Foraging behavior of Microplitis croceipes (Cresson) (Hymenoptera: Braconidae), a parasitoid of Heliothis (Lepidoptera: ...

MISSING IMAGE

Material Information

Title:
Foraging behavior of Microplitis croceipes (Cresson) (Hymenoptera: Braconidae), a parasitoid of Heliothis (Lepidoptera: Noctuidae) species
Physical Description:
xii, 213 leaves : ill. ; 29 cm.
Language:
English
Creator:
Eller, Fred Joseph, 1957-
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 183-212).
Statement of Responsibility:
by Fred Joseph Eller, III.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001581911
notis - AHK5829
oclc - 22973978
System ID:
AA00003319:00001

Full Text











FORAGING BEHAVIOR
OF
Microplitis croceipes (CRESSON)
(HYMENOPTERA: BRACONIDAE),
A
PARASITOID
OF
Heliothis
(LEPIDOPTERA: NOCTUIDAE)
SPECIES






By

FRED JOSEPH ELLER, III








A
DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF
THE UNIVERSITY OF FLORIDA
IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY







UNIVERSITY OF FLORIDA

1990












ACKNOWLEDGMENTS

I wish to thank Dr. Jim Tumlinson for his support, advice,

patience, and independence allowed throughout the course of this

research. I would also like to thank Mr. Vic Bauder, Ms. Peggy Brennan,

Ms. Barb Dueben, Mr. Bob Heath, Mr. Rob Murphy, Mr. Tommy Proveaux, Mrs.

Joan Russo, Mrs. Elaine Turner, Mr. Ted Turlings and the rest of the

staff of the Insects Attractants, Behavior, and Basic Biology Research

Laboratory, ARS-SEA, USDA for their considerable help. The members of

my graduate committee, Drs. Joe Lewis, Jim Nation, Reece Sailer

(deceased), and Merle Battiste deserve credit for their encouragement

and guidance. I would also like thank the staff of the Entomology and

Nematology Department for their support and especially Dr. Stratton Kerr

for his considerable help in dealing with the University bureaucracy.

Mr. Victor Chew cheerfully provided statistical advice and analyses. I

thank Mrs. Christine Gaetke for her generous support. Finally, I wish

to thank my wife, Denise, for her support, patience and understanding

during this research.













TABLE OF CONTENTS

page

ACKNOWLEDGMENTS.................. ................................... ii

LIST OF TABLES...................................................... vi

LIST OF FIGURES..................................................... ix

ABSTRACT ............................................................ xi

CHAPTERS
I LITERATURE REVIEW AND RESEARCH AIMS.......................... 1

Biology of Microplitis croceipes (Cresson)................ 1
The Biological Control of Heliothis Species.............. 1
Taxonomy and Systematics................................ 1
Adult Description........................................ 2
Distribution and Seasonal Occurence..................... 2
Host Range............................................... 3
Host Plant Associations.................................. 3
Rearing................................ ................ 6
Life Cycle............................................... 9
Ecology................................................. 16
Use of Kairomones by Entomophagous Insects................. 20
Host Selection Process.................................. 20
Kairomone-Mediated Host Selection....................... 20
Host and Prey Insects Producing Kairomones............... 22
Entomophagous Insects Using Kairomones................... 22
Long-Range Kairomones................................... 22
Short-Range Kairomones................................... 30
Miscellaneous Responses to Kairomones................... 36
Sexual Differences in Kairomonal Responses.............. 36
Host and Prey Location versus Food or Mate
Location............................................... 37
Uses for Kairomones .................................... 38
Considerations in the Study and Use of
Kairomones ......................................... 39
Research Aims .............................................. 40






II OLFACTOMETRIC STUDIES OF HOST-LOCATION BY
Microplitis croceipes (CRESSON)........................ 42

Introduction............................................ 42
Materials and Methods................................... 42
Results................................................. 49
Discussion.............................................. 54

III SOURCE OF VOLATILES MEDIATING THE HOST-LOCATION Flight
BEHAVIOR OF Microplitis croceipes (CRESSON) ........... 58

Introduction........................................... 58
Materials and Methods................................... 58
Results.................................................. 64
Discussion.............................................. 75

IV INTRASPECIFIC COMPETITION IN Microplitis croceipes
(CRESSON)............................................. 79

Introduction............................ ................ 79
Materials and Methods ................................... 79
Results and Discussion.................................. 82

V PARASITOID ALTERED HOST BEHAVIOR AND PASSIVE HOST
DISCRIMINATION IN Microplitis croceipes (CRESSON)..... 90

Introduction............................ ................ 90
Materials and Methods ................................... 91
Results.................................................. 94
Discussion........................................... .... 95

VI ISOLATION AND IDENTIFICATION OF A LONG-RANGE KAIROMONE
USED BY Microplitis croceipes (CRESSON)............... 103

Introduction............................................ 103
Materials and Methods................................... 104
Results................................................. 109
Discussion.............................................. 114

VII FACTORS AFFECTING THE OVIPOSITION BEHAVIOR OF Microplitis
croceipes (CRESSON) TOWARDS AN ARTIFICIAL SUBSTRATE... 122

Introduction............................................ 122
Materials and Methods................................... 123
Results................................................. 127
Discussion. ............................................. 135







VIII INTERACTIONS BETWEEN Microplitis croceipes (CRESSON) AND
A NUCLEAR POLYHEDROSIS VIRUS OF Heliothis zea
(BODDIE).............................................. 141


Introduction................................... ..
Materials and Methods ...........................
Results .........................................
Discussion.......................................


141
142
144
150


IX EFFECT OF HOST DIET ON THE FLIGHT RESPONSE OF
Microplitis croceipes (CRESSON) ........................


Introduction....................................
Materials and Methods ...........................
Results..........................................
Discussion... ................................ ..


155
156
160
170


X SUMMARY AND CONCLUSIONS ................................... 178


REFERENCES .............................................


....... 183


BIOGRAPHICAL SKETCH............................................ 213












LIST OF TABLES


Table Page


1-1. Cultivated host plants on which Microplitis croceipes has
been reported to attack Heliothis............................. 4

1-2. Non-cultivated host plants on which Microplitis croceipes has
been reported to attack Heliothis............................. 7

1-3. Mean lengths of stages for Microplitis croceipes............... 17

1-4. Identified long-range kairomones used by entomophages
attacking scolytids........................................... 24

1-5. Identified long-range kairomones used by entomophages
attacking non-scolytids........................................ 27

1-6. Identified short-range kairomones which elicit kinetic,
tactic, antennal and abdomenal host examination behaviors in
entomophagous insects ......................................... 33

.-7. Identified short-range kairomones which elicit oviposition
in entomophagous insects ..................................... 35

2-1. Response of inexperienced and experienced female Microplitis
croceipes to four concentrations of plant-host complex odors.. 51

2-2. Response of inexperienced female Microplitis croceipes to
individual components of the plant-host complex................ 53

2-3. Response of inexperienced female Microplitis croceipes to
five concentrations of collected plant-host complex volatiles
placed on filter paper ........................................ 55

3-1. Flight types exhibited by Microplitis croceipes to
plant-host complex, simulated plant-host complex, and
control odors ................................................. 65

3-2. Flight types exhibited by female Microplitis croceipes to
individual components removed from the plant-host complex..... 68

3-3. Flight types exhibited by female Microplitis croceipes to
individual components of the plant-host complex............... 70

vi







3-4. Flight types exhibited by female Microplitis croceipes to
plant-host complex, plant-host complex + artificial damage,
feeding larvae, and nonfeeding larvae.......................... 73

4-1. Percent parasitoid emergence and mean emergence times......... 83

5-1. Effect of parasitization on the attractiveness of Heliothis
larvae and their feces ......................................... 96

6-1. Percentage of female Microplitis croceipes making complete
flights to gravity flow fractions of Heliothis feces extracts. 111

6-2. Percentage of female Microplitis croceipes making complete
flights to high performance liquid chromatography fractions
of Heliothis feces extracts................................... 112

7-1. Effect of artificial oviposition substrate shape on response
of Microplitis croceipes...................................... 128

7-2. Effect of color of artificial oviposition substrate on
ovipositional response of Microplitis croceipes............... 130

8-1. Virus transmission via ovipositor experiment: fate of
experimental larvae......................................... ... 145

8-2. Virus transmission via ovipositor experiment: developmental
times for viral mortality, Microplitis croceipes cocoon
formation, and Heliothis zea pupation......................... 147

8-3. Virus-parasitoid competition experiment: fate of
experimental larvae and developmental times for viral
mortality...................................................... 148

9-1. Effect of a single preflight experience on the choice flight
response of Microplitis croceipes: Cotton versus
artificial diet............................................... 164

9-2. Effect of a single preflight experience on the choice flight
response of Microplitis croceipes: Wild geranium
versus artificial diet........................................ 165

9-3. Effect of a single preflight experience on the choice flight
response of Microplitis croceipes: Cotton versus
wild geranium........................................ ......... 167

9-4. Effect of a single preflight experience on the choice flight
response of Microplitis croceipes: Cowpea versus
artificial diet............................................... 168


Page


Table










9-5. Effect of a single preflight experience on the choice flight
response of Microplitis croceipes: Cotton versus
cowpea. ...................................................... 169


9-6. Effect
flight
versus

9-7. Effect
flight
versus


of repeated (4X) preflight experiences on the choice
response of Microplitis croceipes: Cotton
cowpea.................................................. 171

of repeated (2X) preflight experiences on the choice
response of Microplitis croceipes: Cotton
cowpea................................................. 172


viii


Table


Page











LIST OF FIGURES


Figure Page


2-1. Perspective view of four-choice olfactometer.................. 44

2-2. Schematic diagram of air inlet system for each arm of
4-choice olfactometer. (A) Olfactometer. (B) Parasitoid
trap vial. (C) Odor chamber. (D) Humidifying vial.
(E) Flowmeter.................................................. 45

3-1. Effects of control, plant-host complex, and simulated
plant-host complex odors on behaviors exhibited by
Microplitis croceipes......................................... 66

3-2. Effects of removing individual components of the plant-host
complex on behaviors exhibited by Microplitis croceipes....... 69

3-3. Effects of individual components of the plant-host complex on
behaviors exhibited by Microplitis croceipes................... 72

3-4. Effects of artificial damage on the plant-host complex and
the effect of active larval feeding on behaviors exhibited
by Microplitis croceipes....................... ................ 74

4-1. Effect of time between initial parasitization and
superparasitization on percent female Microplitis croceipes
emergence using mated and unmated females ..................... 85

6-1. Percentage complete flights by female Microplitis croceipes
towards five dosages of a hexane extract of Heliothis feces... 110

6-2. Gas-liquid chromatograph of HPLC fraction 16 on OV-101 (A)
and CPS-1 (B) columns.......................................... 113

6-3. Electron impact (A) and chemical ionization (B) mass spectra
of HPLC fraction 16........................................... 115

6-4. Structure of trans-phytol ((E)-3,7,11,15-tetramethyl-2-
hexadecen-l-ol. .............................................. 116

6-5. Proton magnetic resonance spectra of synthetic trans-phytol
(A) and HPLC fraction 16 (B).................................. 117







Figure


6-6. Percentage complete flights by female Microplitis croceipes
towards five dosages of synthetic trans-phytol................ 118

7-1. Effect of blue food color concentration on oviposition
behavior of Microplitis croceipes towards an artificial
substrate..................................................... 132

7-2. Effect of number of agar drops per dish on oviposition
behavior of Microplitis croceipes towards an artificial
substrate..................................................... 133

7-3. Effect of agar drop size on percent oviposition (A) and
mean number of eggs laid (B) by Microplitis croceipes.......... 134

9-1. Effect of host species/diet combination and experience on the
no-choice flight response of Microplitis crociepes............ 161

9-2. A comparison of preflight experiences on the no-choice flight
response of Microplitis croceipes............................. 162


Page













Abstract of Dissertation Presented to the Graduate School of the
University of Florida in Partial Fulfillment of the Requirements for the
Degree of Doctor of Philosophy

FORAGING BEHAVIOR OF Microplitis croceipes (CRESSON)
(HYMENOPTERA: BRACONIDAE), A PARASITOID OF
Heliothis (LEPIDOPTERA: NOCTUIDAE) SPECIES

By

Fred Joseph Eller, III

May 1990

Chairman: J. H. Tumlinson, III
Major Department: Entomology and Nematology


The parasitoid, Microplitis croceipes (Cresson) (Hymenoptera:

Braconidae), is an important parasitoid of Heliothis species

(Lepidoptera: Noctuidae). The following research examined several

aspects of M. croceipes' foraging behavior.

In both an olfactometer and wind tunnel, females but not males,

were attracted to plant-host complex odors. Experience with the PHC

enhanced subsequent parasitoid response. Although leaves are somewhat

attractive, larval feces are the most important single cue eliciting

host-location behaviors in M. croceipes.

Females responded in a dose-dependent fashion to hexane extracts

of larval feces. The feces extract was fractionated by gravity flow

chromatography and high performance liquid chromatography (HPLC).

Trans-phytol was isolated and identified from an HPLC fraction and

synthetic trans-phytol elicits host-location behaviors in M. croceipes.






Progeny survival rates for a superparasitizing female are equal

to those for the female ovipositing first when the two ovipositions

occur on the same day. However, progeny survival rates for a

superparasitizing female decrease with increased time between the

initial oviposition and the superparasitization.

The difference in the attractiveness of unparasitized and

parasitized larvae increases with the length of parasitization.

Parasitized larvae produce less feces than do unparasitized larvae;

however, on an equal weight basis their feces are equally attractive.

M. croceipes can use quantitative differences in feces production rates

to passively discriminate between unparasitized and parasitized

Heliothis larvae.

Females with a larval oviposition experience laid more eggs in

artificial oviposition substrates (AOSs) than did inexperienced females.

Dark colored AOSs elicited more ovipositions than light AOSs. Although

c sitions per petri dish increased with increased AOS density

S,.JS'/dish), ovipositions per AOS decreased. The preferred size of AOS

was found to be ca. 35-40 pl in volume.

Female parasitoids did not transmit a nuclear polyhedrosis virus

(NPV) of Heliothis from infected larvae to uninfected larvae via

oviposition. Although parasitization did not decrease virus-induced

mortality, it did increase time to virus-induced mortality.

Oviposition per se had little effect on female flight response.

Innate preferences, priming and associative learning all play a role in

the foraging behavior of female M. croceipes. Repeated experiences can

affect the preferences of females.












CHAPTER I
LITERATURE REVIEW AND RESEARCH AIMS


Biology of Microplitis croceipes (Cresson)


The Biological Control of Heliothis Species

The tobacco budworm, Heliothis virescens (F.) and the corn earworm

(also known as the cotton bollworm and the tomato fruitworm), Heliothis

zea (Boddie) (Lepidoptera: Noctuidae) are two of agriculture's most

important pests (Sparks 1981). Problems associated with the chemical

control of these pests (ie. resistance, residues and cost) have given

impetus to the development of alternate control measures (Stadelbacher

et al. 1984). Attention has been especially focused on the biological

control of these two pests. Microplitis croceipes (Cresson)

(Hymenoptera: Braconidae) is the most commonly reported larval

parasitoid of Heliothis and it is believed to offer great potential as a

biological control agent against Heliothis (Knipling and Stadelbacher

1983). Currently, M. croceipes is under extensive study throughout the

United States both in federal laboratories and at state universities.

Taxonomy and Systematics

Microplitis croceipes is a member of the subfamily Microgasterinae

and was originally described in the genus Microgaster by Cresson (1872).

