiti which
they were unable to confirm. The present records prove that S. didactylus does occur
on Hispaniola and is distributed in at least 7 provinces of the Dominican Republic.


We are indebted to J. E. Trager (University of Florida) for identification of ants
from mole cricket digestive tracts, to K. B. Nguyen (University of Florida) for identifi-
cation of the nematode, to G. B. Edwards (Division of Plant Industry) for identification
of the spider, to C. R. Artaud (Division of Plant Industry) for translation of the abstract
into Spanish, and to Padre J. Cicero (Inst. Politecnico Loyola, San Crist6bal) for provid-
ing mole crickets from locality 1. We thank T. J. Walker (University of Florida) and D.
A. Nickle (USDA, Washington, DC) for identification of mole crickets and manuscript
reviews. The mole cricket specimens have been deposited in the US National Museum
of Natural History and in the Florida State Collection of Arthropods. This is University
of Florida, Institute of Food and Agricultural Sciences, journal series no. 8139.


BARRETT, O. W. 1902. La change, 6 grillotalpa (Scapteriscus didactylus Latr.) en
Puerto Rico. Bol. Estac. Exp. Agric. Puerto Rico 2: 1-19.
CASTNER, J. L., AND H. G. FOWLER. 1984a. Gut content analyses of Puerto Rican

Bland: Acridid Mating Behavior 483

mole crickets (Orthoptera: Gryllotalpidae: Scapteriscus). Florida Ent. 67: 479-
AND 1984b. Distribution of mole crickets (Orthoptera: Gryllotalpidae:
Scapteriscus) and the mole cricket parasitoid Larra bicolor (Hymenoptera:
Sphecidae) in Puerto Rico. Florida Ent. 67: 481-84.
LEDRU, P. A. 1957. Viaje a la isla de Puerto Rico en el afio 1797, ejecutado por una
comisi6n de sabios franceses, de orden de su gobierno bajo la direcci6n del capitan
Nicolas Baudin, con objeto de hacer indagaciones y colecciones relatives a la
historic natural; conteniendo observaciones sobre el clima, suelo, poblaci6n, ag-
ricultura, comercio, character y costumbres de sus habitantes (2nd Spanish ed.,
translated from the original French ed. of 1810). Inst. Lit. Puertorriquefia, Univ.
Puerto Rico. xxiii + 178 p.
MATHENY, E. L. 1981. Contrasting feeding habits of pest mole cricket species. J.
Econ. Ent. 77: 444-45.
NICKLE, D. A., AND J. L. CASTNER. 1984. Introduced species of mole crickets in
the United States, Puerto Rico, and the Virgin Islands (Orthoptera: Gryllotal-
pidae). Ann Ent. Soc. America 77: 450-65.
WOLCOTT, G. N. 1936. "Insectae Borinquenses" a revised annotated check-list of the
insects of Puerto Rico. J. Agric. Univ. Puerto Rico 20: 1-631.


Biology Department, Central Michigan Univ.,
Mt. Pleasant, MI 48859


Melanoplus tequestae Hubbell is a small, brachypterous, forb-feeding grasshopper
inhabiting sand scrub and xeric oak forests of central Florida. Pair formation between
silent males and females occurred when the male slowly approached the female or
waited until her general activities brought her closer to him. Females raised their
ventrally colored hind femora and extended their tibia to signal non-receptiveness.
There was no apparent courtship before mounting, but after mounting males shook
their hind femora before copulating. Mounting and copulation could last from a few
hours to 1 1/2 days. Male aggressive signaling was by bursts of femur shaking.


Melanoplus tequestae Hubbell es un pequeno braquiptero saltamonte habitando
malezas en terrenos arenosos y bosques de cedros adaptados a ambientes secos en el
centro de la Florida. El apareamiento entire machos silenciosos y hembras ocurri6
cuando el macho se acerc6 lentamente a la hembra, o esper6 a que las actividades
generals de ella la trajera mas cerca de 61. Las hembras levantaban su colorida f6mur
trasero ventralmente, y extendian su tibia para senalar que no eran receptivas. Aparen-
temente no hubo cortejo antes de montar, pero duspues de montar, los machos sacuden
su f6mur posterior antes de copular. Montaje y coplaci6n pudiera durar de unas pocas
horas a 112 dia. La serial de agresividad del macho son sacudiones del femur.

Florida Entomologist 70(4)

Melanoplus tequestae Hubbell is a small, brachypterous, melanopline acridid in the
Puer Group that occurs locally in the central Highlands of peninsular Florida throughout
most of the year (Hubbell 1932). It has been collected in sand scrub habitat, open oak
forests on sandy slopes and in a few other mixed herbaceous/woody, sandy habitats.
Little is published on the biology of the numerous Melanoplus species found along the
coastal plain of Florida, Alabama and Georgia, and there is no literature on the mating
behavior of M. tequestae. It does not stridulate or use wing movements for pair forma-
tion or courtship. Femoral and body movements and possibly pheromones, are the most
likely modes of communication. This study provides information on the mating behavior
of M. tequestae.


Adults and nymphs of M. tequestae were collected at Archbold Biological Station in
south-central Florida between 18 February and 13 March, 1986 and put in a 21 x 25 x
25 cm terrarium for observation and filming. The terrarium had sand and vegetation
that resembled the habitat and an overhead 100 watt incandescent bulb provided a
temperature ranging from 26 to 29C. The photophase ranged from 14 to 16 hr. Individu-
als were maintained in a second terrarium under the same conditions. Fresh lettuce
was available in both cages. A Bolex super-8 movie camera was used to film behavior.
Behavioral descriptions are a composite of observations derived from 5 adult males, 6
adult females and 2 last-instar females.


Specimens of M. tequestae were scarce and only 15 were collected during a 3-week
period. These were found along dry, sandy firetrail edges that surrounded southern
ridge sandhills, sand pine scrub and scrubby flatwoods (Abrahamson et al. 1984). Gras-
ses and forbs were sparse and low to the ground because new growth was just begin-
Feeding activity in terraria on plants from the grasshoppers' habitats suggested that
this species was primarily a forb feeder. Individuals briefly nibbled Rhynchelytrum
repens (Willd.) C. E. Hubbard (natalgrass), Eremochloa ophiuroides Hack (centipede-
grass), Andropogon brachystachyus Chapm. (shortspike bluestem), and a Cyperus sp.
(sedge). They fed intermittently on new leaves of forbs such as Froelichia floridana
(Nutt.) Moq. (Amarantaceae), and the composits Pityopsis graminifolia (Michx. Nutt.
and Heterotheca subaxillaris (Lam.) Britt and Rusby (camphorweed).

Pair Formation
Males were 11.5 to 12.5 mm long and females were 18.0 to 19.5 mm. Forty six pair
formations were observed, resulting in 22 attempts of males to mount females, 18
occasions when females assumed a defensive posture and 6 occasions where no further
mating behavior occurred. Males were very alert and appeared to have good vision.
Otte (1977) noted that lack of sound production would likely result in a more acute visual
mode or an increase in importance of a lesser sensory mode such as olfaction. Males
approached females by slowly walking or by leaping from as much as 22 cm away to
land 3 to 4 cm from females. After approaching the female, the male's body became
more rigid as he faced her from the side at angles ranging from about 400 to 900 to the
long axis of her body. His antennae were parallel and pointed toward her. Males
mounted females by leaping on them within 3 to 90 sec (mean = 65 sec) after facing
Alternatively, a female might randomly move toward a motionless male as she slowly
searched for food, seemingly unaware of his presence. Females moved somewhat more

December, 1987

Bland: Acridid Mating Behavior 485

continuously than males based on the average number of times 11 individuals were in
motion. One hundred 10-sec observations were made every 15 min over a 3-day period.
Males were in motion 36 times and females 46 times. Uvarov (1977) noted that females
of flightless grasshoppers must be sufficiently restless to be noticed by males.
Two females were highly unreceptive and assumed obvious hind femur-raising defen-
sive postures, a behavior that is used by acridids to repel other grasshoppers in most
cases (Steinberg and Willey 1983). As a male slowly approached a female from a
laterocaudal position, she responded when he was 4 to 5 cm away by stiffening her
abdomen and holding it parallel to the ground. She took 4 to 5 sec to turn and face away
from the male, and then raised her hind legs so that the yellow ventral portion of the
femora was vertical and the tibia extended. The bright yellow color was quite noticeable
on an otherwise gray-brown female. After 1 to 3 min she turned so she was at a 90
angle to the male then lowered her legs. On another occasion, a different female voided
a fecal pellet and began twisting her abdomen while opening and closing her ovipositor
valves. The male's antenna, on the side closest to the fecal material, pointed at the
pellet and then at the female abdominal movement. The female turned away from the
male and again raised her legs for a few min. She oviposited 1 day later and was mating
the following day.
Males responded to femur-raising by females in one or more ways. The types of male
responses and the number of males responding during 18 femur-raising occasions were
as follows: (1) 2 to 3 sec of head-thorax weaving repeated several times (11); (2) a few
slight up-down body movements (4); (3) periodic opening and closing of terminalia (5);
(4) several 0.5 to 1 sec bursts of silent, short-stroked hind femora shaking (3); (5)
remaining nearly motionless for 1 min (1), 2 min (3), 3 min (8), 4 min (4), 5 min (1) or
9 min (1). A male never attempted to mount a female when her hind femora were raised.
Steinberg & Willey (1983) point out that this behavior prevents contact and mounting,
it may maintain a temporary pair bond and reveal identity and, in the case of a Trimero-
tropis species, it may be a female's test of an individual male's fitness by stimulating
repeated courtship. However, in the case of M. tequestae there was no apparent
courtship, and femur-raising prevented any attempt of males to mount females.
Unreceptive females raised their hind femora at approaching last instar nymphs but
the latter continued their activity. Hind femora of last-instar females were not yellow
and were not raised when approached by stalking males. Males rarely attempted to
mount nymphs perhaps because olfactory cues were lacking. Although acridid phero-
mones used for recognition in pair formation are unknown, Steinberg & Willey (1983)
suggest that the accurate recognition of females by males of a Trimerotropis species is
due to pheromones.

Courtship Before Mounting

This species, like Melanoplus species in general (Otte 1970), appeared to have no
courtship behavior after pair formation and before the male mounts the female. Al-
though 3 males were observed to produce a few bursts of femur shaking during a
female's femora-raising defensive posture, it could not be determined if this was an
attempt at courtship or a common disturbance (aggressive) response as described by
Otte (1970).