Muesebeck (1922) subsequently assigned croceipes to the genus

Microplitis and placed M. nigriceDs in synonymy with it.








Microplitis croceipes can be keyed to subfamily using Marsh

(1963),to genus using Marsh (1971), and to species using Muesebeck

(1922). Marsh (1978) provides a key which separates M. croceipes from

other braconids attacking Heliothis species.

Adult Description

Descriptions of adult M. croceipes are provided by Cresson (1872),

Quaintance and Brues (1905), Muesebeck (1922), Winburn and Painter

(1932), and Marsh (1978). An adult is figured by Quaintance and Brues

(1905) and a photograph is provided by Lewis (1970a). Generally,

adults have dark heads, antennae, thoraxes and wings, golden legs, and

yellow-orange abdomens with dark genital regions. Females can be

distinguished from males by their short ovipositors, larger dark genital

region, and shorter antennae (4 mm versus 6 mm) (Bryan et al. 1969a,

Lewis 1970a, Danks et al. 1979).

Distribution and Seasonal Occurrence

Microplitis croceipes is widely distributed and occurs throughout

most of the United States. It occurs from New Jersey to Georgia, west

to New Mexico, Arizona, Utah and Oregon (Muesebeck et al. 1951, Krombein

and Burks 1967, Marsh 1978).

Vickery (1925) reported M. croceipes emerging from Heliothis

collected during April in Texas. Rathman and Watson (1985) reported M.

croceipes attacking Heliothis as early as April and May in Arizona.

Microplitis croceipes have also been collected in Arizona between June

and October (Butler 1958). Soteres et al. (1984) reported M. croceipes

attacking Heliothis between May and September in Oklahoma, and

Stadelbacher et al. (1984) reported parasitization of Heliothis by M.








croceipes between April and August in Mississippi. Microplitis

croceipes has also been reported to attack Heliothis between June and

September in North Carolina (Neunzig 1969) and June through October in

Texas (Puterka et al. 1985).

Host Range

Microplitis croceipes has long been recognized as an important

parasitoid of H. zea and H. virescens (Quaintance and Brues 1905, Lewis

and Brazzel 1966, 1968, Snow et al. 1966, Bottrel et al. 1968, Neunzig

1969, Lewis and Snow 1971, Young and Price 1975, Smith et al. 1976,

Harding 1976, Marsh 1978, Stadelbacher et al. 1984, King et al. 1985).

Microplitis croceipes has also been reported to occur on other Heliothis

species, including H. subflexa (Gn.) (Lewis 1970b, Smith et al. 1976),

H. phloxiphaga Grote and Robinson (Bryan et al. 1969a, Rathman and

Watson 1985), and H. obsoleta (Fab.) (Vickery 1925).

Although Spodoptera frugiperda (J. E. Smith), S. exigua (Hubner)

and Trichoplusia ni (Hubner) (Lepidoptera: Noctuidae) are accepted by M.

croceipes for oviposition, they are accepted less readily than Heliothis

and parasitoid development does not occur on these species (Lewis

1970b). Microplitis croceipes will only rarely oviposit in Cadra

(-Ephestia) cautella (Walker), Plodia punctella (Hubner) and

Elasmopalpus lignosellus (Zeller) (Lepidoptera: Pyralidae) and again

parasitoid development does not occur on these species (Lewis 1970b).

Host Plant Associations

Cultivated host plants. Microplitis croceipes has been reported to

attack Heliothis larvae on a wide variety of cultivated crops (Table 1-

1). Mueller (1983) suggests that plant species may vary in their







Table 1-1. Cultivated host plants on which Microplitis croceipes has
been reported to attack Heliothis.


Plant Genus Family Reference


alfalfa Medicago


corn


Zea


cotton Gossypium


Leguminosae














Graminaceae








Malvaceae


Winburn and Painter 1932

Butler 1958

Bottrell et al. 1968

Young and Price 1975

Smith et al. 1976

Soteres et al. 1984

Puterka et al. 1985

Bibby 1942

Butler 1958

Smith et al. 1976

Puterka et al. 1985

Butler 1958

Bottrell et al. 1968

Neunzig 1969

Graham et al. 1972

Shepard and Sterling 1972

Burleigh 1975

Young and Price 1975

Smith et al. 1976

Burleigh and Farmer 1978

Pitre et al. 1978

Pair et al. 1982








Table 1-1--continued.


Plant Genus Family Reference


cotton Gossypium





potato Solanum

sesame Sesamum

sorghum Sorghum














soybean Glycine










tobacco Nicotiana

tomato Lycopersicum


Malvaceae





Solanaceae

Pedaliaceae

Graminaceae














Leguminosae










Solanaceae

Solanaceae


Stadelbacher et al. 1984

King et al. 1985

Puterka et al. 1985

Puterka et al. 1985

Pair et al. 1982

Winburn and Painter 1932

Bibby 1942

Butler 1958

Graham et al. 1972

Young and Price 1975

Smith et al. 1976

Puterka et al. 1985

Smith et al. 1976

Pitre et al. 1978

McCutcheon and

Turnipseed 1985

Stadelbacher et al. 1984

Neunzig 1969

Lewis and Brazzel 1966

Graham et al. 1972







attractiveness, host accessibility and/or masking of attractiveness to

M. croceipes. For example, parasitization of H. zea on corn is reported

to be low (Lewis and Brazzel 1968, Neunzig 1969, Graham et al. 1972,

Smith et al. 1976), and it has been suggested that the protected feeding

habit of H. zea on corn is the reason (Lewis and Brazzel 1968, Smith et

al. 1976) and that corn lacks the appropriate attractive chemicals

(Smith et al. 1976, Nordlund and Sauls 1981).

Mueller (1983) reported that parasitization of Heliothis on tomato

was also low and suggested that defensive compounds (eg. alpha tomatine)

may be the reason. However, Graham et al. (1972) and Lewis and Brazzel

(1968) reported 27% and 75% parasitization of Heliothis on tomato by M.

croceipes, respectively.

Non-cultivated host plants. Although H. zea and H. virescens have

been reported to attack a wide variety of non-cultivated host plants

(Chamberlain and Tenhet 1926, Barber 1937, Brazzel et al. 1953, Neunzig

1963, Snow and Brazzel 1965, Snow et al. 1966, Graham and Robertson

1970, Roach 1975, Stadelbacher 1979), M. croceipes is reported to attack

Heliothis on non-cultivated plants relatively infrequently (Table 1-2).

Rearing

Methods for maintaining laboratory colonies of M. croceipes on

bollworms have been described by Bryan et al. (1969b), Lewis and Burton

(1970) and Powell and Hartley (1987). Microplitis croceipes has also

been reared on budworms according to the methods of Vinson et al.

(1973). Methods for maintaining host insect colonies have been

described by Brazzel et al. (1961), Vanderzant et al. (1962), Berger









Table 1-2. Non-cultivated host plants on which Microplitis croceipes
has been reported to attack Heliothis.


Plant Genus Family Reference


crimsom clover










cranesbill

















daisy fleabane



ground cherry

johnsongrass

peppergrass

redstem filaree


Trifolium










Geranium


Leguminosae










Geraniaceae


Erigeron Compositae


Physalis

Sorghum

Lepidium

Erodium


Solanaceae

Graminaceae

Cruciferae

Geraniaceae


Polygonum Polygonaceae


Lewis and Brazzel 1968

Meuller and Phillips

1983

Stadelbacher et al.

1984

Snow et al. 1966

Lewis and Brazzel

1966

Smith et al. 1976

Meuller and Phillips

1983

Stadelbacher et al.

1984

Rathman and Watson

1985

Smith et al. 1976

Roach 1975

Butler 1958

Rathman and Watson

1985

Smith et al. 1976


smartweed









Table 1-2--continued.


Plant Genus Family Referenece


spiderflower

Texas bluebonnet

Texas paintbrush

tobacco

velvet-leaf

vetch


Cleome

Lupinus

Castilleja

Nicotiana

Abutilon

Vicia


Capparidaceae

Leguminosae

Scophulariaceae

Solanaceae

Malvaceae

Leguminosae


Lewis and Brazzel 1968

Eger et al. 1982

Eger et al. 1982

Graham et al. 1972

Smith et al. 1976

Meuller and Phillips

1983









(1963) and Pantana (1969). In addition, research on the in vitro

rearing of M. croceipes is currently in progress (Greany 1986).

Life Cycle

Adult emergence and longevity. Adults emerge from their cocoons by

cutting a cap off the end of the cocoon (Winburn and Painter 1932, Bryan

et al. 1969a). A cocoon from which an adult has emerged is figured by

Quaintance and Brues (1905). Males emerge slightly before females (8

and 9 days post-cocoon formation, respectively) (Bryan et al. 1969a,

Lewis and Burton 1970, Hopper 1986). Lewis (1970a) reported tha-

emergence occurs in the morning slightly after illumination, howev..

some parasitoids emerge before illumination (personal observation). In

the laboratory, adults are relatively long-lived, and have been reported

to live up to several weeks if water and sugars are provided (Bryan et

al. 1969a, 1969b, Lewis and Snow 1971).

Mating behavior. Evidence exits for a sex pheromone in M.

croceipes. Males become excited and fan their wings in the presence of

females (Lewis 1970a, Lewis and Snow 1971), and females have been

demonstrated to attract males in the field (Powell and King 1984).

In the laboratory, mating takes place as soon as both sexes are

present (Bryan et al. 1969a, Lewis 1970a). Unmated females produce only

male progeny (ie. M. croceipes is arrhenotokously parthenogenic) (Bryan

et al. 1969a, Lewis and Snow 1971). Mated females produce both male and

female progeny at a ratio of approximately unity (Bryan et al. 1969a,

Lewis and Burton 1970, Lewis and Snow 1971, Hopper and King 1984a).

King et al. (1985), however, reported a female:male sex ratio of 1.6:1.0

from field-collected Heliothis.






10

Host finding. Powell and King (1984) reported that although female

M. croceipes were active all day, their peak activity was in the

morning. Hopper and King (1986) reported that in field cages containing

wild geranium or cotton, the number of Heliothis parasitized by M.

croceipes increased linearly with host density (range of 3 to 24

larvae/m2).

Gross et al. (1975) reported that prerelease exposure to larval

feces increased parasitization rates for M. croceipes. Lewis et al.

(1976) reported that the presence of Heliothis feces both retained M.

croceipes and caused increased parasitization of Heliothis larvae.

Drost et al. (1986) reported that female M. croceipes would fly upwind

in a wind tunnel to an odor source consisting of H. zea larvae feeding

on cowpea seedlings and that preflight exposure to larval feces enhanced

his response. Elzen et al. (1987) reported that M. croceipes was

attracted to leaf volatiles, larval frass and Heliothis larvae. Lewis

and Tumlinson (1988) found that associative learning is an important

aspect of M. croceipes' foraging behavior. They reported that M.

croceipes associates hexane extractable volatiles in larval feces

(conditioned stimulus) with a water extractable fraction of larval feces

(unconditioned stimulus). Lewis and Gross (1989) reported high field

parasitism rates (ie. 42-71%) for M. croceipes after prerelease exposure

to frass from artificial diet-fed Heliothis larvae.

Lewis and Jones (1971) found that a substance present in feces,

hemolymph and salivary secretions from H. zea, H. virescens and H.

subflexa elicited decreased orthokinesis and intense antennation by

female M. croceipes. Jones et al. (1971) subsequently identified 13-









methylhentriacontane as the most active compound. Lewis et al. (1988)

reported that the R and S stereoisomers of 13-methylhentriacontane did

not differ in their effects on M. croceipes. Sauls et al. (1979)

demonstrated that female M. croceipes responded (orthokinetic and

antennal behaviors) more strongly to frass of Heliothis larvae feeding

on cowpea seedlings than to the frass of larvae feeding on artificial

diet. Nordlund and Sauls (1981) reported that M. croceipes responded

(orthokinetic and antennal behaviors) most strongly to frass from

cotton-fed larvae, less strongly to frass from soybean-fed larvae, and

ast strongly to frass from either artificial diet-fed larvae or corn-

fed larvae. Teague and Phillips (1982) reported that Elcar, a

commercial formulation of Heliothis nuclear polyhedrosis virus

containing larval remnants, elicited stopping and antennation in M.

croceipes.

Although Mueller (1983) reported wasp rearing regime (ie. its

nosts' diet) did not affect subsequent wasp preference, Drost et al.

(1988) found that wasps reared on hosts fed cowpeas responded in a wind

tunnel more strongly to larvae feeding on cowpeas than wasps reared on

hosts fed artificial diet. Female M. croceipes exposed to both feces

from larvae fed cotton and feces from larvae fed beans parasitized more

larvae on cotton than on beans in a field cage containing larvae on both

plants, and those exposed to both feces from larvae fed bean and feces

from larvae fed tomato parasitized more larvae on beans than on tomato

(Mueller 1983). Drost et al. (1988) reported inexperienced females

preferred hyacinth bean complex over both cotton and cowpea complexes.






12

The antennae of M. croceipes, presumably which are the sense organs

used during host location (both long- and close-range) and host

acceptance, were found to possess trichoid, placoid, fluted and smooth

sensillae (Norton and Vinson 1974).

Oviposition behavior. The preoviposition period is short, usually

less than one day (Bryan et al. 1969a, Lewis and Snow 1971). Although

M. croceipes can attack and develop in all instars of Heliothis, except

late fifth instars (Lewis 1970b), M. croceipes exhibits instar/size

preferences. Bryan et al. (1969a) reports that M. croceipes prefers

early instars while Lewis (1970b), Hopper and King (1984a) and Hopper

(1986) report that third instars are preferred. The sizes (ie. lengths)

preferred by M. croceipes have been reported to be 6-12 mm, 10-12 mm,

and 11-20 mm (King et al. 1985, Quaintance and Brues 1905, Powell and

King 1984, respectively). These preferences may be a result of smaller

larvae being difficult to find (Lewis 1970b) and larger larvae

aggressively defending themselves and being able to dismember M.

croceipes with their mandibles (Hermann and Morrison 1980). Hopper

(1986) has also reported some fitness correlates for instar preferences.

When a female encounters a host larva, she raises her forelegs,

holds her antennae in a 'C-shape' ('ready-to-strike' pose), then in one

swift movement, springs forward onto the larva and oviposits through the

integument with a quick thrust of her ovipositor (Lewis 1970a). Females

do not probe with their ovipositors prior to oviposition (Hermann and

Morrison 1980). Tilden and Ferkovich (1988) reported that M. croceipes

could be induced to oviposit in agarose drops soaked with Heliothis

hemolymph. The active component was found to be heat labile, less than








12,000 daltons, non-hexane extractable, and not degraded by trypsin or

protease. Further progress towards the identification of the

oviposition stimulating kairomone is reported by Heath et al. (1989).

Fecundity and parasitoid eggs. Lewis and Snow (1971) found that

females lay an average of 300 eggs, with a range of 32-632. The eggs of

M. croceipes have been described and figured by Lewis (1970a) and Greany

(1986). Greany (1986) also reported that hemolymph components are

important for parasitoid egg development. Ferkovich and Dillard (1986)

reported that M. croceipes' eggs do not utilize host proteins but absorb

iree amino acids instead and synthesize proteins de novo. Tilden and

Ferkovich (1987) studied the regulation of protein synthesis during egg

development.

Vinson (1974) suggested that M. croceipes' eggs were not

encapsulated due to ionic properties conferred by the calyx fluid.

Stoltz et al. (1976) reported that virus-like particles in the

reproductive tract of M. croceipes may play a role in countering host

defense responses. Vinson (1977) found that these particles could

inhibit the encapsulation of eggs of Cardiochiles nigriceps Viereck

(Hymenoptera: Braconidae) in H. zea (a non-permissive host for C.

nigriceps) but did not prevent the encapsulation of the hatched larvae.

Sroka and Vinson (1978) reported that M. croceieps' eggs escape

encapsulation by inhibiting melanin formation, but not by inhibiting

phenoloxidase activity.

Vinson and Lewis (1973) reported that approximately 750 teratocytes

are released into the hemolymph of Heliothis when M. croceipes' eggs

hatch. The function of these teratocytes is still unclear; however,





14

they do not function as a direct source of nutrition nor do they inhibit

Heliothis' ability to encapsulate foreign materials (Vinson and Lewis

1973).

Parasitoid larvae. There are three larval instars in M. croceipes

and each has been described and figured (Lewis 1970a, Greany 1986).

Quaintance and Brues (1905) also provide a figure of a M. croceipes

larva. Microplitis croceipes larvae do not feed on the tissues of their

hosts, but obtain their nutrition instead from host hemolymph (Vinson

and Lewis 1973). Microplitis croceipes larvae possess an anal vesicle

i.ch has been described by Edson et al. (1977). Edson and Vinson

(1976, 1977) report that the anal vesicle functions in bicarbonate and

ammonia excretion, absorption of trehalose, glucose, salts and amino

acids, but not in respiration or osmoregulation.

Behavior and physiology of parasitized hosts. After parasitization

hy H. croceipes, Heliothis larvae continue to feed but at a much reduced

.'e, grow slower and become progressively sluggish (Quaintance and

Brues 1905, Bryan et al. 1969b, Jones and Lewis 1971, Dahlman and Vinson

1980, Webb and Dahlman 1985). Jones and Lewis (1971) reported that the

reduced growth rate and respiration rate of parasitized larvae was

caused by the calyx fluid injected during oviposition. Hopper and King

(1984b) report that parasitized larvae, in addition to feeding less,

move less often and damage fewer cotton squares and bolls than

unparasitized larvae. Lewis (1970b) reported that no matter what instar

was parasitized, subsequent host molts occurred on schedule and

progressed to the fifth instar, although head capsule size was reduced.

Powell (1988) reported that the earlier a Heliothis larva was







parasitized by M. croceipes, the less likely the host larva will reach

its fifth instar.

Parasitization of Heliothis affects the number and kinds of

proteins in the host's hemolymph (Barras et al. 1972). Glycogen levels

are depressed in parasitized larvae relative to unparasitized larvae

(Dahlman and Vinson 1980). Dahlman and Vinson (1975a, 1977) found

elevated trehalose levels in parasitized larvae and reported that the

calyx fluid was responsible for this effect.

Parasitized larvae acquire a characteristic pale yellow color by

the time the parasitoid larvae are full grown (Quaintance and Brues

1905, Winburn and Painter 1932). Parasitized Heliothis larvae form

pupal cells like unparasitized Heliothis larvae; however, they do not

pupate (Lewis 1970b, Jones and Lewis 1971, Webb and Dahlman 1985).

About eight days post-parasitization, full-grown M. croceipes larvae

bore out of the third to fifth abdominal pleuron, near the first pair of

prolegs, leaving a black scar on the host (Quaintance and Brues 1905,

Winburn and Painter 1932, Bryan et al. 1969a, 1969b, Lewis 1970a).

Although the host remains alive after parasitoid emergence, it does not

feed or develop further and eventually dies (Quaintance and Brues 1905,

Winbur and Painter 1932, Bryan et al. 1969a, Lewis 1970a). Webb and

Dahlman (1986) report that this is a result of M. croceipes inhibiting

prothoraciotropic hormone or ecdysone release.

Cocoon and diapause. The cocoon spun by M. croceipes larvae is

oval-elongate, 5-7 mm long, 3 mm in diameter, smooth but with several

coarse longitudinal ridges (Quaintance and Brues 1905, Winburn and

Painter 1932, Lewis 1970a). Cocoons are figured by Quaintance and Brues









(1905) and Lewis (1970a). The prepupae within the cocoon is the stage

which diapauses and overwinters (Bryan et al. 1969a). Diapause in M.

croceipes is induced by low temperatures: rearing temperatures of 15, 20

and 0C induced 100, 60 and 0% diapause, respectively, in M. croceipes

(Bryan et al. 1969a). Winburn and Painter (1905) reported that late in

the fall 98% entered diapause.

Developmental times. Eggs hatch approximately 36-48 hours after

oviposition (Lewis 1970a, Vinson and Lewis 1973). Mean development

times for egg plus larval and pupal stages for M. croceipes are shown

Table 1-3. The lengths of both egg plus larval and pupal stages tend to

decrease with increased temperature. However, at 35"C, pupal mortality

was 100%. In addition, developmental times are affected by host species

(Mueller 1983), instar parasitized (Jones and Lewis 1971, Hopper and

King 1984a, Powell 1988), and host diet (Mueller 1983).

Ecology

Intraspecific competition and host discrimination. Although two M.

croceipes have been observed to emerge from a single host larva, in 95%

of the cases of superparasitism only one parasitoid emerges from a ho

(Bryan et al. 1969a, Lewis and Burton 1970). Barras (1971) reported

that the first parasitoid to hatch has an advantage over parasitoids

hatching later except when the later larvae hatch more than 48 hours

after the first. Intraspecific host discrimination has been studied

previously in M. croceipes (Lewis and Snow 1971, Vinson and Guillot

1972), though both studies found that M. croceipes did not discriminate

between unparasitized and once-parasitized Heliothis larvae. In

addition, Lewis and Gross (1989) reported the distribution of M.










Table 1-3. Mean lengths of stages for Microplitis croceipes.


Rearing Length of Stage (days)

Temperature Reference

(C) Egg plus larval Pupal


9.0



(diapaused)



12.2



7.5



6.6



7.2



(died)



6.0


12.0


Quaintance and

Brues 1905

Bryan et al.

1969a

Bryan et al.

1969a

Bryan et al.

1969a

Bryan et al.

1969a

Bryan et al.

1969a

Bryan et al.

1969a

Lewis 1970a

Vinson and

Lewis 1973


34.5


16.0


10.4


32.2


26.6

27


8.0

9.0








croceipes eggs dissected from field-collected Heliothis did not reveal

any evidence of host discrimination. Barras (1971) however, reported

that M. croceipes avoids duplication of parasitization in the field.

Interspecific competition and host discrimination. When both C.

nigriceps and M. croceipes are present in a single Heliothis virescens

larva, C. nigriceps generally wins (Lewis and Brazzel 1968), however M.

croceipes did not discriminate between unparasitized H. virescens larvae

and H. virescens larvae parasitized by C. nigriceps. (Vinson and Guillot

1972).

Vinson and Ables (1980) reported that M. croceipes generally

outcompeted the egg-larval parasitoid Chelonus insularis Cresson

(Hymenoptera: Braconidae) and that M. croceipes did not discriminate

between larvae parasitized by C. insularis and unparasitized larvae.

Hyperparasitoids. Species within two families are known to be

hyperparasitic on M. croceipes: Perilampus sp. (Hymenoptera: Chalcidae)

(Quaintance and Brues 1905, Soteres et al. 1984), and Mesochorus sp.

(Hymenoptera: Ichneumonidae) (Quaintance and Brues 1905, Bottrell et al.

1968, Soteres et al. 1984).

Pathogen associations. Both the bacterium Serratia marcescens and

the fungus Nomuraea rilevi can depress M. croceipes survival (Bell et

al. 1974, King and Bell 1978). The microsporidium Vairimorpha did not

affect M. croceipes' survival, but did increase its developmental time

(Hamm et al. 1983). Under certain conditions, both M. croceieps and N.

rilevi could develop within a single host larva (King and Bell 1978).

Microplitis croceipes has been reported to transmit S. marcescens

horizontally (ie. within a generation: larva to larva) and Vairimorpha








horizontally but not vertically (ie. between generations: adult to

larva) (Bell et al. 1974, Hamm et al. 1983).

Pesticide interactions. Felton and Dahlman (1984) reported that

the fungicide Maneb was more toxic to M. croceipes than to its Heliothis

host. Horton et al. (1986) reported that the fungicides benomyl and

thiabendazole both reduced successful M. croceipes larval emergence.

Powell and Scott (1985) found that M. croceipes adults were relatively

tolerant to residues of flucythriante, fenvalerate, and thiodicarb.

Powell et al. (1986) reported that organophosphates were generally more

toxic to adult M. croceipes than pyrethroids. Bull et al. (1987) also

reported that adult M. croceipes were somewhat tolerant to pyrethroids.

Host plant resistance. Mueller (1983) reported that Heliothis

larvae feeding on cotton were more suitable hosts than those feeding on

either beans or tomatoes and suggested that these differences may be due

to plant defensive compounds (eg. alpha tomatine). Although McCutcheon

and Turnipseed (1981) did not find lower percent parasitism of Heliothis

by M. croceipes in resistant (eg. variety ED73-371) compared to

susceptible (eg. Bragg) varieties of soybeans, Powell and Lampert (1984)

reported that some resistant varieties (ie. PI227687 and 229358)

adversely affected M. croceipes survival.

Predator interactions. Stark and Hopper (1988) reported that

Chrvsoperla (=Chrvsopa) carnea (Stephens) (Neuroptera: Chrysopidae)

larvae did not discriminate between unparasitized Heliothis larvae and

those parasitized by M. croceipes.








Use of Kairomones by Entomophagous Insects

Host Selection Process.

The generally accepted sequence of successful parasitization has

been divided into five consecutive steps by Vinson (1976). These are 1)

host habitat finding, 2) host finding, 3) host acceptance, 4) host

suitability, and 5) host regulation. Lewis et al. (1976) called the

first two of these steps 'host location', while Vinson (1976) termed the

first three of these steps 'host selection'. Entomophagous insects can

use various stimuli during their host selection process. They can use

visual cues (eg. size, shape, color, texture and movement), acoustical

or vibrational cues, and chemical (both olfactory and gustatory) cues.

Chemicals used as such have been termed 'kairomones' (Gr. kairos,

opportunistic) (Brown et al. 1970, Whittaker and Feeny 1971, Nordlund

and Lewis 1976). Although the usefulness of this term has been debated

(Blum 1974, 1977, Weldon 1980, Pasteels 1982), the term is useful in

scribing instances in which chemicals mediate the exploitation of one

species by another.

Kairomone-Mediated Host Selection

Kairomones may act at any of the various levels of the host

selection process. Kairomone detection has been divided into long-range

and short-range chemoreception by Vinson (1976) and Weseloh (1981).

Long-range kairomones act at distances which preclude chemotaction (ie.

must be olfactory). The majority of long-range kairomones studied

involve attraction (ie. chemotaxis), however, it should be noted that

long-range kairomones could stimulate both search initiation (eg. flight

initiation) (Duelli 1980) and landing responses. Short-range






21

kairomones, on the other hand, act only after contact (ie. chemotaction)

or at very close range. Common responses to short-range kairomones

include orthokinetic, klinotactic and 'examination' behaviors (eg.

antennation and ovipositor probing) and actual oviposition (Hendry et

al. 1976).

To be successful, a parasitoid (or predator) must sequentially

narrow the range of possible habitats where it could place its eggs (or

find its prey) until a suitable habitat is located. The concept of

'behavioral active space', originally developed for pheromones (Bossert

and Wilson 1963), is also very useful in the consideration of kairomone-

mediated host selection. A behavioral active space is proportional to a

chemical's release rate (Q) divided by the behavioral threshold (K) (ie.

Q/K ratio). In order to sequentially narrow the possible habitats to

search, there must be ever-decreasing active spaces for the sequential

host selection behaviors and therefore decreasing Q/K ratios during the

host selection process. Relatively high Q/K ratios are expected for

behaviors exhibited early during host selection, such as flight

initiation and upwind flight, slightly lower Q/K ratios for behaviors

such as hovering and landing, low Q/K ratios for antennal examination

and ovipositor probing behaviors, and very low Q/K ratios for

stimulation of oviposition. Although one chemical could theoretically

elicit the complete host selection sequence from search initiation to

oviposition (by ever increasing behavioral thresholds for the sequential

behaviors), it is much more likely that different chemicals (with

different release rates and/or behavioral thresholds) elicit the various

levels of the host selection process.









Host and Prey Insects Producing Kairomones

It should be obvious that the function of such chemicals is not to

serve as kairomones, but that their use as such is an incidental

consequence. The actual function of these chemicals may be as

pheromones, allomones, hormones, glandular secretions, excretory or

metabolic wastes, cuticular constituents, egg adhesives, etc. It is not

surprising that over evolutionary time entomophages have been selected

to associate these chemicals with their hosts and prey. Kairomones have

been demonstrated to be produced by many insect orders including

fsoptera, Hemiptera, Coleoptera, Homoptera, Diptera, Hymenoptera, and

Lepidoptera.

Entomophagous Insects Using Kairomones

The use of kairomones by entomophagous insects has long been

recognized as evidenced by the early reports of parasitoids attracted to

their host's odors (Townsend 1908, Hase 1923, Stein-Beling 1934, Ullyett

1935, Laing 1937, Marsh 1937, Thorpe and Jones 1937, Bartlett 1941).

Evidence of kairomone utilization exists for most of the major insect

orders with entomophagous members including Hemiptera, Neuroptera,

Coleoptera, Diptera, and especially Hymenoptera.

Long Range Kairomones

Methods for studying long-range kairomones. Many different methods

have been used to study long-range kairomones. Field trapping studies

were used in all cases involving scolytid pheromones except Hansen

(1983) and Payne et al. (1984) who used electroantennograms and Mizell

et al. (1984) who used an olfactometer. Field trapping studies have

also been used to study long-range kairomones of entomophages of non-









scolytids (Mitchell and Mau 1971, Sternlicht 1973, Hagen et al. 1976,

Barbosa et al. 1978, Rice and Jones 1982, Aldrich et al. 1984). Field

studies which measured entomophage densities or parasitization rates,

but did not involve trapping, have also been used (Lairachi and Voegele

1975, Duelli 1980, Harris and Todd 1980, Lewis et al. 1982). In

addition, field observations (Marsh 1937, Tinbergen 1972) and

laboratory/greenhouse observations (Smits 1982) have been used. The

most common method, however, is the use of some sort of olfactometer,

especially 'Y-tube' olfactometers, though both 4-choice olfactometers

(Bouchard and Cloutier 1985, Noldus and Lenteren 1985) and wind tunnels

(Elzen et al. 1987) have also been used.

Long-range kairomones and entomophages of scolytids. Evidence

exists for the use of long-range kairomones by many entomophagous

insects, especially those attacking scolytids (Table 1-4). These

compounds all function as scolytid pheromones and most were fortuitously

discovered to serve as kairomones during field studies (Borden 1977,

Wood 1982). Most of these entomophages are predaceous beetles such as

clerids, trogositids or histerids, although a few are hymenopterous

irasitoids.