Mounting, Courtship, and Copulation

Twenty-two attempted mountings were observed. Generally females jumped vigor-
ously when mounted and sometimes landed upside down several times from the weight
of the male. If he was still mounted after 3 to 6 leaps, she usually offered little further
resistance and spread her femora at a slightly greater angle than normal. If the mount-

486 Florida Entomologist 70(4) December, 1987

ing was unsuccessful, a second attempt, always unsuccessful, might be made after
several minutes.
The following example was typical of pre-copulatory behavior. After the male
mounted, both male and female antennae vibrated for 0.5 to 0.8 sec repeated at 5 to 10
sec intervals. This antennal movement appeared to be a form of courtship. In cases
where copulation occurred relatively quickly, the female antennae simply pointed down
while his pointed up. After 3 min he slid backwards so his head was over her front legs
and his front and middle legs gripped her thorax laterally. After he probed with his
terminalia, it was evident that his abdomen was not long enough to reach the genitalia
of the female's rigid and elongated abdomen, so he slid backwards again to where his
head was over her middle legs. His hind femora were raised slightly so the tarsi could
push against the dorsolateral area of her abdomen. The male's femora were shaken
intermittently during this 7 min positioning but the female's did not move. Typically,
males shook their femora, in a burst of extremely short strokes, 6 to 10 times in less
than a second and at intervals of 45 to 90 sec. Otte (1970) classifies this communicative
behavior as courtship after mounting and Uvarov (1977) points out that tactile courtship
of some type is likely for all acridid species before a female will allow copulation.
The duration and repetition of mountings and copulations were quite variable. The
time between initial mounting and copulation averaged 52 min and ranged from 4 min
to nearly 4 hr (n = 17). Copulation time averaged 1.4 hr and ranged from 15 min to 3.6
hr. The following is an example of a very lengthy and repetitive mounting-copulation
(m-c) combination of a single isolated pair: Day 1, m-c 3 hr in the late morning; Day 2,
continuous m-c 0.5 hr after the lights came on in the morning through the day and night;
Day 3, male dismounted at 3 pm and remounted at 7 pm to remain on the female to
Day 4; Day 4, male dismounted about 10 am; Day 7, m-c in early afternoon through Day
8; Day 8, male dismounted in the morning.
Thornhill & Alcock (1983) discuss prolonged copulation in insects as a type of mate
guarding to allow time for sperm transfer and displacement. In some Orthoptera, sper-
matophore mobilization and transfer may require many hours. Contact guarding, by
remaining coupled or just mounted in position, may be a way to prevent the female
from mating again before oviposition and insure that the male's sperm fertilizes her
eggs. For example, McLain (1980) showed that male Nezara stinkbugs lengthened the
time spent in copula when the density of competitors was high. The density of M.
tequestae in the terrarium was higher than in the field and this situation may have
accounted for part of the lengthy copulation. Pickford & Gillott (1971) showed that
multiple spermatophores were passed to female M. sanguinipes (F.) during prolonged
copulation even though sufficient sperm was available without lengthy copulation.
Riegert (1965) and Pickford & Gillott (1976) hypothesized that extended time may be
needed for spermatophore nutrient availability, since prolonged coupling resulted in
increased fecundity, percent egg hatch and offspring survival in the species. M. teques-
tae populations are relatively low and quite local, especially when compared to M.
sanguinipes, and thus it seems less likely that fecundity, egg hatch or nymph survival
is enhanced by prolonged copulation.
Multiple matings by female insects are not uncommon and the potential advantages
are discussed by Sakaluk & Cade (1983) using crickets as their source of information.
They list benefits such as increased nutrition from additional spermatozoa, improved
access to resources, increased genetic variability and a greater likelihood of reproduc-
tion in case the first mating failed.

During copulation the male silently shook his femora when disturbed by the activity
of the female or a nearby male or female. The shaking consisted of 3 to 5 strokes in 0.5

Bland: Acridid Mating Behavior

to 0.7 sec. Bursts sometimes occurred every 1 to 12 sec for up to several minutes if a
nearby male continued to jump, moved closer or shook his own femora. On one occasion
when a nearby female jumped, the mounted male shook his femora in 14 bursts at 1 to
2 sec intervals. Non-mounted males shook their femora whether facing toward or away
from a mounted male, but when the latter stopped shaking the non-mounted male
stopped also.
In general, the mating behavior of M. tequestae was similar to that of other melano-
plines and less complex than some described by Otte (1970). There were no activities
atypical of the genus. Male vision was good, making them alert to the larger size and
slower movements of females. Males were sexually aggressive on a regular basis in the
confines of a terrarium.


The author thanks M. Deyrup, resident entomologist at Archbold Biological Station,
for making facilities available and S. Van der Kloet, visiting botanist, for plant identifi-
cation. The study was made possible by a research grant from the School of Graduate
Studies, Central Michigan University.


Vegetation of the Archbold Biological Station, Florida: an example of the south-
ern Lake Wales Ridge. Florida Sci. 47: 209-250.
HUBBELL, T. H. 1932. A revision of the Puer Group of the North American genus
Melanoplus, with remarks on the taxonomic value of the concealed male genitalia
in the Cyrtacanthacrinae (Orthoptera, Acrididae). Misc. Publs. Mus. Zool. Univ.
Michigan. 23: 1-64.
MCLAIN, D. K. 1980. Female choice and the adaptive significance of prolonged copu-
lation in Nezara viridula (Hemiptera: Pentatomidae). Psyche 87: 325-336.
OTTE, D. 1970. A comparative study of communicative behavior in grasshoppers.
Misc. Publs. Mus. Zoel. Univ. Michigan 141: 1-168.
OTTE, D. 1977. Communication in Orthoptera, pp. 334-361. In How Animals Com-
municate, ed. T. Sebok. Indiana Univ. Press, Bloomington.
PICKFORD, R. AND C. GILLOTT. 1971. Insemination in the migratory grasshopper,
Melanoplus sanguinipes (Fabr.). Canadian J. Zool. 49: 1583-1588.
PICKFORD, R. AND C. GILLOTT. 1976. Effect of varied copulatory periods of
Melanoplus sanguinipes (Orthoptera: Acrididae) females on egg hatchability and
hatchling sex ratios. Canadian Entomol. 108: 331-335.
RIEGERT, P. W. 1965. Effects of grouping, pairing and mating on the bionomics of
Melanoplus bilituratus (Walker) (Orthoptera: Acrididae. Canadian Entomol. 97:
SAKALUK, S. K. AND W. H. CADE. 1983. The adaptive significance of female multiple
matings in house and field crickets. In Orthopteran Mating Systems: Sexual
Competition in a Diverse Group of Insects, eds. D. Gwynne and G. Morris.
Westview Press, Boulder, CO.
STEINBERG, J. AND R. WILLEY. 1983. The mating system of Trimerotropis
maritima (Acrididae: Oedipodinae), pp. 285-304. In Orthopteran Mating Sys-
tems: Sexual Competition in a Diverse Group of Insects, eds. D. Gwynne and G.
Morris. Westview Press, Boulder, CO.
THORNHILL, R. AND J. ALCOCK. 1983. The Evolution of Insect Mating Systems.
Harvard Univ. Press, Cambridge, MA. 547 pp.
UVAROV, B. 1977. Grasshoppers and Locusts. A Handbook of General Acridology,
Vol. 2. Centre for Overseas Pest Research, London. 613 pp.

488 Florida Entomologist 70(4) December, 1987


Insect Attractants, Behavior, and Basic Biology Research Laboratory,
Agricultural Research Service, U.S. Department of Agriculture,
Gainesville, Florida 32604


Sex pheromone baited traps captured Angoumois grain moths (AGM), Sitotroga
cerealella (Olivier), Indianmeal moth (IMM), Plodia interpunctella (Hubner), almond
moth (AM), Ephestia cautella (Walker), and Mediterranean flour moth (MFM),
Angaesta kueh~iella (Zeller), near storage facilities and in field environments. The
AGM was particularly abundant around a wheat field, and numerous in a feedmill yard
and near peanut storage, and was the species captured most frequently. The IMM and
AM were caught most frequently near peanut storage facilities, although each species
was captured in every environment sampled. The AGM, IMM, and AM were most
common during the warm months and absent during February. By contrast, MFM was
absent during the summer months with peak population levels occurring during Janu-
ary, February, and March.


Trampas cebadas con feromonas sexuales atraparon a alevillas de grano Angoumois,
Sitotroga cerealella (Olivier), alevillas Indianmeal, Plodia interpunctella (Hubner),
alevillas del almendro, Ephestia cautella (Walker), y alevillas Mediterraneas de la
harina, Anagasta kuehniella (Zeller), cerca de lugares de almacenamiento y en medios
ambientales. Alevillas de grano Angoumois fueron particularmente abundantes al-
rededor de un campo de trigo, y numerosas en el area de un molino y cerca de un
almacenamiento de mani, y fue la especie que se catur6 mas frecuentemente. Las alevil-
las Indianmeal y del almendro se atraparon mas frecuentemente cerca de almacenes de
mani, aunque cada especie fue capturada en cada lugar que se muestre6. Las alevillas
de grano Angoumois, Indianmeal, y del almendro, eran mas comun durante los meses
calidos y ausentes durante Febrero. En contrast, alevillas Mediterraneas de la harina
estaban ausentes durante los meses de verano, el auge del nivel de poblaci6n ocurri6
durante Enero, Febrero y Marzo.

The Angoumois grain moth (AGM), Sitotroga cerealella (Olivier), (Gelechiidae) as
well as two phycitine moths, almond moth (AM), Cadra cautella (Walker), and In-
dianmeal moth (IMM), Plodia interpunctella (Hfibner), are important pests of stored
products in the southern United States. Two other potentially destructive phycitine
moths, the Mediterranean flour moth (MFM), Anagasta kuehniella (Zeller) and tobacco
moth, Ephestia elutella (Hibner) also are found in this region. All these species are
commonly encountered in stored-product marketing channels and many infestations
undoubtedly originate in manufacturing or transportation facilities. Many other infesta-
tions are the result of inadequate sanitation; e.g., putting uninfested grain into bins
that contain a residual insect infestation. A third avenue for infestation is the influx of
insects into storage from field environments; it has received little attention and has not

Vick et al.: Phycitid Moths in the Field 489

been recognized as a significant problem. Tactics to avoid infestation would be different
for each of these infestation modes.
Among moths, the AGM has a recognized potential to infest crops in the field before
harvest (Cotton 1963, Stockel 1971), although mechanical harvesting at high moisture
levels followed by mechanical drying was suspected to have lessened this avenue of
entry into storage. AGM populations at Beaumont, Texas, were measured with pher-
omone-baited traps and were heaviest near rice milling and storage areas. Also substan-
tial AGM populations were detected in rice fields and pastures, and small populations
were found in forests (Cogburn & Vick 1981).
Phycitine species do not attach their eggs to grains and are unable to feed on intact
grains. Although these factors would appear to make it unlikely that Phycitinae moths
could survive in field situations, Cogburn & Vick (1981), in fact, found small but measur-
able populations of AM in field situations. Since these populations constitute a pool of
insects that may be attracted to stored grains, we initiated this study to see if popula-
tions of these phycitine moths and AGM exist in field situations in North-Central Florida
and to estimate their magnitude.


Synthetic sex pheromone of the Angoumois grain moth, (Z,E)-7,11-hexadecadien-1-
ol acetate, was purchased from Farchan Division, Story Chemical Co., Willoughby,
Ohio, and purified to >98% by the method of Vick et al. (1979). The pheromone was
formulated into plastic laminated strips (Hercon). A square of the material (0.5 by 0.5
cm) was used to bait each trap. A 1-Rl amount of Phycitinae sex pheromone, (Z,E)-7,11-
hexadecadien-1-ol acetate, was placed in a 250-jl capacity cap (size #3, BEEM
polyethylene embedding capsule C) (Ladd Research Industries, Inc., Burlington, Ver-
mont). Pheromone release rates were ca. 77 and 320 ng/h for the Angoumis grain moth
and Phycitinae pheromones, respectively (Vick et al. 1979). Pairs of traps, one with
each bait, were placed at each trapping location. Pheromone baits were replaced every
2 weeks, and trap liners were changed as needed. Traps were checked once a week
from July 1980 through July 1981. The moths were identified by dissection of the male
Ten trapping locations were selected roughly on a ca. 35 mile line between High
Springs (Alachua County), and Williston (Levy County), Florida. This region of Florida
has sandy-loam soil, diversified agriculture of row crops, truck crops and pasture land
that is interspersed by stands of mixed hardwood and planted and natural pines. Climate
is typical of much of the coastal southeastern United States with humid, hot summers
and mild winters (ca. 25 nights below 0C each winter). Traps were set in the following
habitats: pastures (three locations), cornfields (two locations), feedmill yards (two loca-
tions), peanut warehouse yard (two locations), and a wheat field (one location).


More AGM were caught than any other species (Fig. 1). They were particularly
abundant near the wheat field where more than 3X as many were captured as at any
other location. Substantial numbers were also captured in the vicinity of the peanut
warehouses and feed mills. Fewer AGM were caught at the other two locations.
Three species of Phycitinae moths were caught: IMM, AM, and MFM. The relative
efficiency of the pheromone used for the Phycitinae species is unknown for field condi-
tions. Therefore, conclusions were not drawn on relative abundance of the three species
based on numbers of insects caught in traps. Every location yielded specimens of each

Florida Entomologist 70(4)

December, 1987


I ]



1 2 3 4 5

Fig. 1. Average number of Angoumois grain moth (AGM), almond moth (AM), In-
dianmeal moth (IMM), and Mediterranean flour moth (MFM) males caught per week
over a year period in the following habitats: 1-peanut warehouse yard (two locations),
2-pasture (three locations), 3-corn field (two locations), 4-feedmill yard (two loca-
tions), and 5-wheat field (one location).

species. Predictably, AM and IMM captures were greatest in the vicinity of peanut
warehouses since both species are major pests of this commodity. Traps for these two
species at other locations captured <0.5 insects/trap/week. The MFM by contrast was
no more numerous around the peanut warehouses than at other locations.
The AGM was captured in every month of the year except January, February and
March (Fig. 2). Likewise IMM and AM populations were at a low level during these
months; neither of these species was captured during the month of February. By con-
trast, MFM were only captured between November and May with the peak population
levels occurring during January, February and March.


The presence of AGM in substantial numbers in every habitat that was surveyed
mirrors the results reported by Cogburn & Vick (1981). They reported that MFM and
IMM were rare or absent from trap catches although both species were found in all
habitats in this study. The populations of AM were larger in this study than in the
Beaumont, Texas study.
The capture of the MFM in all trap locations was unexpected, since it is thought to
be rare in Florida. There are only scattered reports of MFM in interdictions of infested
commodities being transported into the state (Florida Division of Plant Industry re-
cords). This species was not found in trapping studies inside a peanut warehouse, a
military food distribution warehouse, or a civilian food distribution warehouse (Vick et

Vick et al.: Phycitid Moths in the Field


o L b

-1 -m
ocl o 0


log (x+1

N -
uic U 6

S- r
riuo 0






z .












Fig. 2. Monthly totals of Angoumois grain moth (AGM), almond moth (AM), In-
dianmeal moth (IMM), and Mediterranean flour moth (MFM) males caught over a year
period in the following habitats: 1-peanut warehouse yard (two locations), 2-pasture
(three locations), 3-corn field (two locations), 4-feedmill yard (two locations), and
5-wheat field (one location).

S l l l l l I I l I ll i I I I


-m m

m m am mg

m m mmm mi

m m m

R mm

mm m mmg
mg m agg

mm m mm mg m

mm mM

mmm mmA
Sm m
m m m


Florida Entomologist 70(4)

al. 1981, 1986). The presence of MFM in only winter trap catches suggests that it may
not be well adapted to the hot, humid summers of the southeastern United States.
However, its seemingly ubiquitous distribution in Alachua and Levy Counties and its
documented preference for milled products (Cotton 1963) suggest the likelihood that a
search of flour and feed mills might yield this insect.
Before the advent of grain storage by neolithic women, many species of stored grain
insects undoubtedly were field insects attacking kernels of grain as they ripened in the
field (Cotton 1963). Other species may have been able to survive on grain found in bird
nests or rodent dens. The advent of human agriculture with its necessary grain storage
provided an ideal habitat for these insects. Many stored-product insects, including the
four species studied in this paper, are cosmopolitan, having been spread in transported
agricultural commodities. Although none of these species are thought to be native to
the United States, it would appear from the results of this study and the study by
Cogburn & Vick (1981) that at least in mild climatic areas, four of these species have
found niches where they can survive apart from the typical storage situation where they
do so well.


COGBURN, R. R., AND K. W. VICK. 1981. Distribution of Angoumois grain moth,
almond moth, and Indianmeal moth in rice fields and rice storage in Texas as
indicated by pheromone-baited adhesive traps. Environ. Entomol. 10: 1003-1007.
COTTON, R. T. 1963. Pests of Stored Grain and Grain Products. Burgess Publishing
Co., Minneapolis, 318 pp.
STOCKEL, J. 1971. Utilisation du piegeage sexuel pour l'etude de replacement de
l'alucite Sitotroga cerealella (Lepidoptera, Gelechiidae) vas les cultures de mais.
Entomol. Exp. & Appl. 14: 39-56.
VICK, K. W., J. KVENBERG, J. A. COFFELT, AND C. STEWARD. 1979. Investigation
of sex pheromone traps for simultaneous detection of Indianmeal moths and An-
goumois grain moths. J. Econ. Entomol. 72: 245-249.
Recent developments in the use of pheromones to monitor Plodia interpunctella
and Ephestia cautella, pp. pp. 19-28. In: Everett R. Mitchell, Ed. Management
of Insect Pests with Semiochemicals: Concepts and Practice. Plenum Press, New
VICK, K. W., P. G. KOEHLER, AND J. J. NEAL. 1986. Incidence of stored-product
Phycitinae moths in food distribution warehouses as determined by sex phero-
mone-baited traps. J. Econ. Entomol. 79: 936-939.


December, 1987

O'Brien: New South American Lophopidae


3009 Brookmont Drive
Tallahassee, FL 32312
Research Associate, Florida State Collection of Arthropods
Division of Plant Industry
Florida Department of Agriculture and Consumer Services
Gainesville, FL 32602


A new species, Carrionia panamensis, is described. The genus Ucayalia Fennah is
synonymized with its senior synonym Carrionia Muir, and Carrionia nigrovittata (Fen-
nah) becomes a new combination. Keys to the 3 genera and 7 species and illustrations
of the frons, vertex, and forewing of each species are provided.


Se describe una nueva especia, Carrionia panamensis. Se sinonimiza el g6nero
Ucayalia Fennah con Carrionia Muir, lo que produce una combinaci6n nueva, Carrionia
nigrovittata (Fennah). Se proven claves para los 3 g6neros y las 7 species, y se
incluyen ilustraciones de la frente, vortex, y ala anterior de cada especie.

The morphological characters of a new species of Lophopidae from Panama are
intermediate between the characters used for the genera Carrionia Muir and Ucayalia
Fennah, so that the latter name is synonymized. This produces a new combination,
Carrionia nigrovittata (Fennah). The new species extends the range of the family into
Panama from Brazil, Peru, Ecuador, and the Guianas.
This paper contains a checklist of the New World species, keys to the genera and
species, brief generic descriptions, a description of the new species, and illustrations of
the frons, vertex, and forewing shape of all of the species. A key to the families of
Fulgoroidea may be found in Fennah (1950) or an illustrated key in O'Brien & Wilson
(1985). Lophopidae fall into the group of Fulgoroidea with the 2nd posterior tarsomere
devoid of spines. In the New World, Ricaniidae, Lophopidae, and Eurybrachidae share
this character, but the one species of Eurybrachidae reported from the New World is
thought to be geographically mislabelled.
It is interesting to note that Carrionia is the only new world genus with the female
anal flap enlarged to produce and hold strands of wax, which Asche, Hoch, and 1 thought
might be a unique synapomorphy which would define the family Lophopidae when we
discovered it in 1985 (unpublished). The 10th segments of female Hesticus and Sil-
vanana are normal fulgoroid anal segments; that is, small and entire, no larger than
the 10th segment of the male of the same species. The specimens I have on hand do not
have wax on this segment, nor on the 7th, 8th and 9th, the usual position for wax plates
in Fulgoroidea.


Florida Entomologist 70(4)

December, 1987

Checklist of New World Lophopidae

Taxon Distribution Type Repository
Carrionia flavicollis Muir Ecuador BMNH
nigrovittata (Fennah) Peru NMNH
panamensis n. sp. Panama LOB
Hesticus pictus Walker Brazil, Br. Guiana BMNH
rufimanus Walker Brazil BMNH
sanguinifrons Muir Ecuador, Peru BPBM
Silvanana omani Metcalf Brazil NCS

Key to the genera of New World Lophopidae

1. Forewings membranous and transparent ............................ Hesticus Muir
1'. Forewings colored and opaque ...................................................... 2
2(1'). Pro and meso femora and tibiae foliately expanded; lateral ocelli absent;
large species 12 mm long or longer; vertex narrow and concave Carrionia Muir
2'. Femora and tibiae not expanded; lateral ocelli present; small species under
10 mm in length; vertex broad and flat ........................... Silvanana Metcalf

Silvanana omani Metcalf and Ucayalia nigrovittata Fennah were illustrated
thoroughly when they were published. Illustrations of Hesticus species from the Ama-
zon are being published (O'Brien, Penny & Arias, in press) and the other two species
of Carrionia are illustrated here.

Carrionia Muir 1931 (fig. 1-3, 6-9, 15-17)
[Type species flavicollis, by monotypy]

= Ucayalia Fennah 1944, NEW SYNONYMY. [Carrionia nigrovittata (Fennah),
Large lophopids, 12-16 mm in length, with yellow pronotum, partially dark head,
and dark wings with apical light bands. Vertex longer than wide, varying from 1.25:1
to 1.9:1, concave; frons longer than broad or as long as broad, lateral angles broadly
produced (fig. 6-8); genae lacking ocelli, with a transverse ridge at level of frontal
angles; clypeus laterally carinate at least at base; pro and meso femora and tibiae
foliately flattened and expanded.

Key to the species of Carrionia

1. Frons mostly black, subequal in length at midline and width at widest part
...................................................................................... fl avicollis M uir
1'. Frons yellow or green with black midline, frons at least 1.3 x longer than
w ide ....................................................... .... ..... .... .................. 2
2(1'). Apex of each forewing with 3 small translucent areas (fig. 16) ...............
.... .......................... ........... .......... ... panamensis n. sp.
2'. Apex of each forewing with transverse translucent bands (fig. 15) ..........
................................................................ nigrovittata (Fennah)


O'Brien: New South American Lophopidae



SI 13 14

Fig. 1-4: Vertex and pronotum of 1. Carrionia nigrovittata Fennah, 2. C. panamen-
sis, n. sp., 3. C. flavicollis Muir, 4. Silvanana omani Metcalf. Fig. 5-8: Frons andd
clypeus (stippling indicates black coloration) of 5. S. omani, 6. C. nigrovittata, 7. C.
panamensis, 8. C. flavicollis. Fig. 9-11: Vertex and pronotum (stippling indicates black
coloration, crosshatching indicates red coloration) of 9. Hesticus rufimanus Walker, 10.
H. pictus Walker, 11. H. sanguinifrons Muir. Fig. 12-14: Frons and clypeus of 12. H.
rufimanus 13. H. pictus, 14. H. sanguinifrons.