There are additional cases involving probable long-range kairomone

use by entomophages of scolytids, however, the results are confounded by

the inclusion of plant-derived compounds (ie. synomones) in the

attractant mixtures (Rice 1968, 1969, 1971, Williamson 1971, Rudinsky et

al. 1971, Lanier et al. 1972, Camors and Payne 1973, Pitman 1973, Kline

et al. 1974, Dyer et al. 1975, Bakke and Kvamme 1978, Moser and Brown












H H H H
co co co co


0 0 0 0
I 1 I Ir
OD



m t c t
cO c0


C C r C
4) (1) W(

l 4 (4 *H ( (
CQ O G 56 P C


-1
0




o *
4-) N
10 H



> H





1*4 *H
*<1 r-
> S


U n u)o).4i
3 011 H
Mc FA j3 a m a to
0 to 0 *0
tof 0 0 (0 to r-4 to to r-l
:3 tr to- Ul 0 t| V 0 ty1 v
44 0 0 0 0 0 .
0000
m0 m 0 p 41 D 41
c 0 IIV 0 0 0 0 0
C O 10 C D 0 0 Q Q O
0 > 0 0 > > 0 > > 9
M ll Ml 49 MIN Q l Ml Q|
o1 4)1 41 6 1 1
HI HI HI HI HI HI DI HI HI GI


44
u






0




-1

,MM
o0


















o
0
4 U


0 0
s a>


















0
V







* *

0


O i
a)
ar


0
O

0 1

04 H
*-I


r-1
o


0)


M-) ---

5.r4 l
4)


*H -


S.' UWV U 0 V
0 H *"4 H *4
t co $4 0
U 0 0 U '- 0
0 U-) U ".4 U) *



E 0 E 0 U
w M ) 0) U 0 U)

0 0 0 900 0 U

r-1 44 441 441 V
41 E4 EI E1 91 &; EI


*4
0
H
,-I
u

*Il
0)
4)
0



MIl


H-
0
)C a
04H
-H
















N *-M
ow oy

C 4)
C 0
04)C
V >



V0 V



> Aw
.94 *-


a
) U) C4


0 0 0 C
4) 4 4- *)4 I C

o 0 0 0 c a


U)



4)
4 0 0
C O 4o 4


*A J .0 C +


4d 6I 6I dI


S'0 ^- -
*i.l -'-4 '0 '0 '0
0 0 *4 *4 *H
'4 )O 0 0 '0 V
*-4 o4 (U U) r -H -4
k O O 0 0 k 0
0 0 0 c 0 .
H t ( 0 H a V




>1 > 03 C C 3r 0
'0 '0 > > 3 W0 0


H EI EIH


V0
& C
C *>
0
4 0
o 0

C 0 r- H
0 0 I .
a o N I
14 Ln A Ln 0
0D H m H X
3~~ 0 C


HI HI


*-M
4-1
C N
0 Iq
4 H
44 -


co

C >

0
U4 0
CO )

C 0
#4





C *

(0 4.
o Qd


0



0 w,
Ea co

e #





r-4
H


o
U) 0)


4 N
*-4 *-0
> z


I.. r- r- OD
H H H H


0 0 0 0
C C G C
C C C C
0) ) < 0
C CCC f


3U
4)


-4
0

*4
U0
4)
HP


a0


a 4
o o
H U

0 M


0
o



0 0r
r-4 .


l 01 d1 Ql d1


(l rA U;l rA


61 61 Ml m l










1978, Dixon and Payne 1979, Reeve et al. 1980, Billings and Cameron

1984).

Long-range kairomones and entomophages of non-scolytids. There are

also many reports of entomophages which do not attack scolytids using

long-range kairomones during their host selection process. The sources

of these chemicals vary widely and their functions are seldom known.

They have been demonstrated to be emitted from several insect stages

including larvae and nymphs (Thorpe and Jones 1937, Williams 1951,

Monteith 1955, 1958, Schmidt 1974, Schuster and Starks 1974, McO

and Pass 1977, Mossadegh 1980, Nettles 1980, Powell and Zhi-Li 1983,

Stafford et al. 1984, Obata 1986), pupae (Marsh 1937, Ullyett 1953,

Edwards 1954, Simmonds 1954, Wylie 1958, Carton 1971, 1974, Barbosa et

al. 1978, Sandland 1980), adults (Laing 1937, Akre and Rettenmyer 1966,

Read et al. 1970, Lewis et al. 1971, Richerson and DeLoach 1972,

Tinbergen 1972, Lairachi and Voegele 1975, Smits 1982, Nordlund et

1983, Bouchard and Cloutier 1985, Noldus and Lenteren 1985), but, as

yet, not from insect eggs.

Long-range kairomones have also been demonstrated to be emitted

from insect by-products such as frass (Hsaio et al. 1966, McKinney and

Pass 1977, Elzen et al. 1987), honeydew (Wilbert 1974, Hagen et al.

1976, Emden and Hagen 1976, Duelli 1980, Bouchard and Cloutier 1985),

and larval-damaged leaves (Nealis 1986). Like scolytids, both sex and

aggregation pheromones of non-scolytid hosts or prey also serve as long-

range kairomones (Mitchell and Mau 1971, Sternlicht 1973, Harris and

Todd 1980). Several pheromones of non-scolytids which also serve as

kairomones have been discovered (Table 1-5).









27 .




0
U0
-M
0

Ir
0
CV









O)
tU




0
*P









0
4)



.,

a)
0
a



















0
$
e















4.4
go
a)








$4


0
a


a4)
Il

VP-
C




O








.

4)

0


a)



'4-4
a)







>1
a0
S4

o

(0 R
0 r4
OH


a '
4 a0














O e
R) 0
o 0
0


0 *H
4 0








C 4







c *3
0
O o
u '


4 0

S0 C
(0 O 0 0 .


, 0 I
0 L 4 4 W
O n + ^ co
'0 % C 0 4. N
*H -- U 0 -


S0
I t
,C I
Q) 0
10 C
a o

(P H

I O>


H 0
> H



I U
r^ 0


4.)
I

I 0
* C
* 0



H o4
>11 0

>1




e i
0 i

C, 'O -"-
I 0 0


uO

r0*

C
0)


(0
a



'a
0
rz
c
Eo
CP


0



0 40








a

0
'O -. 3












0v *4-
a) 0)4 $4 C I I


4 $4
o a $4 4.
Sm Oa




U| o- clM










co





WI

(d 00






x(
k O

O



0 40
0 4 H
) .O H



O 4 H
Sgo- ,-4

U EU OH






N 0
S) 0)


Q CI





a C
























I I I I I
Sr
r-



1 0
























44 44 ") W r-"
H L O U
E -4 DW1 00







Q 00
O 'U 'U H U a)



r-U 4 H






to n ON H
IS-O



V 0





O j 1 V p
0 U N u 0 4U

0' 'U 0' 'U H'- 'U I I
I 4) I ) V ^ -
- a) -. a) H 4)'







29

Chemistry of long-range kairomones. Although all of the compounds

listed in Tables 1-4 and 1-5 are generally considered to be 'identified'

long-range kairomones, several were tested as part of a mixture and thus

the effects of the individual components relative to the blend remain

unknown. In addition, even though most scolytid pheromones exist as

stereoisomers and stereochemistry is very important to the pheromonal

response (Silverstein 1977), often the compounds tested were racemic or

the optical purities not reported. Notable exceptions to this include

Hansen (1983) and Payne et al. (1984). Hansen (1983) tested three sets

of enantiomer pairs and found that the stereoisomers of two of these

sets were equal in activity, and that both stereoisomers of the third

set were active, although they differed significantly. Payne et al.

(1984) reported that only one of the stereoisomers was kairomonally

active even though both enantiomers are produced by the prey species.

Aldrich et al. (1984) also tested enantiomers of the optically active

component of the soldier bug attractant and found that for several

entomophages only one enantiomer was active while for other entomophages

both were active even though the host produces only one stereoisomer.

It is interesting to note that all of the compounds listed in

Tables 1-4 and 1-5 have relatively low molecular weights (ie. between 98

and 254) and are therefore probably relatively volatile. In addition,

most function as sex or aggregation pheromones for the hosts, attesting

to their suitability for long-range chemical communication. Therefore,

these compounds probably have high Q/K ratios (for both host and

entomophage species) and are ideally suited to enable the entomophages






30

to 'eavesdrop' on the host's chemical communication system for their own

long-range host or prey location.

Short-Range Kairomones

Methods for studying short-range kairomones. The most common

method for studying short-range kairomones is a 'petri dish' assay.

Generally, test materials are placed on filter paper which is then

placed in the petri dish and the entomophage exposed to the sample and

its behavior recorded. Petri dish assays can be quite effective and

have been used to study various short-range kairomonal responses

including: orthokinetic, klinokinetic and klinotactic responses. Petri

dish bioassays have also been used to study antennal and abdominal

responses to kairomones as well as effects on parasitization rates

(Lewis et al. 1971).

Kairomonal stimulation of actual oviposition has been studied using

artificial wax eggs (Rajendram and Hagen 1976), parafilm tubes (Arthur

et al. 1969), agar-filled cuticles (Burks and Nettles 1978), and agar

drops (Tilden and Ferkovich 1988).

Host 'seeking' and 'excitement' versus kineses and taxes. Although

the most commonly reported responses to short-range kairomones are host

'seeking' and 'excitement', these terms are imprecise and should be

avoided. The behaviors exhibited could have been more precisely

described in terms of the kairomone's orthokinetic, klinokinetic, and

klinotactic effects on the responding insect (Kennedy 1977, 1978). Only

occasionally have these terms been used to describe kairomone-mediated

behavior. There are examples of direct orthokinesis (Mudd and Corbet

1982, Strand and Vinson 1982), inverse orthokinesis (Waage 1978, 1979,









Chiri and Legner 1982), direct klinokinesis (Edwards 1954, Strand and

Vinson 1982), inverse klinokinesis (Chiri and Legner 1982), and

klinotaxis (Edwards 1954, Waage 1978, 1979, Mudd and Corbet 1982, Strand

and Vinson 1982).

Waage (1978) suggests that these terms can be applied to many cases

in which they were not originally used. For example, there are cases

where direct orthokinesis (Mudd and Corbet 1982), inverse orthokinesis

(Laing 1937, Vinson and Lewis 1965, Leong and Oatman 1968, Corbet 1971,

Lewis and Jones 1971, Cardona and Oatman 1971, Hendry et al. 1973,

Wilson et al. 1974, Leonard et al. 1975, Prokopy and Webster 1978,

Vinson et al. 1978, Cederberg 1983), direct klinokinesis (Vinson and

Lewis 1965, Bragg 1974, Sandland 1980), and positive klinotaxis (Vinson

and Lewis 1965, Bragg 1974, Sandland 1980) can be inferred.

Antennal behaviors. Short-range kairomones very often elicit

antennal responses variously described as antennation (Laing 1937),

antennal examination (Williams 1951), antennal tapping (Weseloh and

Bartlett 1971), antennal palpation (Finlayson 1952), antennal drummii.

(Edwards 1954), antennal rubbing (Vinson and Lewis 1965), and antennal

searching (Sato 1979).

Abdominal behaviors. Short-range kairomones also elicit various

behaviors of the abdomen such as ovipositor probing (Williams 1951),

ovipositor jabbing or thrusting (Corbet 1971), ovipositor unsheathing

(Corbet 1971), ovipositor extension (Moran et al. 1969), oviposturing

(Odell and Goodwin 1984), and drilling attempts (Edwards 1954). Some of

the short-range kairomones responsible for affecting the kineses, taxes,









antennal and abdominal behaviors of entomophagous insects have been

isolated and identified (Table 1-6).

Oviposition and larviposition. Short-range kairomones have also

been demonstrated to elicit actual oviposition (Bartlett 1941, Ullyett

1953, Arthur et al. 1969, 1972, Hegdekar and Arthur 1973, House 1978,

Nettles et al. 1982, Wu and Qin 1982, Odell and Goodwin 1984, Tilden and

Ferkovich 1988) or larviposition (Hsaio et al. 1966, Nettles and Burks

1975, Burks and Nettles 1978, Roth et al. 1978, Odell and Goodwin 1984)

in some entomophages. For most endoparasitoids studied, oviposition is

mediated by hemolymph characteristics (Hendry et al. 1976). Some of

these oviposition stimulating kairomones have been identified (Table 1-

7).

Chemistry of short-range kairomones. Unlike some of the

identified' long-range kairomones in Tables 1-4 and 1-5, all of the

identified short-range kairomones listed in Tables 1-6 and 1-7 are

active as individual compounds. Like some of the long-range kairomones,

many short-range kairomones exist as stereoisomers. However, the

presence of some of these enantiomers in nature is unlikely, especially

those kairomones which are sugars, amino acids, or cholesterol

derivatives, which are generally only found naturally in either the d or

1 forms. For the other kairomones, prediction of the naturally

occurring enantiomer is not easy as for sugars or amino acids.

Recently, the enantiomers of 13-methylhentriacontane were tested and

both were found to be active (Lewis et al. 1988).

The short-range kairomones listed in Table 1-6 have a wide range of

molecular weights (112-652), although they tend to have higher molecular























4-)
0





-,"
4J
90









*C
R 0
urn







OV





I o
0 0



o4 0
. 0

C



.-l 4
w -
0 0
01
















C-I 4-C
u c
Z 0


m(












0
4C .

I0

o a





*H M
S4.3
H O









1 1
S0

nQ)






c 3
U-l rC


0)



U










I-,
0

4R1

44 >4










O
4J 44
o













4J
0





O-
0 0
! >9
H
o *l














C -

0
O H
0 0
o o
u s


0r



0r
43




Of
b-


rr
01






O
4.1



C
0' 0
r-i a


Sr-1 0 C
-. -- -- 0 O0



o l | o 0



4-l 4 0







So .-H O 0 00
O 04 O r p4'r
0r 0EC *a 0 0 0






4 0 > () U U u
a 5 a a \




to ) 0 =I o (o
E 1 Co



E| -- BI Q ^ (( ^ 03|- O '-M '" OQ| I


0 1 00
0 '.0 N
Hn %o r-





0 0 0

IH i rei
i i i


0
Ln



Ie


r-(
H


>o I
41N
04
p **





4-)
O 'O
0

0 *
! O
0 0











Mo (1 C r C o ot
0 o C00 o00 0 CO *0 0 0
co 4- i O L r- w
-- .,.- 4
O H .(d co D V 0 t)
co a G ,-4 r-4 c 1 e-I C o
* XI o 1 0 0 CO 0 0 CO > -
. 4 O M
0 0 4.) 4.) .4 r-4 4.) H.r -
r- a) a) a)
Si 0 r ( *0


0 a -, ) -. 4 0






Sco O l to U O I a
o t 4,) 0 t o o to t
c to V o ^ e a) Qc
7 3 3 a) 0 0 3 0 0 -74 a M
E- E-- 4 > r r.
*0 *0















II II
00
0I U U
0 r-I ( .
H #C 00 0 a
a) 14 1.


4 OO Io 0 4 I) O #*4) 4










- ) 3 o 0 v 0 C *iO)
0 >4 0 0 + 0 4 I








M *n




.H *Q U S O 43 0 O 0 00 0 0 v IX
> -H 4) 4,-) |3 0 r.4 4J t r.







m) w r w 1 .0 4w 0 t k 41) (0
o4 > o 9 )wto >4 9







U 0 U X I U 0 I
0| zI M E 1 4









S4)II
H n4
4.) I I


04 CD U 0 1 0) C- U4
w -4 0 X -,-I r-4 44 a0 1"
> r-4 a) u 0 ) +C4 Cv
0O 5i C a) 5k4 0 44 10
S> C >. 0 0 ) IV I I
4t4 U U .! X )U Q)
UU q W C4 ro



C C



a) CJ in 3 Xi
C H '- 4. i4.
(0 M U Hn j ))t
M 44 N- 0 (0 4)4 I I

** r-4 0 a) X *- rI 4M -0 II II
Cl H Al a) QU U >1 0 X f-4C'





3 5i l l a C 3 a) 0) 0 Hd H
IA 4-I i H '- IA -I .Q 4. U 0n 1











0
Ot
0


0
4)





0
4.)