Carrionia panamensis, new species (fig. 2, 7, 16)

SALIENT FEATURES: Females 14-15.5 mm in length. Dorsal surface of pronotum
yellow. Head pale green with a broad band overlying median carina of frons and clypeus,
posterior 3/4 of vertex, sides of head above eyes, a triangular area on gena below eyes,
apex of clypeus, lateral fields of pronotum, and ventral half of tegulae piceous. Legs
piceous except for diagonal pale band across middle third of fore and meso tibiae.
Tegmina dull rusty brown, appearing black where adpressed to darker hind wings, with
three membranous areas at apex of wing outlined by shiny darker bands. Abdomen red.
TYPES: Holotype female: PANAMA, Canal Zone, Barro Colorado Island, III-2-1967,
Roger D. Akre (LOB). Paratype females, 2: Panama, Coco Solo, 6-1-[19]36 (CAS);
Panama Pr., Cerro Jefe, 700 m. 9 12' N, 790 21'W, 20-V-[19172, Stockwell (HW).

Florida Entomologist 70(4)


:i.:i..y i :il. ....
:... : :. ::. : il

is ,,:::: .-,''. ,. ."
,-- :

V / y




Fig. 15-21: Tegmina (stippling indicates brown or black coloration, crosshatching red
coloration) of 15. C. nigrovittata, 16. C. panamensis, 17. C. flavicollis (tips of tegmina
broken off), 18. S. omani (color pattern not shown), 19. H. rufimana, 20. H. pictus,
21. H. sanguinifrons.

NOTES: The specimen from Cerro Jefe is the largest of the three and slightly differ-
ent from the other two in the shape and coloring of the fore and mid tibiae. The tibiae
are less expanded than in the other two species, being approximately 3.6 x as long as

December, 1987




O'Brien: New South American Lophopidae 497

broad, compared with 3.4 for flavicollis, 3.2 for the other specimens of panamensis,
and 2.3 for nigrovittata. The white band is reduced to a small spot, but the tibiae are
also slightly indented at this point. In every other respect the specimen seems to be
panamensis, and I am treating it as a paratype until males are collected for genitalic
This species is very similar to nigrovittata in color pattern except for the tegmina,
the pale bands on the femora mentioned above, and the increased area of dark markings
on the frons, clypeus, vertex, and gena of panamensis. The tegminal pattern is similar
in the number of dark bands, but in panamensis they are compressed into one corner
while in nigrovittata they are spread across the wing. The apices of the forewings are
missing inflavicollis, but part of the first band is visible (fig. 17).

Hesticus Walker 1862 (fig. 9-14, 19-21)
[Type species pictus, by nonotypy]

Moderate sized lophopids, 9-12 mm in length, with yellowish orange bodies, some-
times marked with red, green, brown or black; wings translucent, sometimes marked
with brown or red. Vertex 1.5 times as long as wide to subequal, chevron shaped,
sometimes lateral carinae raised, vertex concave or flat; frons longer than broad or as
long as broad, sides parallel; ocelli present on genae; frontoclypeal suture continued on
genae but no transverse ridge as in Carrionia, clypeus laterally carinate at base; front
femora and tibiae foliately flattened and expanded, mid and hind femora and tibiae
normal, subequal in size.

Key to the species of Hesticus

1. Frons and clypeus with vertical red stripe; tegmina with veins of stigma
red .................................................... .................... sanguinifrons M uir
1'. Frons or clypeus lacking vertical red stripe, but with 1 or more horizontal
black bands; tegmina without red veins .............................................. .. 2
2(1'). Face with only the clypeus having a horizontal black band; tegmina with
brown marking limited to spots at stigma and apex ......... rufimanus Walker
2'. Face with both frons and clypeus having horizontal black bands; tegmina
with stigmal spot, clavus, and commisural margin to apex brown pictus Walker

Silvanana Metcalf 1947 (fig. 4, 5, 18)
[Type species omani Metcalf]

Small lophopids, about 8 mm in length, with light brown body and wings marked
with darker brown. Vertex longer than wide, parallel sided, flat; frons longer than
broad, slightly concave, lateral angles produced; genae with ocelli on transverse ridge
at level of frontal angles; clypeus with lateral carinae at base, flattened and sunken
between legs; none of legs foliately expanded.


I am indebted to G. Allan Samuelson, who examined the type of Hesticus sanguinif-
rons at the Bernice P. Bishop Museum (BPBM), and to Carol Parron, North Carolina
State University (NCS), and James P. Kramer, National Museum of Natural History
(NMNH), for loan of types, and to Lawrence A. Mound, the Keeper of Entomology of
the British Museum (BMNH) for permission to study specimens there. I especially

498 Florida Entomologist 70(4) December, 1987

thank Henk Wolda, Smithsonian Tropical Research Institute (HW), who loaned the
specimen by which generic synonymy could be established, and Norman D. Penny of
California Academy of Sciences (CAS) who loaned the other specimen of the new
species. I thank the Florida State Collection of Arthopods for a grant to cover publica-
tion charges.
Contribution No. 664, Bureau of Entomology, Division of Plant Industry, Florida
Department of Agriculture and Consumer Services, Gainesville, FL 32602.


FENNAH, R. H. 1944. New neotropical Fulgoroidea. American Mus. Novitates 1265:
.1950. Fulgoroidea of Fiji. B. P. Bishop Museum Bull. 202: 3-122.
METCALF, Z. P. 1947. A new genus of Lophopidae from Brazil. Proc. Entomol. Soc.
Washington 49: 238-240.
--. 1955. General Catalogue of the Homoptera. Fasc. IV. Fulgoroidea, Part 17,
Lophopidae. N. Carolina State Univ., Raleigh. 84 pp.
MUIR, F. 1931. New and little known Fulgoroidea from South America. Proc.
Hawaiian Entomol. Soc. 7: 469-480.
O'BRIEN, L. B., N. D. PENNY, AND J. R. ARIAS. In press. Lophopidae of the
Amazon Basin. Acta Amazonica.
O'BRIEN, L. B. AND S. W. WILSON. 1985. Planthopper Systematics and External
Morphology. pp. 61-102. In Nault, L. R. and J. G. Rodriguez, Eds. The Leafhop-
pers and Planthoppers. Wiley & Sons, New York.
WALKER, F. 1858. List of the specimens of Homopterous insects in the collection of
the British Museum. Supplement. 1858: 1-307.
.1862. Characters of undescribed species of Homoptera in the collection of F.
P. Pasco, F. L. S. J. Entomol. 1: 303-319.


North Carolina State Museum of Natural History
P.O. Box 27647, Raleigh, North Carolina 27611
Research Associate, Florida State Collection of Arthropods
Division of Plant Industry
Florida Department of Agriculture and Consumer Services
Gainsville, FL 32602


The scolopendromorph centipede fauna of North Carolina consists of eight species.
Scolopocryptops sexspinosus (Say), S. nigridius McNeill, Theatops posticus (Say), and
Cryptops hyalinus Say occur in all three physiographic provinces. Theatops

Shelley: North Carolina Centipedes 499

spinicaudus (Wood) is prevalent in the Blue Ridge Province and western Piedmont
Plateau, and Hemiscolopendra punctiventris (Newport) is widespread east of the moun-
tains. Scolopendra viridis Say, ranging from the sandhills to the southeastern coastal
islands, and Scolopocryptops peregrinator (Crabill), in Ashe County, are newly recorded
from the state. The latter, ranging from southern Pennsylvania to northwestern North
Carolina and westward to eastern Kentucky, is elevated from subspecific status under
S. gracilis Wood.


La fauna de cienpies escolopendromorfos de Carolina del Norte, consiste de ocho
species. Scolopocryptops sexsopinosus (Say), S. nigridius McNeill, Theatops posticus
(Say), y Cryptops hyalinus Say ocurren en las tres provincias fisiograficas. Theatops
spinicausus (Wood) es prevalente en la Provincia Blue Ridge y en el oeste de Piedmont
Plateau, y Hemiscolopendra punctiventris (Newport) se extiende al este de las mon-
tafas. Se report por primera vez de este estado a Scolopendra viridis Say, que habitat
desde las lomas arenosas hasta las islas al sudeste de la costa, y Scolopocryptops pereg-
rinator (Crabill) en el Condado de Ashe. El filtimo, se extiende desde el sur de Pennsyl-
vania al noroeste de Carolina del Norte, y hacia el oeste hasta el este de Kentucky, y
se Ilevo a un estado sub-especifico bajo S. gracilus Wood.

The centipede fauna of the southeastern United States is poorly known. Records
can be gleaned from general works on North America by Say (1821) and Wood (1862,
1865), and reviews of Nearctic Lithobiomorpha (Chamberlin 1912a, 1913, 1914, 1917,
1922, 1925a,b). The southeastern Lithobiomorpha and Geophilomorpha were sum-
marized by Chamberlin (1911, 1912b), and Crabill (1960) provided a key to, and commen-
tary on, the Scolopendromorpha north of Mexico. Many nominal southeastern species
have been described in short miscellaneous papers, too numerous to cite here, along
with forms from other parts of the country. Specific regional publications in the past
half-century include incomplete state listings for North Carolina (Brimley 1938, Causey
1940, Wray 1950, 1967), South Carolina (Crabill 1950), Georgia (Chamberlin 1944, 1945),
and Florida (Chamberlin 1958); descriptions of new lithobiomorphs from North Carolina
(Chamberlin 1940a, Causey 1942) and Florida (Chamberlin 1940b); descriptions of new
geophilomorphs from Alabama (Crabill 1953) and Tennessee (Crabill 1958); a report on
the scolopendromorphs in the Coastal Zone of South Carolina (Shelley 1978); and a key
to the five Floridian species of the scolopendromorph genus Cryptops, with a description
of a congener from the tip of the peninsula (Crabill 1969). This key to Cryptops is the
only published identification guide to southeastern chilopods, but its utility is limited
because it refers only to one genus and one state. The most proximal modern key to all
American chilopod taxa is for the north-central states (Summers 1979).
I have been sampling North Carolina centipedes and examining preserved specimens
for 15 years to prepare a checklist of the state's fauna. Scutigera coleoptrata (L.), the
introduced European scutigeromorph that occurs statewide, is the sole representative
of this order, and eight scolopendromorphs are distributed among the three physiog-
raphic provinces. The Lithobiomorpha and Geophilomorpha are more speciose, but their
compositions are uncertain because of pervasive taxonomic and nomenclatorial prob-
lems, which are discussed by Lewis (1981) in his comprehensive text on chilopod biology.
Many years will pass before these difficulties are rectified and a checklist is feasible. A
report on North Carolina scolopendromorphs is possible, however, since all plausible
indigenous species have been recorded.

Florida Entomologist 70(4)

With 21 or 23 pairs of legs the scolopendromorphs are among the better known land
invertebrates because of their generally large size and their ability to deliver a painful
bite if handled.' They are prominent in woodland habitats because the striking yellow,
orange, and blue-green pigmentations contrast markedly with the darker colored sub-
strates. Both families, Scolopendridae and Cryptopidae, occur in North Carolina. The
two scolopendrids are larger (average lengths of specimens in this study are 47.4 mm
for Hemiscolopendra punctiventris (Newport) and 54.9 mm for Scolopendra viridis
Say) and occur only in association with decaying pine logs and stumps, especially be-
neath bark. Occasional specimens of Scolopocryptops sexspinosus (Say)2 also are more
than 50 mm long, but the cryptopids in general are smaller and are encountered in open
litter, under rocks, and under pine and deciduous logs.
The earliest reports of North Carolina Scolopendromorpha are by Wood (1862), who
cited Opisthemega postica (= Theatops posticus Say)) from Goldsboro and S. sex-
spinosus from Salem. Bollman (1888, 1893) recorded Cryptops hyalinus Say from un-
specified localities, Scolopocryptops nigridius McNeill and S. sexspinosus from Chapel
Hill, and Scolopendra woodi Meinert (= Hemiscolopendra punctiventris) from
Beaufort. Brolemann (1896) listed Theatops spinicaudus (Wood) and all these species
except C. hyalinus from unspecified sites in the state. Brimley (1938) and Causey (1940)
gave specific localities for all species except T. spinicaudus, which were repeated by
Wray (1950, 1967). However, the citations of Scolopendra viridis from Duke Forest,
Durham County, must be discounted as based on a misidentification of Hemiscolopendra
punctiventris, which is common in this part of the state. The specimen is no longer
available, and the closest authentic locality for S. viridis is near Fayetteville, Cumber-
land County, some 96 km south of Durham. Thus, the present North Carolina records
are the first verifiable ones for S. viridis, and I also add Scolopocryptops peregrinator
(Crabill), elevated from subspecific status under S. gracilis Wood.
The color of young individuals may be pale but the external anatomical features
mentioned in the key are applicable to all stages, because scolopendromorphs are
epimorphic and hatch with the full complement of legs and pedal segments. Except for
Scolopocryptops peregrinator, only prior North Carolina references are cited in the
ensuing species accounts, and distributions are solely within the state. The diagnoses
are pertinent only among these eight species and will not distinguish them from congen-
ers in other parts of America. North Carolina locality data are presented for
Scolopendra viridis, but generalized range statements are provided for the widely dis-
tributed species, since a detailed listing is prohibitively long. Scolopocryptops pereg-
rinator is known only from the original description, based on the holotype and paratype
from localities in Virginia and Maryland, respectively (Crabill 1952), and from Pine
Ridge, Wolfe County, Kentucky (Branson and Batch 1967). I have access to 24 individu-
als in 13 samples from different sites and report variation, provide detailed locality data,

'Because it is small and its bite cannot pierce skin, Cryptops hyalinus can be col-
lected by hand. The other species should be grasped by forceps behind the head, so that
the animal cannot flex the anterior end and strike the collector.
zConsiderable confusion has existed about the gender of the -ops genus-group suffix,
and as noted by Crabill (1955) both masculine and feminine species-group names have
been published arbitrarily for many scolopendromorph species. Some authors published
Scolopocryptops sexspinosus, while others have reported S. sexspinosa. Citing the
recommendation of the 1953 Copenhagen Decisions on Zoological Nomenclature [p. 51,
paragraph 84(7) (b) (iii)] (Heming 1957), Crabill (1955) attempted to stabilize treatments
by using only the feminine termination. However, this recommendation was recently
superseded by article 30 (a) (ii) of the 1984 edition of the International Code of Zoolog-
ical Nomenclature, which declares that all genus-group names with the -ops ending are
to be considered masculine regardless of derivation or treatment by the author.

December, 1987

Shelley: North Carolina Centipedes

and assess its taxonomic status by comparison with specimens of Scolopocryptops
gracilis from California. In the distribution maps, figures 11-16, open symbols denote
uncorroborated literature records considered reliable, and acronyms for physiographic
provinces are CP, Coastal Plain; PP, Piedmont Plateau; BR, Blue Ridge; RV, Ridge
and Valley; and AP, Appalachian Plateau. Acronyms of sources of preserved study
material are AAW, private collection of Andrew A. Weaver, Wooster, Ohio; ANSP,
Academy of Natural Sciences, Philadelphia; and NMNH, National Museum of Natural
History, Smithsonian Institution, Washington, DC. The invertebrate catalog number
is provided for specimens housed at the North Carolina State Museum of Natural His-
tory, Raleigh.

Key to Families, Subfamilies, Genera, and Species

1. With four ocelli on each side of cephalic plate (Fig. 1) ............. Scolopendridae 2
Ocelli absent (Figs. 5-6, 8) ................................................... Cryptopidae 3
2. Proximotarsi of legs 1-20 with prominent ventrodistal spur (Fig. 2); color
green, with or without lighter lateral stripes; Moore and Richmond to New
Hanover and Brunswick counties ............................ Scolopendra viridis Say
Without this character; color uniformly bluish gray; Piedmont Plateau and
Coastal Plain ................................ Hemiscolopendra punctiventris (Newport)
3. Ultimate legs greatly enlarged and thickened, heavily sclerotized (Figs.
3-4 ...................................................................................... Theatopinae 4
Ultimate legs only slightly larger and more sclerotized than preceding pair ..... 5
4. Prefemora of ultimate legs with dorsal, pigmented, distomedial spine (Fig. 4);
Blue Ridge Province to central Piedmont Plateau Theatops spinicaudus (Wood)
Prefemoral spine absent (Fig. 3); statewide, rare in Blue Ridge Province
..................................................... ..... Theatops posticus (Say)
5. 21 pairs of legs and pedal segments; statewide ................................
..................................... ............. Cryptopinae, Cryptops hyalinus Say
23 pairs of legs and pedal segments .............................. Scolopocryptopinae 6
6. Cephalic plate not margined; tergites 4-22 with complete paramedian sutures
(Figs. 8-9); color light yellowish; Ashe County .....................................
........................ ............................ Scolopocryptops peregrinator (Crabill)
Cephalic plate margined laterally; all tergites without complete paramedian
sutures (Figs. 5-7) .......................................... ........... ......... 7
7. Dorsal surface of second antennomere sparsely hirsute (Fig. 5); color usually
dark brownish-orange, with or without scattered irregular blue splotches;
statewide ................................................. Scolopocryptps nigridius McNeill
Dorsal surface of second antennomere densely hirsute (Fig. 6); color usually
uniformly bright orange to yellowish-orange, without darker patches; state-
wide ........................ .................. Scolopocryptops sexspinosus (Say)

Family Scolopendridae
Hemiscolopendra punctiventris (Newport 1844)
Figs. 1, 11

Scolopendra woodi: Brolemann, 1895:49-50. Brimley. 1938:501. Causey, 1940:65. Wray,
1950:155; 1967:155.
Scolopendra viridis: Brimley, 1938:501. Causey, 1940:65. Wray, 1950:155; 1967:155.
Scolopendra morsitans: Brimley, 1938:501. Wray, 1950:155; 1967:155.
Diagnosis. Color uniformly blue-gray; 21 pairs of legs and pedal segments; prox-
imotarsi without ventrodistal spurs; ocelli present.

Florida Entomologist 70(4)



Figs. 1-10. 1. right side of cephalic plate of Hemiscoloperndra punctiventris, dorsal
view. 2, midbody leg of Scolopendra viridis, ventral view. 3, ultimate legs and terminal
tergite of Theatops posticus, dorsal view. 4, ultimate legs and terminal tergite of T.
spinicaudus, dorsal view. 5, right side of cephalic plate and basal antennomeres of
Scolopocryptops nigridius, dorsal view; the arrow indicates the lateral margination. 6,
right side of cephalic plate and basal antennomeres of S. sexspinosus, dorsal view; the
arrow indicates the lateral margination. 7, tergites 1-5 of S. sexspinosus, dorsal view.
8, right side of cephalic plate and basal antennomeres of S. peregrinator, dorsal view.
9, tergites 1-5 ofS. peregrinator, dorsal view. 10, tergites 1-5 ofS. gracilis, dorsal view.

Ecology. Occurs only in association with decaying pine logs and stumps, usually
under bark.
Distribution. Piedmont Plateau and Coastal Plain.
Remarks. Hemiscolopendra punctiventris is sluggish and easy to collect. A specimen
will sometimes stay motionless for 10 to 15 seconds after exposure, leaving ample time
to reach for forceps and a vial.
Though occurring only in association with pines, H. punctiventris rarely is found
with small and medium-size branches. It is usually encountered on large logs whose

December, 1987


Shelley: North Carolina Centipedes


Fig. 11. Distributions of Hemiscolopendra punctiventris, trangles, and Scolopendra
viridis, dots, in North Carolina.

bark peels cleanly. Loblolly pine (Pinus taeda) seems to be the preferred species, and
the centipede's apparent absence from the mountains may relate to the diminished
occurrence of this tree. Hemiscolopendra punctiventris is abundant in the central Pied-
mont and Coastal Plain and likely to be found in any sub-climax woodland, being particu-
larly plentiful in areas ravaged by the southern pine beetle. Literature records consi-
dered valid are Beaufort, Carteret County (Bollman 1893), and Duke Forest, Durham
County (Causey 1940).