0
0

I





$4
o
0










.*H

.r.

E-4
0





0
*s
.9-4


0





*r4
I














.d
-43







0
*i




U)









(4
a,











H



,I.r
Ji


















aC
+Q)





0cc

C^-


0)
u
4)










44






Sw
0
0 0









S4-4
0










0'
0p -
.C >1
1 H

C 0

1 4)
















0-)


H ir


aC C l c
*-4l .-I U H
Q) P 0a) U
U) r 0r
W (d H


0
N
r r-



U U
0' C


0


M
43
00






- 9
P~ $4


0
$4
4) '
4 C
0 r


0a) *
r-a 4) 0\


V V
*r4 .ia
4-)
rrl -r
V0 *V
0 5 H 0

+ 0 5o


1 9
4) 4)
E-'I E-I


0
C


r 1
0



0) a)


e 5





o
1 f
*4I 0





i-il ^-







weights than the long-range kairomones and are therefore probably less

volatile. This is not surprising assuming that the behaviors they

mediate are only adaptive when exhibited in close proximity (ie. small

active space) to the host or prey. Therefore, one would expect a

relatively low Q/K ratio resulting from a low release rate (Q) and/or a

high behavioral threshold (K).

The molecular weights of the ovipositional stimulants are

relatively low (75-174) compared to the other identified kairomones

discussed earlier. All of these compounds exist as ions in an aqueous

solution and can only be sensed by contacting this solution (eg.

gustation via the ovipositor). Therefore the active space for these

kairomones is restricted to the hemoceol of the host and the final host

selection behavior is exhibited only in a very restricted subset of the

entomophage's environment.

Miscellaneous Responses of Entomophages to Kairomones

Other responses to kairomones include trail laying by ants after

encountering termite produced kairomones (Longhurst and House 1978),

increased longevity (Nordlund et al. 1976, Buleza 1979), increased

fecundity (Nordlund et al. 1976, Buleza 1979), sex ratio changes

(Sandland 1979), the following of ant trails by socially parasitic

staphylinids (Akre and Rettenmyer 1966), and the attraction of social

parasites to ant colonies (H11dobler 1969) or bumblebee nests

(Cederberg 1983). Chemotaxis in an aquatic environment has also been

demonstrated for a predator of mosquito larvae (Barber and Hirsch 1984).








Sexual Differences in Kairomonal Responses

The sex of entomophages attracted to long-range kairomones is only

rarely reported or studied. Therefore, any possible sexual differences

in responses to kairomones remain largely unknown. For entomophages of

scolytids, when sexual effects are noted, both sexes generally respond

(Wood et al. 1968, Bedard et al. 1969, Pitman and Vite 1970, Hansen

1983, Billings and Cameron 1984, Chatelain and Schenk 1984, Payne et

al. 1984), however not always in equal numbers (Vite and Williamson

1970).

For entomophages of non-scolytids, in most cases only females were

tested or those responding were not sexed. In those cases where the

response of males as well as females was tested, in many cases both

sexes were found to respond to the kairomones (Marsh 1937, Mitchell and

Mau 1971, Hagen et al. 1976, Emden and Hagen 1976, Barbosa et al. 1978,

Duelli 1980, Harris and Todd 1980, Aldrich et al. 1984, Obata 1986). In

a few cases females but not males were found to respond to the

lairomones (Tucker and Leonard 1977, Nettles 1982, Powell and Zhi-Li

1983, Bouchard and Cloutier 1985, Elzen et al. 1987). In addition,

Wilbert (1974) found unsexed larval cecidomyiids were attracted to their

aphid prey.

Host and Prey Location versus Mate and Food Location

The use of kairomones by female parasitoids and females of species

with predaceous larvae is obviously adaptive. Similarly, the use of

kairomones by both sexes of species predaceous as adults (Wood et al.

1968) or social parasites (Akre and Rettenmyer 1966, H11dobler 1969) is

not surprising. However, the explanation for the use of kairomones by









male parasitoids (Marsh 1937, Mitchell and Mau 1971, Barbosa et al.

1978, Harris and Todd 1980, Aldrich et al. 1984) is less obvious, though

males may use these chemicals to locate areas to serve as mating

locations (Thornhill and Alcock 1983). On the other hand, the response

of both males and females to these 'kairomones' may actually be a food-

seeking response rather than a host-seeking response. This may be

especially true for entomophages using honeydew as a 'kairomone' (Hagerx

et al. 1976, Emden and Hagen 1976, Duelli 1980), because honeydew has

been demonstrated to serve as food for entomophages (Singh and Sinha

1982, Elliot et al. 1987). Drost et al. (1988) reported no difference

in the flight response of female parasitoids to uninfested flowers and

infested leaves. In addition, the antennation response of M. croceipes

males and females towards honey is indistinguishable from the

antennation response of females towards larval feces (personal

observation).

Uses for Kairomones

There are many proposed uses for kairomones, though most relate t"

increasing rates of parasitization or predation (Lewis et al. 1972).

This increase may be accomplished by 'prestimulating' parasitoids at the

time of release to overcome their escape response (Jones et al. 1973) or

by 'retaining' (Jones et al. 1971) or concentrating (Hagen et al. 1970)

parasitoids or predators in a given area. There has been sc. e success

in increasing parasitization rates or predation rates using kairomones

(Lewis et al. 1972, 1975a, 1976, 1977, Jones et al. 1976).

Kairomones may be useful in monitoring entomophage densities in the

field (Sternlicht 1973). They may also give insights into phylogenetic








relations (Sternlicht 1973) or in the suppression of entomophages

attacking beneficial insects (Aldrich et al. 1984). Gross et al. (1981)

suggests that kairomones may be of some use in the screening of

potential imported natural enemies. Kairomones could also be of great

utility in the mass rearing of parasitoids on factious hosts (Finlayson

1952) or on artificial media (Arthur et al. 1969).

Considerations for the Study and Use of Kairomones

During the study of kairomones, possible effects of both host

species and the host's diet should be considered. The species of host a

parasitoid was reared on is postulated to affect its subsequent host

selection behavior (Walsh 1864) and host diet has been demonstrated to

affect kairomonal responses (Roth et al. 1978, Sauls et al. 1979,

Nordlund and Sauls 1981). In addition, learning has been demonstrated

to be very important during a parasitoid's host selection process.

Vinson et al. (1976, 1977) identified methyl paraben (an antifungal

agent added to artificial diet) as a kairomone from hosts reared on

artificial diet. Lewis and Tumlinson (1988) reported that a parasitoid

will fly to volatile odors encountered in combination with a host-

recognition factor.

The field use of kairomones, especially pest monitoring programs

employing pest pheromones which serve as kairomones, may have

detrimental effects on the entomophages using these chemicals (Pitman

and Vite 1971, Dyer 1973). Similarly, pest management programs using

pheromones to 'trap out' or to disrupt mating may also have detrimental

effects. Kennedy (1984) suggested adjusting the timing of trapping to

avoid parasitoid catches or using a bait which reduces parasitoid









catches. Additionally, Elcar, the commercial nuclear polyhedrosis

virus of Heliothis has been shown to have kairomonal effects on a

parasitoid of Heliothis (Teague and Phillips 1982). The use of this

unpurified virus could adversely affect parasitoid performance, though

the purified virus has no affect on parasitoid behavior.

Research Aims

The cotton bollworm, Heliothis zea (Boddie) and the tobacco

budworm, H. virescens (F.) (Lepidoptera: Noctuidae), probably are the

most important pests of cotton in the United States. Because of

problems associated with the use of pesticides to control these pests

(eg. resistance, environmental contamination, and cost), emphasis has

tfted towards biological control (Sparks 1981). The parasitoid,

Microplitis croceipes (Cresson) (Hymenoptera: Braconidae), is the most

commonly reported parasitoid of Heliothis and is believed to offer great

potential for the biological control of Heliothis species (Knipling and

Stadelbacher 1983). Although understanding the foraging behavior of a

parasitoid is extremely important to its use in an integrated pest

management program (Knipling 1979), relatively few studies have focused

on the foraging behavior of M. croceipes.

The purpose of this research was to examine several aspects of M.

croceipes' foraging behavior to facilitate its incorporation into an

integrated pest management program against Heliothis species. In order

to accomplish this, the following studies were undertaken: 1)

olfactometric studies of host-location by M. croceipes; 2) source of

volatiles mediating host-location by M. croceipes; 3) intraspecific

competition in M. croceipes; 4) parasitoid altered host behavior and







41

passive host discrimination in M. croceipes; 5) isolation and

identification of host-location kairomones used by M. croceipes; 6)

factors affecting the oviposition behavior of M. croceipes towards an

artificial substrate; 7) interactions between M. croceipes and a nuclear

polyhedrosis virus of Heliothis; and 8) effect of host diet and

preflight experience on the flight response of M. croceipes.














CHAPTER II
OLFACTOMETRIC STUDIES OF HOST-LOCATION BY
Microplitis croceipes (CRESSON)


Introduction

As indicated in Chapter I, only a few long-range chemical cues

used by parasitoids during their host-location process have been

identified. Although 13-methylhentriacontane has been isolated from

Heliothis zea (Boddie) (Lepidoptera: Noctuidae) and shown to elicit the

short-range behaviors of antennation and ovipositor probing in the

parasitoid Microplitis croceipes (Cresson) (Jones et al. 1971), n<

long-range chemical cues have been identified for female M. croceipes.

There is evidence that M. croceipes does use olfactory cues during its

host-location process; Drost et al. (1986) reported on the response of

M. croceipes in a wind tunnel to olfactory cues. This chapter describes

olfactometric studies of the response of M. croceipes to olfactory cues

and factors that might affect its host-location behavior.



Materials and Methods

Parasitoids. Microplitis croceipes, reared on H. zea, were

obtained as pupae from the Insect Biology and Population Management

Research Laboratory (Tifton, GA). The pupae were held for emergence in

an environmental chamber maintained at 260C, a relative humidity of ca.







43

70% and a photocycle of 15L:9D. Male and female parasitoids were caged

together to allow mating and were provided with water and honey.

Host Insects. Heliothis zea eggs were obtained from the Insect

Attractants, Behavior, and Basic Biology Research Laboratory

(Gainesville, FL). After hatching, larvae were fed pink-eye,

purple-hull cowpea [Vigna unguiculata (L.)] seedlings.

Olfactometer. The response of M. croceipes to volatile chemicals

was examined in a four-choice olfactometer (Pettersson, 1970) with

dimensions as described by Vet et al. (1981) (Fig. 2-1). Differences

are as follows: internal chamber height was 16 mm, inlet port inside

diam was 9.5 mm, insect trap, odor chamber, and humidifying vials were

of a slightly different design (Fig. 2-2). In addition, illumination

was provided by a 20-watt circular fluorescent lamp 20 cm above the

olfactometer. The complete olfactometer system was housed in a room

maintained at 27-28*C.

A vacuum pump, placed outside the room, drew air through the

olfactometer at a rate of 1200 ml/min (300 ml/min through each of the

four quadrants). Odor fields were visualized using ventilation smoke

tubes (Mine Safety Appliances Co., Pittsburgh, PA). The pressure inside

the olfactometer was never more than 4 cm of water (ca. 2.9 mm Hg) below

atmospheric pressure when measured with an open-end, water-filled

manometer connected to the olfactometer at the center extractor tube.

General Experimental Procedure. All experiments were conducted 3

to 6 hr after parasitoids had experienced "lights-on." Parasitoids were

tested singly by introducing them through the disconnected extractor

tube. The relative time spent in each quadrant and the time of final






















































Figure 2-1. Perspective view of four-choice olfactometer.
































B C


Figure 2-2. Schematic
4-choice olfactometer.
(C) Odor chamber. (D)


diagram of air inlet system for each arm of
(A) Olfactometer. (B) Parasitoid trap vial.
Humidifying vial. (E) Flowmeter.







46

choice were recorded with an Epson HX-20 portable computer with optional

expansion unit. A final choice was defined as the entry of a parasitoid

into an insect trap vial. Time of final choice was defined as time

between introduction into the olfactometer and entrance into an insect

trap vial. Parasitoids were given a maximum of 10 min to make a final

choice. The percent time per quadrant was calculated based on total

time spent in each of the four quadrants before a final choice was made.

Each insect was tested only once and after ca. 10 to 20 insects were

tested, the olfactometer, trap vials, and odor chamber vials were

dismantled, rinsed with absolute ethanol and washed in hot detergent.

System Bias. The response of inexperienced female M. croceipes in

the olfactometer when no odor source was present was examined to

determine if females exhibited preferences toward any quadrant(s).

Twenty-eight 3-day-old female M. croceipes were tested to humidified air

only (controls) in all four quadrants.

Effect of Oviposition Experience and Odor Concentration.

Three-day-old females were used to study the effects of oviposition

experience and odor concentration on female response. Inexperienced

females had no oviposition experience nor any previous exposure to

larvae, frass, or foliage. Experienced females had searched larval

damaged foliage with frass present and had oviposited once in an H. zea

larva just prior to being tested. The odor source consisted of a third

instar H. zea larva feeding on a 4-day-old (ca. 6-cm-tall) cowpea

seedling with the accompanying foliage damage and frass. This odor

source was placed in the odor chamber (Fig. 2C) and was termed one

plant-host complex equivalent (PHCE). Four concentrations of PHCE odor









were tested (1.00, 0.75, 0.50, and 0.25 PHCE). PHCEs of less than one

were produced by splitting the odor flow and adding back humidified air

to regain the original 300 ml/min flow rate. The odor was introduced

into only one quadrant, the other three quadrants contained humidified

air only (controls). Each replicate consisted of an inexperienced and

an experienced female (random order) being tested to each of the four

odor concentrations (in order of increasing concentration) in each of

the four quadrants (in random order) (32 females per replication).

Seven such replications were made.

Male Response. The response of male M. croceipes to one PHCE was

examined using 3-day-old males with no previous exposure to larvae,

frass, or foliage. Seven males were tested to the odor source at each

of the four quadrants (in random order) for a total of 28 males.

Effect of Female Age. The effects of age on female response were

studied using 0.75 PHCE as the odor source. Four age classes were used:

1, 3-4, 6-7, and 10-12 days post-adult emergence. Each replicate

consisted of an inexperienced female in each of the four age classes (in

random order) being tested at each of the four quadrants (in random

order) (16 females per replication). Seven such replications were made.

Inexperienced females were used in this and the following experiments to

investigate the innate response of females.

Source of Attraction. To determine the origin of attractive

volatile chemicals, three possible sources were compared to the PHC for

activity. These were damaged leaves, H. zea larvae, and frass. Damaged

leaf odor was produced by using a wire to make four 2-cm long scratches

on a 4-day-old cowpea seedling just prior to testing. Larval odor was









from a third instar H. zea larva which had fed on cowpeas for 2 days

preceding testing. The larva was placed in the odor source chamber

without foliage or frass. Frass odor was obtained from ca. 10 fresh

(less than 1-hr-old) pellets of frass from third instar H. zea larvae

feeding on cowpea seedlings. Each replicate consisted of an

inexperienced female being tested to each of the four treatments (in

random order) in each of the four quadrants (16 females per replicate).

Seven such replicates were made.

Collection and Assay of Attractive Volatiles. Volatiles were

collected from the plant-host complex to determine if volatile

attractants could be collected and retain their bioactivity. Volatiles

were collected from third instar H. zea larvae actively feeding on

cowpea seedlings. This odor source was contained in a glass chamber

(5.0 cm i.d. X 20.0 cm long). The chamber consisted of two halves

connected by a 50/50 ground-glass joint. The male half held a

coarse-glass frit to provide laminar flow through the chamber. Both

halves had a 24/40 ground-glass joint opposite the 50/50 joint, to

connect the chamber to the rest of the volatile collection system.

Humidified and prefiltered (activated charcoal) air was blown at a rate

of 300 ml/min over 3-5 PHCE for 2-4 hr. Plant-host-complex volatiles

were collected on activated charcoal (ca. 1.5 mm thick and 4 mm diam)

and subsequently extracted with 3-20 pl volumes of methylene chloride

and 2-20 Ml volumes of pentane. The methylene chloride/pentane extract

was then tested at the following concentrations: 0.1, 0.5, 1.0, 5.0, and

10.0 plant-host complex hour equivalents (PHCHE). Fifty pl of each

extract were placed on filter paper (4.25 cm diam), which was then










placed in the odor chamber after the solvent had dried. One replicate

consisted of an inexperienced female being tested to each of the five

concentrations (in order of increasing concentration) in each of the

four quadrants (random order) (20 females per replication). Eight such

replicates were made.

Statistical Analyses. Friedman Rank Sums (FRS) (Conover 1980),

based on percent times per quadrant, were used to test for quadrant

preferences when all four odor fields contained humid air only

(controls). In all other experiments, each treatment was tested against

three controls and preference for the treatment odor field over control

odor fields was tested using FRS. To compare percent times per odor

field between treatments, Duncan's New Multiple Range Test (DNMRT)

(Steel and Torrie 1960) was used after analysis of variance (ANOVA) of

percent transformed by angular transformation. Percent making final

choices were compared with DNMRT after ANOVA of percent making final

choices after angular transformation. For all dose-response data, the

concentration sum of squares were divided into polynomial components and

the reported equations represent the polynomials in X using significant

components. Times of final choice were compared by DNMRT after ANOVA.

Significance levels were 0.05 in all tests.



Results

General Behavioral Observations. After being released into the

olfactometer, the parasitoids were observed to walk on the bottom, top

and sides of the olfactometer, occasionally stopping to groom their

antennae, legs, wings, or abdomen. Parasitoids also made occasional









flight attempts, defecated, and stood motionless. When female

parasitoids crossed from a control field into a sample field, they

frequently began more intensive antennation and also walked more slowly,

though these behaviors were not quantified. Females also were observed

to follow the border between sample and control fields and to make

klinotactic turns after leaving a sample field.

Bias. Female parasitoids spent an average of 29, 28, 16, and 27%

of their time in quadrants 1, 2, 3, and 4, respectively. There was no

significant preference for any quadrant (P>0.25), and no female made a

final choice to humid air only.

Experience and Concentration. The effects of oviposition

experience and odor concentration on female response are shown in Table

2-1. Both inexperienced and experienced females spent significantly

more time in the test odor field than control odor fields at all odor

-centrations. Experienced females spent a greater percent of their

time in the odor field than inexperienced females at all concentrations

tested (P<< 0.005). Although mean percent time spent in the odor field

increased with odor concentration for both types of females, ANOVA

indicated a significant linear effect of PHC odor concentration (X) on

percent time in the test odor field (Y) only for experienced females: (Y

- 30.4X + 57.0; n = 112; r 0.37; P <0.01).

There were significant effects of both experience and

concentration on percent making final choice. Experienced females were

more likely to make final choices than inexperienced females and there

were significant linear effects of odor concentration (X) on percent

making final choice (Y) for both inexperienced females: (Y = 68.6X +









Response of inexperienced and experienced female


Microplitis croceipes to four concentrations of
odors .


plant-host complex


Odor Mean % Mean % making Mean time

Type of cone. time in final of final
b
female (PHCE) odor field choices choice(sec)



Inexperienced

0.25 54.8 46.4 233.8

0.50 60.2 57.1 195.2

0.75 64.8 78.6 146.5

1.00 67.4 96.4 155.9



Experienced

0.25 64.8 64.3 258.0

0.50 75.4 82.1 211.2

0.75 81.7 100.0 117.0

1.00 84.8 100.0 133.6


aTwenty-eight females
combination.


were tested at each experience-concentration


Both types of females spent significantly (P <0.005) more time in the
treatment odor fields than control odor fields at all concentrations
tested using Friedman rank sums.


Table 2-1.








52

26.8; n 28; r 0.69; P <0.01) and for experienced females: (Y 50.OX

+ 55.4; n 28; r 0.62; P <0.01).

The ANOVA of times of final choice revealed a significant effect

only for odor concentration, but no significant effect of experience on

time of final choice (P <0.25). ANOVA revealed a significant decrease

in time of final choice (Y) with increased odor concentration (X) for

both inexperienced females: (Y -93.1X + 238.0; n = 78, r = 0.23;

P <0.05) and experienced females: (Y -181.3X + 291.8; n 97; r -

0.36; P <0.01).

Males. Male parasitoids did not spend significantly more time in

the plant-host complex odor field than control odor fields (P >0.25),

nor did they make any final choices to this odor source.

Age. All age classes tested spent significantly more time (P

<0.005) in the treatment odor fields than in control odor fields,

however, there were no differences in percent time in odor field between

any age classes (P>0.1). The number of females (N 28) making final

choices were 20, 20, 19, and 17 for the 1, 3-4, 6-7, and 10-12 day-old

parasitoids, respectively, and were not significantly different (P

>0.25). There also was no significant effect of age on time of final

choice (P >0.25).

Individual Components. The responses of inexperienced female

parasitoids to individual components of the plant-host complex are shown

in Table 2-2. Female parasitoids spent significantly more time in the

odor fields that contained damaged-leaf odor and frass odor than in

control odor fields. However, females spent significantly more time in

odor fields containing plant-host complex odor than in odor fields that









Table 2-2. Response of inexperienced female Microplitis croceipes to
individual components of the plant-host complex


Mean % Percent Mean Time

Odor Time in Making Final of Final

Source Odor Fieldbc Choicec Choice

(secs)c



Plant-Host

Complex 68.7 **,c 82.1 a 143.5 a



Artificially

Damaged Leaves 50.7 **,b 32.1 b 160.4 a



Frass 46.9 **,b 32.1 b 161.5 a



!! zea larvae 34.5 NS,a 3.6 c 287.5 a



aTwenty-eight females were tested to each odor source.
b** Denotes females spent significantly (P<0.005) more time in treatment
odor fields than in control odor fields, NS denotes no significant
difference (P-0.05) from control (P calculated using Friedman rank
sums).
CColumn means with letters in common do not differ significantly
(P=0.05) using Duncan's New Multiple Range Test.







54

contained damaged-leaf or frass odor. Female parasitoids did not spend

more time in the odor field containing H. zea larval odor than in

control odor fields. Female parasitoids also made significantly more

final choices to damaged leaves and frass than to H. zea larva, but

fewer than to the plant-host complex. The ANOVA on the times of final

choice did not reveal any significant effect of odor source (P >0.25).

Collected Volatiles. The responses of inexperienced females to

the collected plant-host complex volatiles are shown in Table 2-3.

Female parasitoids neither spent significantly more time in the odor

fields that contained the two lowest concentrations (0.1 and 0.5 PHCHE)

than in control odor fields, nor did they make any final choices to

these concentrations. Female parasitoids did, however, spend more time

in the odor fields that contained 1.0, 5.0, and 10.0 PHCHE than in

control odor fields. ANOVA revealed significant linear and quadratic

components of collected volatile concentration (X) on percent time in

the odor field (Y): (Y 21.7 + 16.6X 1.2X2 ; n 160; r 0.51; P

<0.01). ANOVA also revealed significant linear and quadratic components

of collected volatile concentration (X) on percent making final choice

(Y): (Y -3.2 + 14.5X 0.9X2; n 80; r 0.73; P <0.01). The ANOVA

on the times of final choice did not reveal any significant effect of

concentration (P >0.1).



Discussion

The results demonstrate that female, but not male, M. croceipes

respond in a dose-dependent fashion to volatiles from the H. zea

larva/cowpea plant-host complex. Although females without any









Table 2-3. Response of inexperienced female Microplitis croceipes to
five concentrations of collected plant-host complex volatiles placed on
filter papera


Concentration of Mean % Mean Time

Collected Mean % time making final of Final

Volatiles (PHCHE) odor field choices Choice (secs)c


29.0 NS


33.4 NS


47.7 *


71.7 **


10.0


67.2 **


0.0



0.0


12.5



46.9



53.1


177.3



192.5



187.4


aThirty-two females were tested at each concentration.
b* and ** denote significantly (P=0.05 and P-0.005, respectively) more
time spent in treatment odor fields than control odor fields, NS
denotes no significant (P-0.05) difference from controls (P calculated
using Friedman Rank Sums).
CThere were no significant differences between any concentrations in
time of final choice (F-test P-value >0.1).









experience responded (i.e. innately), there was a clear effect of the

oviposition experience on female response. It is not clear whether this

is a case of associative learning (eg. olfactory cues with oviposition)

or a case of a change in state of motivation, although learning appears

to be an important part of host-selection by parasitoids (Arthur 1981).

The effect of "oviposition experience" may not be the oviposition per se

but rather the exposure to damaged leaves, frass, and/or salivary

secretions, etc. In fact, Drost et al. (1986) demonstrated that the

wind tunnel response of female M. croceipes exposed only to frass was no

different than that of females exposed to the plant-host complex

including oviposition. In addition, Lewis and Tumlinson (1988) showed

that females were attracted to novel odors (eg. vanilla extract)

encountered in association with a water soluble component of larval

feces. This type of experience might be used to increase the efficiency

of parasitoids released into the field (Gross et al. 1975).

Age of inexperienced female parasitoids had no apparent effect on

their response in the olfactometer. Drost et al. (1986) reported

similar results for experienced females in wind tunnel studies.

The results of the individual component experiments suggest that

no single component elicits a response equal to the PHC, but that some

combination, acting either additively or synergistically, is required.

Preliminary olfactometer experiments suggest that the sum of all

components may not give the complete activity either, and that some

interaction such as larval feeding or salivary secretions on the leaf

tissues might be required. The source of the volatile attractants will

be discussed in Chapter III.







57

Although female parasitoids spent approximately an equal amount of

time in the odor field that contained either the plant-host complex odor

or the collected volatiles, fewer final choices were made to the

collected volatiles than to the plant-host complex. This is probably

due to the extreme volatility of the collected volatiles, which were

never active for more than 10 min, and possibly to differences in the

ratios of chemicals released by the plant-host complex and the collected

volatiles.

In conclusion, these data clearly indicate that M. croceipes

females respond to olfactory cues from the plant-host complex. The fact

that females are attracted to artificially damaged leaves and to frass

suggests that M. croceipes may use volatile chemicals from these source

as synomones and kairomones (as defined by Nordlund and Lewis, 1976),

respectively during its host-location process. In addition, this

olfactometer could be used to identify volatile attractants for M.

croceipes and these chemicals might be used to improve the effectiveness

of M. croceipes in the field.














CHAPTER III
SOURCE OF VOLATILES MEDIATING THE HOST-LOCATION FLIGHT
BEHAVIOR OF Microplitis croceipes (CRESSON)


Introduction

Although much is known about the source and chemicals that mediate

close-range host-location behaviors (i.e. antennation and decreased

locomotion) in M. croceipes (Lewis and Jones 1971, Jones et al. 1971,

Sauls et al. 1979, Nordlund and Sauls 1981), little is known of the

factors that mediate long-range host-location (i.e. flight initiation,

anemotaxis, and landing). Chapter II describes the use of an

olfactometer to identify some sources as attractants for M. croceipes.

However, the source of volatiles mediating the flight behavior of M.

croceipes toward H. zea feeding on cowpeas, reported by Drost et al.

(1986), are unknown. The purpose of this research was to identify the

sources) of the volatile attractants that mediate the host-location

flight behavior of M. croceipes toward H. zea feeding on cowpeas

reported by Drost et al. (1986).

Materials and Methods

Parasitoids and Host Insects. Microplitis croceipes and Heliothis

zea were obtained and handled as described in Chapter II.

Flight Tunnel. The flight behavior of M. croceipes to volatile

chemicals was examined in a variable windspeed flight tunnel 60 cm X 60

cm in cross-section and 2.4 m long. The sides were made of 6-mm









plexiglass and the top and bottom were made of 3-mm plexiglass. Two

sheets of nylon mosquito netting (ca. 10-cm apart) were used to provide

laminar flow and contain the insects. One sheet of nylon screen (7 x 7

mesh/cm2) was used on the downwind end. A pattern of black and white

stripes under the tunnel was used to provide a visual reference for

flying insects. Air was drawn through the tunnel using a 30-cm diam

blower (Model 2C939A, Dayton Electric Mfg. Co., Chicago, IL) and a 3/4

HP motor (model 6K9498, Dayton Electric Mfg. Co., Chicago, IL).

Windspeed was measured with a hot-wire anemometer (TSI model 1610-12,

St. Paul, MN). Windspeed could be varied from 0.2 m/s to 0.4 m/s using

a damper and was set at 0.25 m/s for all experiments. Air was exhausted

outside the building after being drawn through the tunnel. The tunnel

had two doors on each side allowing access to the inside, and lighting

was provided by four overhead incandescent lights (54-W) (ca. 2400 lux).

The tunnel was housed in a room maintained at 27.5-290C and a relative

humidity of 55-80%.

Odor Inlet System. Test odors were introduced into the flight

tunnel via the odor inlet system. This system consisted of pressurized

air filtered through activated charcoal and metered to 300 ml/min with a

flowmeter (Brooks R-2-15-D tube with carboloy float, Emerson Electric

Co., Hatfield, PA). The air was then humidified (ca. 100% RH) and blown

through a glass odor chamber, which held the test material. The odor

chamber (5-cm ID 20-cm long) consisted of two halves connected by a

50/50 ground-glass joint. The male half had a coarse-glass frit to

provide laminar flow. Both halves had a 24/40 ground-glass joint,

opposite the 50/50 joint, to connect the chamber to the rest of the odor







60

inlet system. The air was then introduced into the tunnel via a 6-mm OD

(4-mm ID) glass tube through a hole in the tunnel floor. A brass elbow

with Teflon ferrules was used to connect the glass tube to a nozzle

(6-mm OD, 4-mm ID drawn to 1-mm ID tip). The nozzle tip was 30-cm from

the tunnel floor. An artificial leaf (6.2 cm2), made of green

construction paper, was attached below the nozzle to serve as a landing

place.

General Experimental Procedure. Test insects ranged from 3 to 10

days of age, however, on any given day the range was only 2 days. All

experiments were conducted 3 to 7 h after onset of photophase. Females

were allowed to search a larval-damaged pea seedling with accompanying

feces and allowed to oviposit in a third instar H. zea 1 min before

being tested. Males were tested without prior exposure to leaves,

larvae, or feces. Parasitoids were tested individually by transferring

them into the tunnel via a 5-dram vial. The vial was placed in a wire

stand, open end up, 30 cm from the tunnel floor and was in the odor

,lume downwind 1.3 m from the odor nozzle.

Behavioral Classifications. Behavioral durations and transitions

were recorded using an Epson HX-20 portable computer. Parasitoid

responses were classified into the following behaviors:

l.In vial: at least one antenna in vial.

2.Out of vial: both antennae out of vial.

3.Angle stance (ANGLSTN): standing on hind four legs with forelegs

in the air and body held at ca. 45 degree angle relative to substrate.

4.Standing/walking: all behaviors except ANGLSTN while on outside of

vial.









5.Take off: flight initiation.

6.Casting (CAST): side to side flight in horizontal plane without

upwind movement.

7.Zig/zagging (ZIG/ZAG): side to side flight in horizontal plane

with upwind movement.

8.Straight flight (STRTFLT): flight directly upwind.

9.Hovering (HOVER): stationary flight usually 5-10 cm from odor

nozzle.

10.Darting (DART): rapid straight flight toward odor nozzle.

11.Land source: landing on artificial leaf or on odor nozzle.

'.Land other: landing anywhere but on artificial leaf or odor

nozzle.

Flight Type Categories. Parasitoid flight responses were

categorized into one of the following types.

1.No flight: parasitoid did not leave vial during the 5 min allowed.

2.Nonoriented flight: nonanemotactic flight of short duration (ca.

1-3 sec) resulting in landing on ceiling or wall.

3.Incomplete oriented flight: parasitoid that exhibited at least two

of following behaviors; CAST, ZIG/ZAG, or STRTFLT, but did not land

on odor source.

4.Complete oriented flight: parasitoid that flew nonstop from vial

and landed on odor source.

Parasitoids making a nonoriented flight or an incomplete oriented

flight were captured with the vial and returned to the original starting

point. Parasitoids were given a maximum of three flights to make a







62

complete oriented flight, but they were tested to one treatment only and

never reused on another day.

Experiment I. Purpose: to characterize the general flight behavior

of female M. croceipes to both plant-host complex odors and control

odors, and to determine whether larval feeding was a necessary element

of the odor source. Male M. croceipes also were tested to determine

whether males were attracted to the plant-host complex odors. The four

treatments were: 1) females to plant-host complex (PHC) odors (= three

3rd instar H. zea larvae feeding on three cowpea seedlings for 5-7 h

before testing), 2) males to PHC odors, 3) females to control odors (-

humidified air), and 4) females to simulated PHC (SIM PHC) odors (=

three cowpea seedlings each with two 9-mm diam holes punched in the

leaves, feces produced by H. zea larvae feeding on cowpeas during 5-7 h

placed on a microscope slide, and three caged H. zea larvae. The four

treatments were tested using a randomized complete block (RCB) design,

with one replication consisting of five parasitoids tested individually

with each treatment. Ten replications were made.

Experiment II. Purpose: to determine what effect the removal of

individual components of the PHC had on the flight behavior of female M.

croceipes. The four treatments were: 1) PHC, 2) PHC minus feces

(PHC-FEC) (- three PHCs with all feces carefully removed and larvae

caged to prevent feeding and defecation during the test), 3) PHC minus

larvae (PHC-LAR) (= three PHCs with only larvae removed), 4) PHC minus

leaves (PHC-LEV) (= three caged third instar larvae plus the feces

produced by three third instar larvae feeding on cowpeas for 5-7 h

placed on a microscope slide), and 5) SIM PHC.









The five treatments were tested in a RCB design, with one

replication consisting of five parasitoids tested individually to each

treatment. Fifteen replications were made.

Experiment III. Purpose: to determine what effect individual

components of the PHC had on the flight behavior of female M. croceipes.

The four treatments were: 1) PHC, 2) feces (= feces produced by three

third instar larvae feeding on cowpeas for 5-7 h placed on a microscope

slide), 3) larvae (- three caged third instar larvae), and 4)

larval-damaged leaves (- three PHCs with both larvae and all feces

carefully removed).

The four treatments were tested in a RCB design, with one

-eplication consisting of eight parasitoids tested individually to each

treatment. Ten replications were made.

Experiment IV. Purpose: to determine the effects of active larval

feeding and artificial damage added to the PHC on the flight behavior of

female M. croceipes. The four treatments were: 1) PHC, 2) PHC +

artificial damage (PHC + Art. Dam.) (- three PHCs each with two 9-mm

holes punched out of the leaves), 3) nonfeeding larvae (LAR + LEV) (=

three third-instar larvae caged to prevent feeding and starved to

prevent defecation plus three undamaged cowpea seedlings), and 4)

feeding larvae (LAR w LEV) (= three third-instar larvae starved to both

delay defecation and to ensure active feeding when placed on three

cowpea seedlings).

The four treatments were tested in RCB design, with one replication

consisting of five parasitoids tested individually to each treatment.

Eight replications were made.









Statistical Analyses. To compare mean percent exhibiting the

flight types, Duncan's new multiple range test (DNMRT) (Steel and Torrie

1960) was used after analysis of variance (ANOVA) of percent

transformed by angular transformation. Mean percent exhibiting

specific behaviors also were compared using DNMRT after ANOVA of

percent transformed by angular transformation. Flight durations for

nonoriented, incomplete oriented, and complete flights were analyzed by

ANOVA. Significance levels were P 0.05 for all tests.



Results

Experiment I. The flight types exhibited by M. croceipes are shown

in Table 3-1. Females exhibited a high percentage of complete flights

and a very low percentage of no flights in response to PHC odors.

Control odors, on the other hand, elicited no complete flights and a

significantly higher percentage of both "no flights" and nonoriented

flights in females than that elicited by the PHC. The SIM PHC elicited

significantly fewer complete flights than elicited by the PHC. Nearly

every male tested exhibited a nonoriented flight when exposed to the PHC

odor. There was no significant treatment effect on percent exhibiting

incomplete oriented flight (F3 27 = 1.83, P <0.10). The percent

exhibiting ANGLSTN and other flight behaviors are shown in Figure 3-1.

Although some females exhibited some ANGLSTN, CAST, ZIG/ZAG, and STRTFLT

when exposed to control odors, it was at significantly lower levels than

that elicited by the PHC or the SIM PHC. Although both the PHC and the

SIM PHC elicited fairly high percentages of all behaviors, the PHC

elicited a significantly higher percentage of all behaviors except









Table 3-1. Flight types exhibited by Microplitis croceipes to
plant-host complex, simulated plant-host complex, and control odors.

b
Mean percent as


Treatment No flight Nonoriented Oriented

complete


Males to plant-host

complex



Females to control



Females to simulated

plant-host complex



Females to plant-

host complex


4 ab



16 b


6 ab


0 a


94 c


0 a


76 b


54 a


40 a


0 a


29 b


48 c


Fifty parasitoids were tested to each treatment.
Column means without letters in common differ significantly using
Duncan's New Multiple Range Test (P 0.05). Rows may not total to
100% due to individuals exhibiting incomplete oriented flights.





















PERCENT EXHIBITING (N=50)
100
FEMALES TO d
PHC 90 -


FEMALES TO c
SIM PHC c
70

FEMALES TO 60
CONTROL
50 b
MALES TO PHC / /

40

"/ // b
20 b b

10

-a a a
ANGLSTN CAST ZIG/ZAG STRTFLT HOVER DART
BEHAVIOR




Figure 3-1. Effects of control, plant-host complex, and simulated
plant-host complex on behaviors exhibited by Microplitis croceipes.
Bars within a cluster without letters in common differ significantly
using DNMRT (P = 0.05).








HOVER. Males exhibited a very low percentage of all behaviors when

exposed to the PHC odor.

Experiment II. The flight types exhibited by females to individual

components removed from the PHC are shown in Table 3-2. There was no

significant treatment effect on percent exhibiting nonoriented flight

(F 5- 1.21, P <0.25). Neither the removal of the leaves nor the
4, 56
removal of the larvae resulted in significantly fewer complete flights

than that elicited by the complete PHC. The removal of the feces,

however, did elicit significantly fewer complete flights than did the

complete PHC. The SIM PHC elicited significantly fewer complete flights

than did the PHC-leaves, even though both treatments contained feces and

larvae. There was no significant treatment effect on percent exhibiting

incomplete oriented flight (F, 56- 0.72, P <0.25).

The percent exhibiting the various behaviors in response to the

removal of individual components from the PHC are shown in Figure 3-2.

The PHC-FEC elicited the lowest percent for all behaviors, except HOVER,

and elicited a significantly lower percent than did the PHC for CAST,

ZIG/ZAG, and STRTFLT. Both the PHC-LEV and the PHC-LAR were

statistically equivalent to the complete PHC in the elicitation of all

behaviors recorded.

Experiment III. The flight types exhibited by females to the

individual components of the PHC are shown in Table 3-3. Larvae alone

elicited no complete flights and also elicited significantly more "no

flights" than any other treatment. Larval damaged leaves elicited

significantly fewer complete flights and significantly more no flights

than elicited by the complete PHC. Feces alone, on the other hand, was










Table 3-2. Flight types exhibited by female Microplitis coceipes to
individual components removed from the plant-host complex.

b
Mean percent as


Treatment No flight Nonoriented Oriented

Complete



Plant-host complex 13 a 24 a 51 c



Simulated plant- 35 b 37 a 19 a

host complex

Plant-host complex 8 a 36 a 47 be

minus leaves

Plant-host complex 15 a 29 a 40 be

minus larvae

Plant-host complex 20 a 36 a 37 b

minus feces



aSeventy-five parasitoids were tested to each treatment.
Column means without letters in common differ significantly using
Duncan's new multiple range test (P 0.05). Rows may not total to
100% due to individuals exhibiting incomplete oriented flights.




















PHC


PHC-LEAVES



PHC-LARVAE



PHC-FECES


ANGLSTN CAST ZIGZAG STRTFLT HOVER DART
BEHAVIOR


Figure 3-2. Effects of removing individual components of the plant-host
complex on behaviors exhibited by Microplitis croceipes. Bars within a
cluster without letters in common differ significantly using DNMRT (P =
0.05).









Table 3-3. Flight types exhibited by female Microplitis croceipes to
individual components of the plant-host complex.


Mean percent as


Treatment No flight Nonoriented Oriented

complete



Plant-host complex 13 a 33 a 48 c



Feces 23 ab 31 a 41 c



Larval damaged leaves 25 b 43 a 28 b



H. zea larvae 71 c 28 a 0 a


bEighty female parasitoids were tested to each
Column means without letters in common differ
Duncan's new multiple range test (P 0.05).
100% due to individuals exhibiting incomplete


treatment.
significantly using
Rows may not total to
oriented flights.








71

statistically equivalent to the complete PHC in the elicitation of both

"no flights" and complete flights. There was no significant treatment

effect on percent exhibiting incomplete oriented flight (F, 27 0.61,

P <0.25).

The percent exhibiting the various behaviors in response to the

individual components are shown in Figure 3-3. Larvae alone elicited

very low percent of all behaviors, significantly less than the other

treatments for all behaviors. The larval damaged leaves elicited a

significantly lower percent of all behaviors, except ANGLSTN and HOVER,

than did the complete PHC. Feces alone was statistically equivalent to

the PHC in the elicitation of all behaviors recorded.

Experiment IV. The flight types exhibited by females in Experiment

IV are shown in Table 3-4. The PHC + Art. Dam. elicited significantly

fewer complete flights than did the PHC. The PHC elicited both

significantly more complete flights and significantly fewer "no flights"

than did either the nonfeeding larvae or the feeding larvae. Nonfeeding

larvae and feeding larvae were equivalent statistically in the

elicitation of both complete flights and "no flights." There was no

significant treatment effect on percent exhibiting incomplete oriented

flight (F3, 15 = 1.85, P <0.10).

The percent exhibiting the various behaviors are shown in Figure 3-

4. The PHC and PHC + Art. Dam. were statistically equivalent in the

elicitation of all behaviors. Nonfeeding larvae and feeding larvae also

were equivalent statistically in the elicitation of all behaviors

recorded.



















PHC


FECES


DAMAGED
LEAVES

LARVAE


b
b
b

c
C
c c
/C

a -b

",b I ib




a / /
,. /,_1


ANGLSTN CAST ZIG/2AG STRTFLT
BEHAVIOR


HOVER


DART


Figure 3-3. Effects of individual components of the plant-host complex
on behaviors exhibited by Microplitis croceipes. Bars within a cluster
without letters in common differ significantly using DNMRT (P = 0.05).


PERCENT


EXHIBITING (N=80)









Table 3-4. Flight types exhibited by female Microplitis croceipes to
plant-host complex, plant-host complex + artificial damage, feeding
larvae and nonfeeding larvae.

b
Mean percent as


Treatment No flight Nonoriented Oriented

Complete



Plant-host complex 2.1 a 50.0 a 45.9 c



Plant-host complex 18.1 ab 45.9 a 25.9 b

+ artificial damage



Feeding larvae 28.0 bc 48.8 a 15.0 a



Nonfeeding larvae 45.4 c 35.9 a 8.5 a



bForty-eight parasitoids were tested to each treatment.
Column means without letters in common differ significantly using
Duncan's new multiple range test (P = 0.05). Rows may not total to
100% due to individuals exhibiting incomplete oriented flights.























PHC c
S bc
I I

PHC+ART.
DAMAGE b
V///2 70 a

FEEDING 60
LARAVE
50 ab b
NON-FEEDING b
LARVAE 40 b



30
2/aab



10 a eaa a


ANGLSTN CAST ZIGZAG STRTFLT HOVER DART
BEHAVIOR




Figure 3-4. Effects of artificial damage on the plant-host complex and
the effect of active larval feeding on behaviors exhibited by
Microplitis croceipes. Bars within a cluster without letters in common
differ significantly using DNMRT (P = 0.05).









For all four experiments, the ANOVAs of flight durations did not

reveal any significant treatment effects for nonoriented flights,

incomplete oriented flights, or complete flights. The overall mean

flight durations were 2.5 sec (n 427), 14.9 sec (n = 86), and 36.3 sec

(n 315), respectively.



Discussion

Although female M. croceipes did not fly upwind to control odor

sources, females, but not males, flew upwind to odors from the PHC.

Drost et al. (1986), testing females only, reported similar findings.

Elzen et al. (1987) reported that male M. croceipes were not attracted

to odor sources attractive to females. The observation that the PHC

odors attract only females suggests that the attraction is a

host-seeking response.

Although some females exhibited ANGLSTN when exposed to control

odors, this behavior was exhibited to a much higher degree in response

to PHC odors. This suggests that PHC odors elicit this preflight

behavior in females. In addition, because the PHC odors elicited fewer

no-flights than did the control odors, PHC odors also appear to

stimulate flight initiation in females. Although chemical odors have

been demonstrated to elicit flight in male moths (Shorey 1964), as far

as is known, this is the first report of a chemical odor eliciting

flight initiation in a resting parasitoid.

Because only the PHC-FEC elicited significantly fewer complete

flights than did the PHC and because of all the individual components

tested, only the feces treatment was statistically equivalent to the









PHC, feces appear to be the most important factor mediating the

host-locating flight behavior of M. croceipes. Lewis and Jones (1971)

demonstrated that H. zea feces mediated the close-range behaviors of

decreased locomotion and antennation, and subsequently

13-methylhentriacontane was identified as the most active compound

(Jones et al. 1971). Heliothis zea feces were found to be attractive to

M. croceipes females in an olfactometer bioassay (Chapter II) and a wind

tunnel study (Elzen et al. 1987). Feces have been demonstrated to

attract parasitoids in only a few cases (McKinney and Pass 1977, Hsaio

et al. 1966), and LeCompte and Thibout (1986) reported that feces were

unimportant during host-location for an ichneumonid.

Heliothis zea larval odor did not seem to have any effect on the

host-location flight behavior of M. croceipes. M. croceipes females

were similarly unattracted to H. zea larvae in an olfactometer (Chapter

II) although Elzen et al. (1987) reported M. croceipes exhibited

odor-directed flight to H. virescens larvae. Rotheray (1981) also

reported larvae as being unattractive, though many authors have reported

larvae as being attractive to their respective parasitoids (Thorpe and

Jones 1937, Monteith 1955, Mossadegh 1980, Nettles 1 Y'). It is

interesting that in this case, feces seems to be a better indicator of

larval presence than the larva itself.

The observation that the SIM PHC that contained neither active

larval feeding nor larval damaged leaves was attractive at all,

demonstrates that neither active feeding nor larval damage is necessary

for the attraction of M. croceipes. The insignificant role of feeding

larvae was further demonstrated in the comparison of LAR + LEV (no









feeding or frass) and LAR w LEV (feeding but no frass) both of which

were equally low in the attraction of M. croceipes. Feeding damage has

been reported to be attractive for another parasitoid (Nealis 1986).

Although Drost et al. (1986) reported that M. croceipes females

were not attracted to undamaged pea seedlings (n = 9), in this study,

there was a slight attraction to LEV + LAR (no feeding or frass).

Because the LAR alone elicited no attraction, this attraction is

probably due to a slight attraction to undamaged leaves. Many authors

have reported parasitoids being attracted to their host's food,

regardless of host presence (Thorpe and Jones 1937, Monteith 1955,

Arthur 1962, Read et al. 1970, Camors and Payne 1972, Nettles 1980,

Powell and Zhi-Li 1983, Elzen et al. 1984b, Elzen et al. 1986).

Although the SIM PHC was attractive, it was significantly less

attractive than the PHC, suggesting that either something attractive was

missing or that something inhibitory was present. Because the SIM PHC

was less attractive than the PHC-LEV, even though both contained f

and larvae, it is possible that there was something inhibitory about

SIM PHC, specifically the artificially damaged leaves. This was

demonstrated when the PHC + Art. Dam. was shown to be less attractive

than the PHC. The inhibitory nature of the artificial damage was a

somewhat unexpected finding considering artificial damage was found to

be attractive to M. croceipes in an olfactometer study (Chapter II).

This difference probably is due to different techniques used to damage

the leaves. Artificial damage has been demonstrated to be attractive to

some parasitoids (Elzen et al. 1983).







78

It was somewhat surprising that there were no treatment effects on

flight durations for any flight type. Although some odor sources

elicited only a few complete flights, females did not take any longer to

make the flight, suggesting that the "decision" to make an oriented

flight is made early, perhaps before flight initiation.

In summary, odors from the plant-host complex stimulate both flight

initiation and the subsequent host-location flight behaviors in female

M. croceipes. Larval feces are the most important single component of

the plant-host complex in stimulating both flight initiation and

host-locating flight behaviors in female M. croceipes. Heliothis zea

larvae alone are unattractive, and leaves are only very slightly

etractive whether or not there is larval feeding. Artificial damage

can inhibit the host-locating flight behavior of female M. croceipes.












CHAPTER IV
INTRASPECIFIC COMPETITION IN Microplitis croceipes (CRESSON)


Introduction

Although host discrimination (i.e., the ability to distinguish

between unparasitized and parasitized hosts) (Lenteren 1981) has been

studied previously in M. croceipes (Lewis and Snow 1971, Vinson and

Guillot 1972), both studies reported that M. croceipes did not

discriminate between unparasitized and once-parasitized Heliothis

larvae. This lack of host discrimination leads to superparasitism and

subsequently intraspecific larval-larval competition. Little is known

about intraspecific competition in M. croceipes except that

approximately 95% of superparasitized hosts produce only one parasitoid

(Bryan et al. 1969a, Lewis and Burton 1970). Therefore, intraspecific

competition in M. croceipes generally results in the elimination of

conspecifics within the host larva.

Interspecific competitive outcomes are easily assessed by the

parasitoid species emerging from the host (e.g. Lewis and Brazzel 1968,

Vinson and Ables 1980). Intraspecific competitive outcomes are much

more difficult to determine (Chow and MacKauer 1984) and the survival

rates of M. croceipes larvae resulting from superparasitization are

unknown. Genetically marked parasitoids have been used in some

intraspecific competition experiments, however, the marked individuals

were found to be weaker competitors (Bakker et al. 1985). In this







80

study, I report a novel means for determining intraspecific competitive

outcomes for an endoparasitoid.

The purpose of this study was: 1) to determine if mated and

unmated females of an arrhenotokous (unmated females produce only male

offspring) parasitoid could be used to determine competitive outcomes

between conspecifics within a host (by sex ratios of emerging

parasitoids), and 2) to determine how the time interval between the

initial parasitization and the superparasitization affected the survival

rates of the parasitoid larvae resulting from these ovipositions. This

information could then be used to provide an adaptive explanation for

the lack of host discrimination reported earlier and also predict how M.

croceipes should respond towards host larvae parasitized for various

lengths of time.



Materials and Methods

Parasitoids. Microplitis cro-eipes were handled as described in

Chapter II. Unmated females were obtained by holding cocoons in

individual cells for emergence. Mated females were taken from a cage in

which both males and females were present and mating was assumed. For

all experiments, parasitoids were used at 3-5 days postadult emergence.

All ovipositions were by naive females (i.e., females without prior

oviposition experience) because naive females are more likely to

superparasitize than experienced females (Lenteren and Bakker 1975).

Host Insects. Heliothis zea (Boddie) larvae were obtained as

described in Chapter II and were reared on artificial diet (Guy et al.

1985) under conditions described in Chapter II.









Experimental Procedure Because genetically marked M. croceipes

were unavailable, but M. croceipes is arrhenotokous (Bryan et al. 1969a,

Lewis and Snow 1971), competitive outcomes were determined by using

mated and unmated females in the competition experiments and comparing

the sex ratios of the emerging parasitoids. For example, if the progeny

from the initial oviposition have an advantage over progeny from

superparasitizations, and the initial oviposition was by an unmated

female and the superparasitization was by a mated female, most of the

emerging parasitoids will be males. The nine treatments were as

follows: mated only (M) larvae parasitized by a mated female; unmated

only (U) larvae parasitized by an unmated female; both same day (BSD)

- larvae parasitized by both a mated and an unmated female (random

order) within 10 minutes of each other; larvae parasitized by a mated

female subsequently superparasitized by an unmated female after 2, 4, or

6 days, (M2U),( M4U), and (M6U), respectively; larvae parasitized by an

unmated female subsequently superparasitized by a mated female after 2,

4, or 6 days, (U2M), (U4M), and (U6M), respectively. Initial

ovipositions were in premolt second instars (ca. 6 days post egg-hatch)

and all ovipositions were observed individually in a 9-cm diameter petri

dish. Only one oviposition was allowed per parasitoid.

After parasitizations, larvae were held individually in plastic

jelly cup trays (Ignoffo and Boenig 1970) with ca. 3.5 cm3 of artificial

diet. Larvae were checked daily for emergence of parasitoid larvae from

the host, host pupation, and parasitoid adult emergence.

One replication consisted of the nine treatments applied to nine

sets of ten larvae. The entire experiment was replicated ten times.







82

Percent parasitoid larval emergence, percent parasitoid adult emergence,

and percent female emergence were analyzed by analysis of variance

(ANOVA) after angular transformation of percent. Regression analyses

were performed to determine how time between ovipositions affected the

sex ratio of the emerging parasitoids. Significance levels were P =

0.05 for all tests.



Results and Discussion

Female parasitoids readily stung Heliothis larvae in all

treatments. However, to insure that a 'sting' resulted in an actual

oviposition, four sets of ten larvae were superparasitized 0, 2, 4, and

6 days after the initial parasitization. These larvae were subsequently

dissected (within 10 minutes) and the number of parasitoid eggs and

parasitoid larvae (eggs/larvae) recorded. The numbers of eggs/larvae

were 19/0, 10/10, 8/9, and 6/10, respectively. Although the number of

parasitoid eggs was somewhat lower for larvae parasitized for 6 days, it

was difficult to find the eggs in these hosts because of their larger

size and cloudy hemolymph. Even so, a relatively high percentage of

observed stings resulted in actual oviposition, therefore I assumed that

a sting resulted in an oviposition.

Percent larval and adult parasitoid emergences are shown in Table

4-1. There were no significant treatment effects on either percent

larval or adult emergence (F8,72 = 1.29, P>0.25 and F8,72 = 1.97,

P>0.05, respectively). Therefore, mated and unmated females are

apparently equally likely to oviposit. In addition, superparasitization

of hosts did not produce either more or less progeny than once-











CO

- *- 0
LA Ln 0o
H- H- H-


0 )









I)


* I
H H





L LO Ln
r-H r-
1-1 i-


in n i


* *
in in in
r-i 11 .-


LO o

OVONN
O Co 0 CO
H rH


le

o n
oo cr


o N

e 0

N CO
T CO


CO
<-t \ ^r 'a
o\ w o\ o


Si i
a! illilil S^


Cl r-I -










O H H H
Lo ON' r- o





o co co


~rn


4*
I4
0










4.'




j*i


4j
()




I
I

.A









parasitized hosts. Therefore, any observed differences in parasitoid

sex ratios between treatments are not due to the hosts overcoming the

initial parasitization but succumbing to the superparasitization.

The ANOVA of percent female emergence revealed significant

treatment effects (F8,72 24.8, P<0.005). Figure 4-1 shows percent

female emergence for the nine treatments. The solid line represents the

regression equation for percent female emergence as a function of the

time between an oviposition by a mated female followed by an oviposition

by an unmated female. The broken line represents the regression

equation for percent female emergence as a function of the time between

an oviposition by an unmated female followed by an oviposition by

mated female. The slopes of these lines were of opposite sign but o0

equal absolute value (t66 0.03, P>0.5) (Steel and Torrie 1960).

Percent female emergence for treatment BSD was midway between that for

treatments M and U, indicating that when initial parasitization and

superparasitization occurred on the same day, progeny from mated and

unmated females had equal survival rates. Although the fitness

conferred by superparasitization is less than that for attacking an

unparasitized host, it is not zero and may be adaptive under certain

circumstances (Bakker et al. 1985). These results provide an adaptive

explanation for the lack of discrimination by M. croceipes reported in

earlier studies (Lewis and Snow 1971, Vinson and Guillot 1972). In

addition, the equal progeny survival rates suggest there is no intrinsic

superiority of one sex of parasitoid larvae over the other in larval-

larval competition.
















100
90
80
70
60
50
40
30
20
10
0


Days Between Ovipositions


Figure 4-1. Effect at time between initial parasitization and
superparasitization on percent female Microplitis croceipes emergence
using mated and unmated females. BSD is used when initial oviposition
superparasitization both occurred on the same day.


(Y=24.4 + 4.9X )
Mated Only Mated-*Unmated



,SD
&- -... Unmated- Mated
-- -. (Y=29.3 5.0X )
Unmated Only 0










As time increases between initial parasitization and subsequent

superparasitization, progeny survival rates decrease for the

superparasitizing female while progeny survival rates increase for the

female ovipositing initially. Older parasitoid larvae are often

reported to have a competitive advantage (Browning and Oatman 1984, Chow

and MacKauer 1984). From this, one would predict increased avoidance of

parasitized larvae (i.e.the exhibition of host discrimination) with

increased time since the initial parasitization. Although this study

and previous studies (Lewis and Snow 1971, Vinson and Guillot 1972)

indicate that M. croceipes does not discriminate at the host acceptance

level of host selection, M. croceipes can discriminate between

unparasitized and once-parasitized hosts under more natural conditions

(Chapter V and unpublished data, F. WAcker).

The equal magnitudes of the slopes of the two regressions furtl--

indicates that there is no intrinsic superiority of one sex of

parasitoid larvae over the other in larval-larval competition. This is

in contrast to Labeyrie and Rojas-Rousse (1985) who reported males were

inferior to females in larval-larval competition.

Of the 700 Heliothis larvae superparasitized in this experiment,

only 17 (2.4%) produced two parasitoid larvae. In two of these 17 cases

neither reached adulthood (treatments M2U and U2M), in six cases only

the first to emerge reached adulthood (1, 2 and 2 males from treatments

M2U, U2M and M4U, respectively and 1 female from treatment M2U), in six

cases only the second to emerge reached adulthood (1 and 2 males from

treatments M2U and M4U, respectively and 2 and 1 females from treatments

U2M and U4M, respectively), and in three cases both reached adulthood (2










males and one female followed by three males all from treatment M2U).

In the three cases where both parasitoid larvae reached adulthood, the

first to emerge was of normal size while the second to emerge was

invariably of subnormal size. Interestingly, there was no double

emergence of parasitoid larvae from treatments BSD, M6U, U6M. For

treatment BSD, this may have been a result of intense physical combat

between mandibulate first instar parasitoid larvae. Although only first

instar M. croceipes possess sickle-shaped mandibles (Lewis 1970a, Greany

1986), apparently this does not insure their success over older

conspecifics. These older parasitoid larvae probably defeat younger

conspecifics through physiological suppression or selective starvation

(Steinberg et al. 1987). This 'suppression' of younger conspecifics is

virtually complete when the older larvae have a six day head-start as

seen in treatments M6U and U6M which had sex ratios equal to those for

treatments M only a!d U only, respectively, and no double emergences.

Parasitoid emergence times are shown in Table 4-1. For treatment

M, the mean larval emergence time for both males and females was 8.0

days post-parasitization. The mean male and female adult emergence

times for treatment M were 15.3 and 16.1 days post-parasitization,

respectively. These emergence time data indicate that the mechanism of

protandry in M. croci .pes reported by Bryan et al. (1969a) and Lewis and

Burton (1970) is a result of accelerated pupal-adult development of

males compared to females and not accelerated larval development of

males.

Larval development time was longer for the second parasitoid than

that for the first parasitoid, even after subtracting the time between









the initial oviposition and the superparasitization (Table 4-1). In

treatments M2U and U2M, the larval developmental times (after

subtracting 2 day interval between ovipositions) for second to emerge

parasitoids were 2.0 and 2.5 days longer than that for first to emerge

parasitoids, respectively. In treatments M4U and U4M, the larval

developmental times (after subtracting 4 day interval between

ovipositions) for second to emerge parasitoids were 2.4 and 2.0 days

longer than that for first to emerge parasitoids, respectively. Both

the first to emerge and the second to emerge parasitoids spent ca. 7.0

days between larval emergence and adult emergence. So, in addition to a

decreased chance of survival, parasitoid larvae from a

superparasitization which co-develop with a parasitoid larva from an

initial parasitization also exhibit both a decreased developmental rate

and decreased size. These in turn may result in decreased longevity,

fecundity, and searching rate (Hawkins and Smith 1986, Narayanan

Subba Rao 1955, Waage 1986).

In summary, for arrhenotokous parasitoids with no intrinsic

superiority of one larval sex over the other, mated and unmated females

can be used to determine intraspecific competitive outcomes by comparing

the sex ratios of the emerging parasitoids. For M. croceipes, progeny

survival rates for a superparasitizing female are approximately equal to

those for the female ovipositing initially when the ovipositions occur

on the same day. However, progeny survival rates for a

superparasitizing female decrease with increased time between the

initial oviposition and the superparasitization. Therefore, the

avoidance of parasitized hosts (ie. the exhibition of host