Scolopendra viridis Say 1821
Figs. 2, 11

Diagnosis. Color green, with or without pale lateral stripes; 21 pairs of legs and
pedal segments; proximotarsi of legs 1-20 with ventrodistal spurs: ocelli present.
Ecology. Occurs only in association with pine logs and stumps, usually under bark.
Distribution. Southeastern North Carolina, from the sandhills to the coastal islands
of New Hanover and Brunswick counties. As this is the first report of S. viridis from
North Carolina specifically and its range in the state is limited, detailed locality data
are provided. The sight record is by the author.
Richmond Co., 26 km W Hoffman (A1870) and 6.6 km NE Rockingham (2979).
Moore Co., 8.3 km W Aberdeen (2964). Cumberland Co., Fayetteville (A370) and 16.6
km NNW Fayetteville (sight record). Hoke Co., 6.4 km SW Ashley Heights (2945).
Scotland Co., Wagram (2976) and 15.2 km W Laurinburg (2981). Bladen Co., 8.0 km
NE Kelly (2918). New Hanover Co., 11.4 (328) and 13.6 km (2899) S Wilmington; 7.2
km NW Carolina Beach (A1449); 1.6 km SW Carolina Beach (1091); and Carolina Beach
St. Pk. (A3919-A3921). Brunswick Co., Boiling Springs Lakes (2255) and Southport
Remarks. On the average about 7 mm longer than H. punctiventris, S. viridis is
also much swifter and will usually move the instant it is uncovered. One must therefore
be ready with forceps in hand before stripping a prospective log. The two scolopendrids
occur in the same biotopes and appear to occupy closely similar niches; I have never
found them together.
In his key to American scolopendromorph genera, Crabill (1960) stated that the
proximotarsal spurs occurred on the first 15 to 20 legs in Scolopendra, and Summers
(1979) indicated that they were present on the anterior legs in general in S. viridis. In

Florida Entomologist 70(4)

the nearly 30 North Carolina individuals taken in this study, spurs are present on all
but the ultimate leg pair.
Crabill (1960) reported that S. viridis ranges northward along the Atlantic Coast
from Florida to southern Virginia, and it should therefore be expected throughout
our Coastal Plain. However, repeated searches in suitable hatitats north of Cumberland,
Bladen, and New Hanover counties have only yielded H. punctiventris. In 1983 I spent
a day near the Virginia border in the Dismal Swamp area of northeastern North
Carolina but did not find either scolopendrid. Thus, I cannot confirm Crabill's report,
but it is logical since the Dismal Swamp-southeastern Virginia area is a known biogeo-
graphical boundary and the northern distributional limit for the bald cypress (Taxodium
distichum) (Radford et al. 1968) and such southern vertebrates as the yellow-bellied
turtle (Trachemys s. scripta), brown water snake (Nerodia taxispilota), southern dusky
salamander (Desmognathus auriculatus), southern toad (Bufo terrestris), squirrel tree-
frog (Hyla squirella), and little grass frog (Limnaoedus ocularis) (Conant 1975). With
all the field work in North Carolina's Coastal Plain, S. viridis should have been discov-
ered if it occurred north of the southeastern corner. Perhaps the Virginia population,
if it still exists, is isolated from the main range. Otherwise, southeastern North Carolina
is the northern distributional limit along the Atlantic Coast.

Family Cryptopidae
Subfamily Cryptopinae
Cryptops hyalinus Say 1821
Fig. 12
Cryptops hyalinus: Brimley, 1938:501. Causey. 1940:66. Wray, 1950:156; 1967:156.
Diagnosis. Color uniformly yellow; 21 pairs of legs and pedal segments; ultimate
legs not conspicuously broader than preceding pair; ocelli absent.
Ecology. Occurs primarily under bark of decaying pine logs and stumps; less com-
monly found in litter and in association wth deciduous logs.
Distribution. Statewide; apparently less common in the Coastal Plain.
Remarks. The smallest scolopendromorph in North Carolina, C. hyalinus is abun-
dant under bark of loblolly pine logs in the eastern Piedmont, where it can be mistaken
for geophilomorphs, which it resembles in length and color. It is a common urban
chilopod in the Raleigh-Durham-Chapel Hill area and probably occurs in all major pied-
mont cities. In the mountains it occurs in association with hardwood logs. Although
known from all three physiographic provinces, C. hyalinus has only been collected

BR / pp CP

121 J. T L-


12 *' >i

Fig. 12. Distribution of Cryptops hyalinus in North Carolina.


December, 1987

Shelley: North Carolina Centipedes 505


-- -, .- ,- - T ; -r i l ,/ % o

Fig. 13. Distributions of Theatops posticus, dots. and T. spinicaudus, squares, in
North Carolina.

three times in the Coastal Plain, one each from Cumberland, Sampson, and New
Hanover counties. As considerable chilopod sampling has taken place in the eastern
counties and countless logs have been peeled, this diminished occurrence seems real
and not a reflection of insufficient field work. The small size of C. hyalinus should not
disproportionately affect its discovery in the Coastal Plain in comparison to other re-
I have examined most specimens of Cryptops collected in North Carolina, and all
are C. hyalinus, even ones taken in urban areas. Cryptops hortensis Leach, a European
centipede introduced into urban environments in several eastern states, may occur in
the major cities, but is currently unknown.

Subfamily Theatopinae
Theatops posticus (Say 1821)
Figs. 3, 13

Theatops posticus: Brolemann, 1895:50. Brimley, 1938:501. Causey, 1940:66. Wray,
1950:156: 1967:156.
Diagnosis. Color yellow to yellowish-orange; 21 pairs of legs and pedal segments;
ultimate legs much larger and more heavily sclerotized than preceding pair, prefemora
unarmed; ocelli absent.
Ecology: Primarily inhabits moist deciduous litter, occasionally found in predomi-
nantly pine litter.
Distribution. Statewide; rare in mountains, particularly at high elevations.
Remarks. The records from Goldsboro, Wayne County (Wood 1862, Bollman 1888)
and Duke Forest, Durham County, and Greensboro, Guilford County (Causey 1940,
Chamberlin 1951), are considered valid.

Theatops spinicaudus (Wood 1862)
Figs. 4, 13

Theatops spinicauda: Brolemann, 1896:50-51.
Diagnosis. Color yellow to yellowish-orange; 21 pairs of legs and pedal segments;
ultimate legs much larger and more sclerotized than preceding pair, prefemora with
dorsal, distomedal spines; ocelli absent.

Florida Entomologist 70(4)

December, 1987


Fig. 14. Distribution of Scolopocryptops nigridius in North Carolina.

Ecology. Primarily inhabits moist deciduous litter, occasionally found in predomi-
nantly pine litter.
Distribution. Blue Ridge Province to central Piedmont Plateau.
Remarks. In North Carolina the ranges of T. spinicaudus and posticus overlap,
with the former prevalent in the west and the latter more common in the east (Fig. 13).
Theatops posticus is rare in the mountains and generally absent from higher elevations,
where T. spinicaudus is common, particularly in the Black and Great Smoky Mountains.
Most collections of T. posticus in the Blue Ridge province are from low elevations in
Cherokee and Clay counties, which have piedmont characteristics. The chilopods have
similar ecological preferences but have not been found syntopically. The records of T.
spinicaudus from Mt. Pisgah, Haywood-Transylvania counties (Wray 1950, 1967), are
considered valid.

Subfamily Scolopocryptopinae
Scolopocryptops nigridius McNeill 1887
Figs. 5, 14

Otocryptops nigridius: Brolemann, 1895:50. Brimley, 1938:501. Wray, 1950:156;
Scolopocryptops nigridius: Causey, 1940:66.
Diagnosis. Color dark brownish-orange, with or without irregular purple spots; 23
pairs of legs and pedal segments; ultimate legs not conspicuously thicker than preceding
pair; 2nd antennomeres sparsely hirsute; ocelli absent; head margined laterally; tergites
without complete paramedian sulci.
Ecology. Usually found in moist predominantly pine or deciduous litter also occurs
under rocks and either hardwood or pine logs but rarely under bark.
Distribution. Statewide.
Remarks. The record from Duke Forest, Durham County (Causey 1940), is consi-
dered valid.

Scolopocryptops sexspinosus (Say 1821)
Figs.6-7, 15

Scolopocryptops sexspinosus: Brolemann, 1895:50.
Otocryptops sexspinosus: Brimley, 1938:501. Causey, 1940:67. Wray, 1950:156;


Shelley: North Carolina Centipedes 507

BR / pp CP

Fig. 15. Distributions of Scolopocryptops sexspinosus, dots, and S. peregrinator,
star, in North Carolina.
Diagnosis. Color bright orange to yellowish-orange, without purple patches; 23 pairs

of legs and pedal segments; ultimate legs not conspicuously thicker than preceding legs;
2nd antennomeres densely hirsute; ocelli absent; head margined laterally; tergites with-
out complete paramedian sulci.
Ecology. Usually found in moist litter, either predominantly pine or deciduous; also
occurs under rocks and either hardwood or pine logs but rarely beneath bark.
Fig. 15. Distributions of Scoopocrde.ptops sexspinosus, dots, and S. peregnator,
Remarks. The records from Salem, Forsyth County (Wood 1862, Bollman 1893);olina.

Diagnosis. Chapel Hill, Orange County (Bollman 1893); Greensboro, Gullford Cwithout purple patches; 23 pairsrlin
1951)of legs and Linvilpedal segments; ultimate legs not conspicuously thicker than preceding legs;
2nd antennomeres densely hirsute; ocelli absent; head margined laterally; tergites with-
out complete paramedian sulci.

(Causey 1940) are considered valid. A specimen was found in moist litter, either predominantly pingestive or deciduous; al
ocurs under crocks aned snake (Tantia cor pine logs but rarely beson County, 11.2 km
SE Red Spribution. Statewide.
The two common specrds from Salem, Forsyth County (Woodthe same habitats, occur1893);
Chapel Hill, are about equally (Bollman 1893); Greensboro, Gullford Copically unty (Chamberlin
log.1951); and Linville, Burke Cd Uetz (1979) county, and S. sexspinosus exclusively in leaf litter in east-cen-ties
(Causey 1940)l Illinois, and I have rarely encounter valid. Either species und in the digestive tract of ang logs

in North Carolina. However, Lee (1980) found 63 specimens of S. sexspinosus under
southebark and none in littcrowned snake (Tantral Ohio. Externata) colleted in Robeson County, 11.2 kmeach

other, but they are readily distinguished by the characters in the key and diagnoses.
SE Red Springs.

ScolThe two common species of Scoloocryptops nosus occupy the same habitats, occur55 mm
statewide, are about equally abundant, and can even occur syntopically under the same
log. Summers and Uetz (1979) found S. sexspinosus exclusively in leaf litter in east-cen-
tral Illinois, and I have rarely encountered either species under bark of decaying logs
in North Carolina. However, Lee (1980) found 63 specimens of S. sexspinosus under
bark and none in litter in central Ohio. Externally, the two centipedes resemble each
other, but they are readily distinguished by the characters in the key and diagnoses.
Scolopocryptops sexspinosus tends to be larger, occasionally growing to nearly 55 mm
in length. In any part of North Carolina, urban or rural, one can hardly spend 15
minutes sifting through litter or turning over logs without encountering one of these

Scolopocryptops peregrinator (Crabill 1952), new status
Figs. 8-9, 15-16

Otocryptops gracilis peregrinator Crabill, 1952:124-126, Figs. 4-5.
Scolopocryptops gracilis peregrinator: Crabill, 1960:12. Branson and Batch, 1967:86.
Diagnosis. Color light yellow; 23 pairs of legs and pedal segments; ultimate legs not
conspicuously thicker than preceding pair; 2nd antennomeres sparsely hirsute; ocelli
absent; head not margined laterally; complete paramedian sulci beginning on 4th tergite.

Florida Entomologist 70(4)


Fig. 16. Distribution of Scolopocryptops peregrinator.

Ecology. The two specimens from Pine Ridge, Wolfe Co., Kentucky, were taken ii
March and April 1966 from beneath decaying leaves (Branson & Batch 1967). No habitat
information accompanies the newly reported samples.
Distribution. Southern Pennsylvania to northwestern North Carolina and westward
to eastern Kentucky; in North Carolina, known only from Ashe County (Figs. 15-16)
The range includes parts of the Piedmont Plateau, Blue Ridge, Ridge and Valley, anc
Appalachian Plateau physiographic provinces. Material was examined from the follow-
ing new localities.
PENNSYLVANIA: Lebanon Co., Lebanon, Oct 1892 (NMNH); and unknown site in
"western" part of state (not plotted in fig. 16), date unknown (NMNH).
DISTRICT OF COLUMBIA: Catholic University Campus; Apr 1893 (NMNH).
VIRGINIA:3 Frederick Co., east of Winchester, date unknown (NMNH). Rappahan-
nock Co., 12.8 km NW Sperryville, Oct 1957 (NMNH). Pulaski Co., near Claytor Lake,

'In Virginia, S. peregrinator has also been taken from an unknown site in Shenan-
doah National Park (NMNH), which spans over 160 km (100 miles) of the Blue Ridge
Province and occupies parts of eight counties. For convenience a dot is placed centrally
in the Park area in figure 16.

December, 198'


Shelley: North Carolina Centipedes


12 Oct 1959 (NMNH); and Draper Mtn., 12 Sep 1959 (NMNH). Patrick Co., Pinnacles
of Dan, 14 Apr 1958 (NMNH).
WEST VIRGINIA: Greenbrier Co., White Sulphur Springs. Kate's Mtn., 20 Apr 1968
NORTH CAROLINA: Ashe Co., 4.8 km NW Lansing along SR 1367, 0.3 km SW jet.
SR 1368, 13 Apr 1984 (A4215).

Remarks. Despite considerable collecting in its known range the past century, S.
peregrinator has only been taken 16 times. It may therefore be rare and population
sizes may be small, but seasonality could also explain the paucity of specimens. The
samples were collected in September, October, March, and April, suggesting that sur-
face activity may be restricted to cooler months when less field sampling usually occurs.
Concentrated investigations in the fall and spring may provide a truer picture of the
abundance and distribution of S. peregrinator.
I examined the holotype, the Kentucky specimens, and 25 individuals from the above
sites plus the paratype locality, Woodside, Montgomery County, Maryland, and com-
pared these with specimens of S. gracilis from California. All specimens of S. pereg-
rinator agree closely with the anatomical description by Crabill (1952). The degree of
pilosity of the second antennomere varies but is always considerably less than that of
the third and more distal articles, and the paramedian sulci on the third tergite extend
beyond midlength on a few individuals. Both species lack lateral marginations on the
cephalic plates, but the distinctions between S. peregrinator and gracilis cited by Crabill
(1952) are readily apparent when the two are placed side by side and seem of comparable
magnitude to those between S. sexspinosus and nigridius. The tergal sulcal differences
on segments 2-4 are particularly noteworthy (complete paramedian sulci begin on seg-
ment 4 in S. peregrinator and segment 2 in S. gracilis), as shown in figures 9-10 in
contrast to the pattern in S. sexspinosus (similar to that in S. nigridius) in Figure 7.
The ranges of S. gracilis and peregrinator are segregated by over 3,200 km (2,000 miles)
and neither extant linkage nor the occurrence of an intervening form with the "non-mar-
gined" cephalic plate has been demonstrated. The supposed occurrence of S. gracilis in
Houston, Texas (Chamberlin 1943a), is surely a labeling mixup, as the sample was taken
at the same time and by the same collector, Russell Scott from September to December
1941, as the holotype of the milliped Sigmoria (Falloria) houstoni Chamberlin (1943b).
Shelley and Whitehead (1986) showed that the milliped actually occurs in southeastern
Tennessee, and the chilopod record is probably equally erroneous. I could not locate the
specimen in the NMNH collection in April 1986, and the Houston record of S. gracilis
must be discounted until confirmed with fresh material. Shelley and Whitehead (1986)
defined subspecies in the xystodesmid milliped tribe Apheloriini as "taxa which are
reasonably homogeneous throughout their ranges but which connect with other such
taxa through intergrade or intermediate forms," and where range disjunctions occurred
the populations were considered either separate species or consubspecific. This ap-
proach seems applicable to chilopods and S. peregrinator deserves elevation on a geog-
raphical basis alone, since gene flow is impossible with S. gracilis in California. The
anatomical differences substantiate this change and reflect lengthy geographical isola-
tion, if the chilopods ever shared a common ancestor as the "non-margined" cephalic
plates suggest. Considering the structural differences and the vast geographical segre-
gation, an equally plausible interpretation of the cephalic similarity is that it represents
convergence and that scolopocryptopines displaying it evolved independently in both
eastern and western North America. The true relationships of S. gracilis and pereg-
rinator may therefore lie elsewhere; the latter, for example, may be sister to S.
rubiginosus Koch, a more proximal midwestern species, which exhibits complete
paramedian sulci on most tergites. An analysis of relationships in Scolopocryptops is
beyond the scope of the present study, but peregrinator and gracilis have been repro-

510 Florida Entomologist 70(4) December, 1987

ductively isolated for a long time. I therefore formally elevate peregrinator to the
specific level.


I thank Jonathan Coddington, National Museum of Natural History, Smithsonian
Institution, for loans of non-type specimens and the holotype of Otocryptops gracilis
peregrinator. The following curators provided specimens from their institutions: Nor-
man I. Platnick (American Museum of Natural History), Selwyn S. Roback (Academy
of Natural Sciences), Larry E. Watrous (Field Museum of Natural History), G. B.
Edwards and Howard V. Weems, Jr. (Florida State Collection of Arthropods), and
Herbert W. Levi (Museum of Comparative Zoology, Harvard University). Comparative
California material of Scolopocryptops gracilis was loaned by Charles L. Hogue, Natu-
ral History Museum of Los Angeles County; Andrew A. Weaver, Wooster, Ohio; and
Dr. Coddington. Gerald Summers and Andrew A. Weaver provided advice, specimens,
records from their collections, and constructive comments on drafts of the manuscript.
Alvin L. Braswell and David L. Stephan collected numerous centipedes while on her-
petological field trips; the latter discovered the North Carolina specimen of Scolopo-
cryptops peregrinator. Figures 1-10 were prepared by Renaldo G. Kuhler, North
Carolina State Museum scientific illustrator. This study was supported in part by a
short term visitor award from the Smithsonian Institution, which enabled the author
to search for and examine pertinent specimens during a visit to the National Museum
of Natural History in April 1986.
Contribution No. 663, Bureau of Entomology, Division of Plant Industry Florida
Department of Agriculture and Consumer Services, Gainesville, FL 32602.


BOLLMAN, C. H. 1888. A preliminary list of the Myriapoda of Arkansas, with descrip-
tions of new species. Ent. Amer., 4: 1-8.
--. 1893. The Myriapoda of North America. Bull. U. S. Natl. Mus. No. 46, 210 pp.
BRANSON, B. A. AND D. L. BATCH. 1967. Valley centipedes (Chilopoda; Symphyla)
from northern Kentucky. Trans. Kentucky Acad. Sci., 28: 77-89.
BRIMLEY, C. S. 1938. Insects of North Carolina. North Carolina Department of
Agriculture, Division of Entomology, Raleigh, NC, 560 pp.
BROLEMANN, H. W. 1896. Liste de Myriapodes des Etats-Unis, et principalement de
la Carolina du Nord, faisant parties des collections de M. Eugene Simon. Ann.
Soc. Ent. France., 65: 43-70.
CAUSEY, N. B. 1940. Ecological and systematic studies on North Carolina myriapods.
Unpublished Ph. D. Thesis, Zoology Department, Duke University, Durham,
NC, 181 pp.
.1942. New lithobiid centipedes from North Carolina. J. Elisha Mitchell Sci.
Soc., 58: 79-83.
CHAMBERLIN, R. V. 1911. The Lithobiomorpha of the southeastern states. Ann. Ent.
Soc. America. 4: 32-48.
.1912a. The Henicopidae of America north of Mexico. Bull. Mus. Comp. Zool.,
57: 3-36.
1912b. The Geophiloidea of the southeastern states. Bull. Mus. Comp. Zool.,
54: 407-436.
1913. The lithobiid genera Nampabius, Garibius, Tidabius, and Sigibius.
Bull. Mus. Comp. Zool., 57: 39-104.
S1914. The genus Watobius. Bull. Mus. Comp. Zool., 57: 107-112.
S1917. The Gosibiidae of America north of Mexico. Bull. Mus. Comp. Zool., 57:

-- i

Shelley: North Carolina Centipedes 511

S1922. Further studies on North American Lithobiidae. Bull. Mus. Comp.
Zool., 57: 259-382.
1925a. The Ethopolidae of America north of Mexico. Bull. Mus. Comp. Zool.,
57: 385-437.
1925b. The genera Lithobius, Neolithobius, Gonibius, and Zinapolys in
America north of Mexico. Bull. Mus. Comp. Zool., 57: 441-504.
.1940a. On six new lithobiid centipeds from North Carolina. Proc. Biol. Soc.
Washington, 53: 75-78.
.1940b. Two new lithobiid chilopods from burrows of the Florida Pocket
Gopher. Ent. News, 51: 48-50.
1943a. Some records and descriptions of American chilopods. Proc. Biol. Soc.
Washington, 56: 97-108.
1943b. Some records and descriptions of American diplopods. Proc. Biol. Soc.
Washington, 56: 143-152.
1944. Some centipeds from Georgia. Ent. News, 55: 32-35.
1945. On some centipedes from Georgia. Canadian Ent., 77: 215-216.
-- 1951. Records of American millipeds and centipeds collected by Dr. D. Eldon
Beck in 1950. Great Basin Nat., 11: 27-35.
1958. Some records of chilopods from Florida. Ent. News, 69: 13-14.
CONANT, R. 1975. A Field Guide to Reptiles and Amphibians, Second Edition.
Houghton Mifflin Co., Boston, 429 pp.
CRABILL, R. E., JR. 1950. On a collection of centipedes from western South Carolina.
Ent. News, 61: 199-202.
.1952. A new subspecies of Otocryptops gracilis (Wood) from the eastern
United States, together with remarks on the status of Otocryptops nigridius
(McNeill) and a key to the species of the genus now known to occur east of the
Rocky Mountains (Chilopoda: Scolopendromorpha: Cryptopidae). Ent. News, 63:
1953. A new himantariid from the eastern United States (Chilopoda:
Geophilomorpha: Himantariidae). Bull. Brooklyn Ent. Soc. 48: 85-88.
1955. A checklist of the Chilopoda known to occur in Kentucky. Ent. News,
66: 257-261.
.1958. A new schendylid from the eastern United States, with notes on distri-
bution and morphology (Chilopoda: Geophilomorpha: Schendylidae). Ent. News,
59: 153-160.
S1960. A new American genus of cryptopid centipede, with an annotated key
to the scolopendromorph genera from America north of Mexico. Proc. U. S. Natl.
Mus., 111: 1-15.
- 1969. A new Floridian Cryptops, with a key to the state's species (Chilopoda:
Scolopendromorpha: Cryptopidae). Proc. Biol. Soc. Washington, 82: 201-204.
HEMING, F. (ed.). 1957. Copenhagen Decisions on Zoological Nomenclature. Addi-
tions to and Modifications of, the Regles Internationales de la Nomenclature
Zoologique. International Trust for Zoological Nomenclature, London, 135 pp.
LEE, R. E. 1980. Summer microhabitat distribution of some centipedes in a deciduous
and coniferous community of central Ohio (Chilopoda). Ent. News, 91: 1-6.
LEWIS, J. G. E. 1981. The Biology of Centipedes. Cambridge University Press, Cam-
bridge, England, 476 pp.
RADFORD, A. E., H. E. AHLES, AND C. R. BELL. 1968. Manual of the Vascular
Flora of the Carolinas. University of North Carolina Press, Chapel Hill, NC,
1183 pp.
SAY, T. 1821. Descriptions of the Myriapodae of the United States. J. Acad. Nat. Sci.
Philadelphia, 2: 102-114.
SHELLEY, R. M. 1978. Class Chilopoda. p. 221, In: Zingmark, Richard G., ed., An
Annotated Checklist of the Biota of the Coastal Zone of South Carolina. Univer-
sity of South Carolina Press, Columbia, SC, 364 pp.
--- AND D. R. WHITEHEAD. 1986. A reconsideration of the milliped genus Sig-
moria, with a revision of Deltotaria and an analysis of the genera in the tribe

512 Florida Entomologist 70(4) December, 1987

Apheloriini (Polydesmida: Xystodesmidae). Mem. American Ent. Soc. No. 35,
223 pp.
SUMMERS, G. 1979. An illustrated key to the chilopods of the north-central region of
the United States. J. Kansas Ent. Soc., 52: 690-700.
-- AND G. W. UETZ. 1979. Microhabitats of woodland centipedes in a streamside
forest. American Midl. Natur., 102: 346-352.
WOOD, H. C. 1862. On the Chilopoda of North America, with a catalogue of all the
specimens in the collection of the Smithsonian Institution. J. Acad. Nat. Sci.
Philadelphia, Ser. 2, 5: 2-52.
.1865. The Myriapoda of North America. Trans. American Philosoph. Soc., 13:
WRAY, D. L. 1950. Insects of North Carolina, Second Supplement. North Carolina
Department of Agriculture, Raleigh, NC, 181 pp.
.1967. Insects of North Carolina, Third Supplement. North Carolina Depart-
ment of Agriculture, Raleigh, NC, 181 pp.

-L -- e-- a a a a -a a a----L--


Department of Entomology
Virginia Polytechnic Institute and State University
Blacksburg, VA 24061


During our study of the genus Chionaspis in North America, two new species were
found. Chionaspis hamoni, n. sp. is described from Salix spp. in Florida, and Chionas-
pis gilli, n. sp. is described from Tamarix spp. in the southwestern United States. The
adult females of both species are illustrated. The diagnostic characters, hosts, distribu-
tions, relationships with other species in the genus are also given and discussed.


Se encontraron dos species nuevas durante nuestro studio del g6nero Chionaspis
en Norte America. Chionaspis hamoni, nueva especie, es descrita de Salix spp., en la
Florida y C. gilli, nueva especie, es descrita de Tamarix spp. en el suroeste de los
Estados Unidos. Se ilustran las hembras adults de ambas species. Se presentan y
discuten los caracteres diagn6sticos, hospederos, distribuci6n, y relaciones con otras
species del genero.

Adult females of the 59 known species in the genus Chionaspis are similar to one
another in general appearance and morphology. The pygidial lobes bear the characters
most useful for identification. However, identification is not simply a matter of discern-
ing differences between specimens, and the non-specialist may find it exceedingly dif-
ficult to identify an unknown specimen for the following reason. Several species in the
genus have recently been discovered to be polymorphic dimorphicc or trimorphic); this

Liu & Kosztarab: New Species of Chionaspis

polymorphism is associated with different feeding sites on the same host. Furthermore,
because of their polymorphism, some morphs considered earlier as distinct species have
now been combined as different forms of the same species. The morphological variation
within one of these polymorphic species is often greater than that between some other
genera of diaspidids. Consequently, not only may specimens be hard to place correctly,
but some specific identifications reported in the earlier literature may not serve as
reliable records.
The following descriptions are based on external morphological characters, especially
of the pygidial lobes, dorsal macroducts and other characteristics of the pygidial margin.
The first species (C. hamoni) was considered as a variant of C. longiloba Cooley, while
the second species (C. gilli) was identified as C. etrusca Leonardi by many early en-
Measurements are given in microns, unless stated otherwise. The measurements or
ranges given before the parenthesis refer to the holotype, while those inside the paren-
thesis refer to the paratype series.

(Figure 1)

Suggested Common Name: Florida willow scale.
Test of Adult Female (Fig. 1A): Elongate oystershell-shaped; white or dirty white,
sometimes of same color as bark, thus difficult to detect; rather small, about 1.5-2.0 mm
long, and moderately convex. Exuviae terminal, occupying about one third of total
length of test, yellowish brown. Ventral test very thin, white.
Body of Adult Female (Fig. 1B): Spindle-shaped, produced laterally and broadened
posteriorly, widest at abdominal segment I or II. On microscopic slide, 1577 (715-1177)
long and 686 (445-637) wide.
Pygidial Margin (Fig. 1C): Oval or somewhat triangular, 255 (211-250) long and 539
(402-471) wide at base, with sclerotized apex. Median lobes well developed, broad and
stout, with fine serrations on the rounded margins, base of each lobe wider than or
nearly equal to its length; two lobes contiguous or with a narrow separation at mesal
base; a sclerotized sigmoid or crescentlike bar at base of each lobe; median yoke distinct,
stout and well sclerotized, connecting the two lobes. Second pair of lobes bilobed, con-
spicuous; inner lobule elongate, close to median lobe, somewhat oblique; outer lobule
shorter and smaller; both lobules with fine marginal serrations. Third pair of lobes with
only inner lobule distinct, this lobule even larger than that of second pair; outer lobule
reduced; both lobules with similar serrations as those of second pair. Gland spines,
macroducts and setae well developed.


Macroducts (Fig. 1D, E): Two-barred, 6.6-22.8 long and 7.6-9.5 wide, arranged in
submedian groups on abdominal segments II-VI, in submarginal groups on II-V, in
marginal groups on III-VII and laterad of median lobes on each side of body; number
in each group variable and with ranges as follows: submedian: 8-11 (4-11) on abdominal
segment II, 13-14 (6-14) on III, 11 (4-12) on IV, 8 (4-8) on V, and 5-8 (3-7) on VI;
submarginal: 13-17 (6-16) on II, 12-13 (5-13) on III, 10-12 (6-12) on IV, and 6-7 (5-9) on
V; marginal: 1-2 on III, 2-3 on IV, 2 on each of V and VI, respectively, and 1 laterad
of median lobe.
Small Macroducts (Fig. 1F): Two-barred, 6.7-14.3 long and 2.9-7.6 wide, clustered
on marginal-submarginal areas of abdominal segments I-III and thoracic segments of


Florida Entomologist 70(4)



December, 1987





L /
,, f
'. ** '

3/ Y ; c .... i I

Fig. 1. Chionaspis hamoni, n. sp., adult female. A, test; B, body (dorsal/ventral
view); C, pygidial margin; D and E, regular macroducts; F, small macroduct; G, micro-
duct; H, body seta; I, antenna; J, anterior spiracle; K, trilocular pores near spiracle; L,
gland spine; M and N, gland tubercles; O, small macroduct; P, microduct; Q, quin-
quelocular pore from perivulvar pore groups.

each side of body; number varies widely with ranges as follows: 1-2 on prothorax, 2-7
on mesothorax, 6-12 on metathorax, 4-7 on abdominal segment I, 4-10 on II, and 2-5 on
Microducts (Fig. 1G): One-barred, 7.6-9.5 long and 1.4-2.0 wide at orifice, scattered
on dorsal surface from cephalic region to pygidium; number of each body region or
segment as follows: 5-8 on cephalic region, 3-6 on prothorax, 2-5 on each of mesothorax
and metathorax, 2-6 on each of abdominal segments I-II, and few or absent on abdominal
segments III-V and pygidium.

Liu & Kosztarab: New Species of Chionaspis 515

Setae (Fig. 1H): Hairlike, long and slender, 3.8-11.4 long; 5-8 scattered on cephalic
region, 1 on each margin of thoracic segments and abdominal segments I-III; also 1 on
each margin of abdominal segments IV-VI and laterad on pygidial lobes, these latter
much longer and thicker than those on anterior part of body, 11.4-19.0 long.
Anal Area: Anal opening circular or slightly oval, 11.9 (11.9-13.8) in diameter, lo-
cated about one fourth anterior on pygidium, opposite to median group of perivulvar
pores of ventral derm under microscope. Distance from anterior margin of anus to
mid-point of line of abdominal segments V and VI, 73 (67-89), and from posterior margin
to base of median lobes, 158 (122-148).


Antennae (Fig. 11): Reduced to small tubercles, 8.9 (7.9-9.9) long, 9.9 (7.9-11.9) wide
at base, each with 1 long, usually curved and slender fleshy seta, 19.8 (15.8-21.7) long,
with 2 short and stout sensory terminal setae, 2.9-3.8 long. Distance between antennal
bases, 106.7 (49.4-106.7).
Clypeolabral Shield: Similar in general appearance to other species in this genus,
168 (105-160) long and 120 (98-119) wide.
Labium: Cuplike, 49.0 (34.3-47.4) long and 58.3 (43.5-63.7) wide.
Spiracles: Anterior spiracle (Fig. 1J), 27.7 (23.7-29.6) long and atrium 8.5-9.5 in
diameter, each with 6-8 (3-9) associated trilocular pores (Fig. 1K), each pore 2.3-3.1 in
diameter; posterior spiracles similar in shape to anterior ones but slightly smaller, 27.7
(15.8-23.7) long, each with 2 (1-5) associated trilocular pores; sometimes trilocular pores
in cluster, especially those near anterior spiracles; this clustering makes counting dif-
Gland Spines (Fig. 1L): Marginal, well developed, 20.0-38.0 long, each with an
associated one-barred microduct, arranged on abdominal segments IV-VII and laterad
of median lobe on each side of body; usually 4 (2-5) on IV, 2 (1-2) on V, 2 (1-2) on VI
and VII, respectively, and 1 laterad on median lobe.
Gland Tubercles (Fig. 1M, N): Conical basally and spinelike distally, 9.5-15.2 long
and 2.9-4.8 wide at base, each with an associated one-barred microduct, located sub-
marginally on thoracic segments and abdominal segments I-III, and becoming gradually
smaller anteriorly; usually 2-3 (0-3) on prothorax, 2 (0-2) on mesothorax, 1-2 (0-3) on
metathorax, 2-3 (0-4) on abdominal segment I, 4-6 (2-6) on II, 5 (2-7) on III.
Small Macroducts (Fig. 10): Same shape, size and location and nearly same number
as, or slightly more numerous than, those on dorsal surface.
Microducts (Fig. 1P): Shape, size and location about same as for those on dorsal
surface except for being more numerous on cephalic region, prepygidial segments and
pygidial area.
Setae (Fig. 1H): Same as those on dorsal surface except for their number being
Vulvar Area: Vulvar opening situated about middle of pygidium; perivulvar pores
quinquelocular (Fig. Q), clustered in 5 groups median: 25 (15-24); anterolateral: 26-27
(15-35); and posterolateral: 13-16 (10-23). Diameter of pore, 3.9-5.5.

Material Examined:

Holotype: 1 adult female on 1 slide, Steinhatchee, Florida (FL), XII-2-1981 by F.
McHenry on Salix nigra, holotype and most paratypes are deposited in the Florida
State Collection of Arthropods (FSCA) unless indicated otherwise in parentheses.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs