Title Page
 Table of Contents
 List of Tables
 List of Figures
 Literature review and research...
 Analysis of the reproductive behavior...
 Epidermal glanos in the gvipositor...
 Behavioral and electrophysiological...
 The function of sex pheromones...
 Identification of a sex pheromone...
 Summary and conclusions
 Biographical sketch

Group Title: role of sex pheromones in the reproductive isolation of Heliothis species (Lepidoptera: Noctuidae) /
Title: The role of sex pheromones in the reproductive isolation of Heliothis species (Lepidoptera: Noctuidae) /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00099101/00001
 Material Information
Title: The role of sex pheromones in the reproductive isolation of Heliothis species (Lepidoptera: Noctuidae) /
Physical Description: xi, 131, 1 leaves : ill. ; 28 cm.
Language: English
Creator: Teal, Peter Edmund Allan, 1953-
Publication Date: 1981
Copyright Date: 1981
Subject: Heliothis   ( lcsh )
Pheromones   ( lcsh )
Insects -- Behavior   ( lcsh )
Entomology and Nematology thesis Ph. D
Dissertations, Academic -- Entomology and Nematology -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis (Ph. D.)--University of Florida, 1981.
Bibliography: Bibliography: leaves 120-131.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Peter Edmund Allan Teal.
 Record Information
Bibliographic ID: UF00099101
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000295074
oclc - 07841643
notis - ABS1416


This item has the following downloads:

PDF ( 5 MBs ) ( PDF )

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
    List of Tables
        Page vi
    List of Figures
        Page vii
        Page viii
        Page ix
        Page x
        Page xi
    Literature review and research aims
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
    Analysis of the reproductive behavior of heliothis virescens (F.) under laboratory conditions
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
    Epidermal glanos in the gvipositor of heliothis virescens (F.)
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
    Behavioral and electrophysiological assay of extracts from two sex pheromone gland sites of female heliothis virescens (F.)
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
    The function of sex pheromones in the reproductive isolation of heliothis virescens (F.) from heliothis subflexa (GN.) under laboratory conditions
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
    Identification of a sex pheromone of heliothis subflexa (GN.) and field trapping studies using different blends of components
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
    Summary and conclusions
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
    Biographical sketch
        Page 132
        Page 133
        Page 134
        Page 135
Full Text







Copyright 1981


Peter E. A. Teal

For Margaret Louise Ferguson Teal and Barbara Allison Teal

with all my love.



I wish to thank the staff of the Insect Attractants, Behavior,

and Basic Biology Research Laboratory, AR-SEA, USDA, for their

support and help throughout this project, the Natural Sciences and

Engineering Research Council of Canada for providing me with a

postgraduate scholarship which enabled me to conduct this research,

and the Department of Entomology and Nematology of the University

of Florida for enabling me to continue my education. The members

of my graduate committee and staff of the chemistry section of the

IABBBRL were also exceedingly helpful and deserve considerable

credit for putting up with me. I would particularly like to thank

D. L. Chambers, R. R. Heath, M. D. Huettel, J. R. McLaughlin and

M. M. Brennan. Finally, I wish to thank Jim Tumlinson for his

time, help, ideas, knowledge, and ability to help me maintain a

love for my work and keep my head pointed in the right direction.


ACKNOWLEDGMENTS . . . . . . . . . . . iv

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

LIST OF FIGURES. . . . . . . . . . . vii

ABSTRACT . . . . . . . . . . . ix


CONDITIONS . . . . . . . . . 21

VIRESCENS (F.) . . . . . . . . 43


CONDITIONS . . ... . . . .... .78



REFERENCES . . . . . . . . ... . . . 120

BIOGRAPHICAL SKETCH . . . . . . . . ... . 132


Table Page

1. Courtship rejections of H. virescens males as indicated
in small cage laboratory videotape studies . . ... 27

2. First-order behavioral transitions in successful matings
of H. virescens observed in laboratory wind tunnel studies. 35

3. x2 Comparisons of behaviors elicited by 1 FE
concentrations of the test samples . . . . .... 70

4. x2 Comparisons of behaviors elicited by 10 FE
concentrations of the test samples . . . . .. 71

5. x2 Comparisons of behaviors elicited by 1 and 10
FE extract concentrations. . . . . . . . 73

6. Comparison of the effects of test samples on temporal and
behavioral aspects of H. virescens male reproductive
behaviors . . . . . . . . . . . . 75

7. Equivalent chain length units (ECL) of GC standards on the
SP2340 and cholesterol cinnamate capillary columns as
calculated using saturated C14-C16 acetates as the func-
tional retention index . . . . . 104

8. Chemical components isolated from H. subflexa ovipositor
washes . . . . . . . ... . . . . 101

9. Comparison of cone trap captures of male H. subflexa using
females, synthetic chemicals, and crude extracts. .... .106

10. Comparison of sticky trap captures of H. subflexa males
using different blends of components. . . . . ... 109


Figure Page

1. Precourtship reproductive behaviors of initially inactive
male H. virescens. . . . . . . . . . . 30

2. Comprehensive ethogram of 30 successful matings. . . ... 34

3. Most frequent behavioral paths in successful couplations
based on standard normal deviates having probabilities <0.05
in a binomial test . . . . . . . . ... . 38

4. Ventrolateral aspect of the terminal abdominal segments of
female H. virescens. . . . . . . . .... . 47

5. Longitudinal section through the ovipositor. . . . .. 47

6. Cross-section through the GI area of a partially extended
ovipositor . . . . . . . . ... .. .. . .49

7. Undifferentiated epidermal cells overlying the hindgut . 49

8. Columnar and cuboidal cells forming GI . . . . .. 49

9. Columnar gland cells of GI . . . . . . . . 52

10. Longitudinal section through the dorsal valves indicating 2
glandular types . . . . . . . . . 52

11. Section through columnar epidermal cells of the dorsal
valves . . . . . . . . . . . . 54

12. Section through trichogenous gland of dorsal valves. ... 54

13. Section through trichogenous gland . . . . .... 54

14. Mean EAG responses of 10 males to glandular extracts . .. 62

15. Behavioral sequences evoked by 1 FE concentration of each
test extract and the solvent blank in the flight tunnel. . 64

16. Behavioral sequences evoked by 10 FE concentrations of each
extract and the solvent blank in the flight tunnel . . 66

Figure Page

17. Male hovering at ca. 3 cm from dispenser. . . . ... 68

18. Same male after landing on dispenser. . . . . ... 68

19. Male exposing his hairpencils on dispenser. . . . ... 68

20. Male searching dispenser. . . . . . . . ... 68

21. Ventral view of male searching dispenser . . . ... 68

22. Homosexual mating attempt near dispenser. . . . ... 68

23. Behaviors performed by initially inactive H. subflexa
males released downwind from actively calling H. virescens
females . . . . . . . . . . . . . 84

24. Behaviors performed by initially inactive H. subflexa
males released downwind from a 1 FE sample of the H.
virescens pheromone gland extract . . . . . .. 86

25. Behaviors performed by initially inactive H. virescens
males released downwind from actively calling H. subflexa
females . . . . . . . ... . . . . 91

26. Behaviors performed by initially inactive H. virescens
males released downwind from a 1 FE sample of the H.
subflexa pheromone gland extract. . . . . . ... 93

27. Chromatogram of standard compounds eluting from the
SP2340 column . . . . . . . . .. . 103

28. Chromatogram of components in the ovipositor extracts
eluting from the SP2340 column. . . . . . . ... 103

29. Chromatogram of standard compounds eluting from the
cholesterol cinnamate column. . . . . . . .. 103

30. Chromatogram of components in the ovipositor extracts
eluting from the cholesterol cinnamate column ...... 103

31. Chemicals identified from pheromone gland extracts of
H. zea, H. virescens, and H. subflexa . . ..... .. 119

Abstract of Dissertation Presented to the Graduate Council of the
University of Florida in Partial Fulfillment of the Require-
ments for the Degree of Doctor of Philosophy



Peter Edmund Allan Teal

June 1981

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

Chemical, behavioral, morphological, and physiological aspects of

the sex pheromone communications systems of Heliothis virescens (F.)

and H. subflexa (Gn.) were studied under laboratory and field conditions.

Behavioral analyses of the semiochemically induced inter- and

intra-specific reproductive interactions between H. virescens and H.

subflexa were conducted in the laboratory. The reproductive behavior

of H. virescens was broadly categorized into 2 distinct phases dependent

upon the interactive forces governing each. Precourtship behaviors

include female calling, male activation, and orientation behaviors,

all of which rely on a unidirectional flow of behavioral information

and are rather stereotyped in their sequence. Courtship involves

numerous variably committed interactions between the sexes. Nonethe-

less, a common courtship sequence is apparent, with several points at

which reproductive isolation from closely related species could be

effected. H. subflexa males were incapable of orienting toward

calling H. virescens females, indicating that semiochemically induced

reproductive isolation acts at a distance. H. virescens males were

capable of orienting toward calling H. subflexa females but the majority

of males either failed to enter into or complete courtship. Therefore,

close range reproductive isolation appears to maintain the genetic

integrity of this pair. The pheromone blend extracted from H. subflexa

female glands did not induce H. virescens males to enter taxis, indi-

cating that there are differences between the pheromone blend released

by calling females and that extracted from the glands.

The ovipositor of H. virescens was found to have 2 areas of

glandular tissue. The 1st area, located in the intersegmental membrane

between abdominal segments 8 and 9 + 10, has the form of a ventro-

lateral chevron. The 2nd glandular area, situated in the dorsal

valves papillaee annales), is composed of an area of columnar epidermal

cells, similar to cells in the intersegmental membrane, and of trichogenous

glands associated with tubular setae. Extracts from these 2 glandular

areas were assayed for their ability to elicit both EAG responses and

male sexual behavior in bioassays. EAG studies indicated that extracts

from both glandular sites were perceived by the male antennae, and

wind tunnel bioassays confirmed the role of extracts from each glandular

site in the elicitation of male sexual behaviors. However, due to

disparate response levels to individual and combined extracts, it is

hypothesized that each glandular site maintains a different pheromone

blend and the admixture of the 2 sites is necessary for maximizing

male sexual behaviors.

Gas chromatographic and mass spectral analysis of extracts

obtained from the ovipositor of calling H. subflexa females revealed

the presence of 8 compounds which were identified as hexadecanal, (Z)-

9-hexadecenal, (Z)-ll-hexadecenal, (Z)-7-hexadecen-l-ol acetate, (Z)-

9-hexadecen-l-ol acetate, (Z)-ll-hexadecen-l-ol acetate, (Z)-9-hexadecen-

1-ol, and (Z)-11-hexadecen-1-ol. Although the whole blend was found

to be an effective male attractant, the deletion of the alcohols from

the blend increased trap captures considerably. Further, although the

binary mixture of (Z)-9-hexadecenal and (Z)-ll-hexadecenal caught some

male H. subflexa, significant increases in captures were noted when

the 3 acetate components were included in the blend.



The subfamily Heliothidinae (Lepidoptera: Noctuidae) is composed

of an assemblage of 14 genera including some 158 species (Hardwick,

1970). Until recently the major North American pests of this sub-

family were considered members of a single genus, Heliothis, which

included Heliothis armigera (Hbn) and H. virescens (F.) (Brazzel et

al., 1953). The North American form of H. armigera was subsequently

given species status (Common, 1953) and renamed Heliothis zea (Boddie)

by Todd (1955). Through further reclassification (see Hardwick, 1965)

the original species, H. armigera, was divided into 6 species groups

and reclassified in the genus Helicoverpa (Hardwick, 1965). The basis

of Hardwick's classification lies in the presence of cornutii on the

male vesica and the female appendex bursae which is coiled or alter-

natively dilated and constricted. The genus Heliothis is, according

to Hardwick (1970), distinguished from Helicoverpa in that male Heliothis

have a denticulate bar at the base of the vesica adjacent to the right

margin of the aedeagus and females have an appendex bursa of approxi-

mately the same size as the funds bursae. To avoid the taxonomic

arguments surrounding this species, Helicoverpa (Heliothis) zea is

herein referred to as Heliothis zea.

Based on Hardwick's monograph (1970) the genus Heliothis is

composed of 13 species including H. virescens (F.) and H. subflexa

(Gn). The latter species is so morphologically similar to H. virescens

that its species status was under considerable question until McElvare

(1941) suggested the restoration of H. subflexa to the specific rank

on the basis of an absence of wing banding and internal genitalical

differences. Brazzel et al. (1953) negated the use of wing banding,

noting that only male H. subflexa had immaculate wings. Their studies

on genitalic structure were, however, in general agreement with those

of McElvare (1941). Similarly Furr et al. (1974) have concluded that

male genitalic differences were sufficient to segregate both species

and H. virescens male X H. subflexa female hybrids.

The eggs and first 2 larval instars of both species are identical;

however, in the 3rd instar H. virescens diverges, developing spicuoles

on setigerous tubercles on abdominal segments 1, 2, and 8, and a basal

process on the oral face of the mandible which becomes tooth-like in

later instars (Brazzel et al., 1953). Larval H. subflexa are distin-

guished by their habit of feeding solely on ground cherry, Physalis

spp., and development of short microspines on the basal half of the

abdominal tubercles (Brazzel et al., 1953). The pupae of both species

are apparently indistinguishable.

Seasonal Distributions

Due to the polyphagous nature of both H. zea and H. virescens

range limitations are apparently solely dependent upon seasonal climatic

conditions, with northward extensions governed by the maximum distances

travelled by individuals of successive summer generations (Hardwick,

1965). Hardwick states that few, if any, individuals are capable of

overwintering as pupae above 40N latitude; hence pest outbreaks in

these northern areas (see Beirne, 1971) are the result of seasonal

influxes. The number of generations per year increases from north to

south, there being 4-5 generations in Louisiana, Arkansas, and North

Florida (Brazzel et al., 1953; Snow and Brazzel, 1965), with the

penultimate generation overwintering in pupal diapause while the last

is generally killed by the first hard frost (Gentry et al., 1971). In

subtropical areas there is no cold-induced diapause and hence a

continuum of generations per year (see Hardwick, 1965). In addition

to the effects of climate on range limitations, H. subflexa is further

limited by the distribution of its host plant, the ground cherry.

Although little is known about the population dynamics of this species

it probably closely approximates that of H. virescens (see Brazzel et

al., 1953).

Life Cycles

Exceptionally detailed descriptions of the life histories of H.

virescens and H. subflexa have been given by Brazzel et al. (1953) and

that of H. zea by Hardwick (1965). As all 3 have quite similar life

cycles, the following is a generalized synopsis of the observations of

the above authors.

Eggs are laid over an extended period, between dusk and dawn,

although late in the season at higher latitudes adults can be found

ovipositing in the late afternoon and evening. Oviposition occurs on

or near flowers, with terminal growth being the preferred site.

Larvae emerge from their eggs after an incubation period of 3-14 days,

depending upon climatic conditions, and proceed to move through the

terminal buds to feed from the undersides of leaves. Larvae of

middle stadia then move to the reproductive parts of the plant which

not only provide protection from predators but also from pesticides

during the final phases of larval development. Both the number of

larval instars (5-7) and larval feeding period are variable and depend-

ent upon temperature and the nutritive value of the food consumed.

Larval development may be completed in as little as 2 weeks at 27-300C

when feeding on cotton bolls, or as prolonged as a month on tomato or

tobacco leaf buds at temperatures in the low 200C's. In response to

food having low protein content, such as alfalfa, larvae of mid and

later instars may become carnivorous, feeding on the pupae of other

Lepidoptera, or cannabilistic when in high numbers.

After completion of feeding, larvae enter a prepupal stage of 2-3

days during which time they migrate down the host plant, burrow into the

soil, and construct a pupal cell at 3-23 cm beneath the surface. Pupal

periods are variable and again dependent upon climatic factors. In

northern and arid climates pupae enter an arrested state of development

during inclement months while during the growing season and in southern

temperature climes the pupal period is generally 12-33 days.

Reports on adult longevity are somewhat ambiguous; mean values of

5.5 and 8.4 days are reported for unmated and mated females by Callahan

(1958a), while values reported by both Hardwick and Brazzel et al. are

at least twice that length. The longer figures seem more reasonable in

view of the long northward migrations, with pest outbreaks of both H.

zea and H. virescens being reported as far north as central Saskatchewan

(Beirne, 1971), and estimated northward migration of 440 km by at least

some individuals of each generation (see Hardwick, 1965).

Host Plants and Economic Damage

Although both H. zea and H. virescens show considerable reference

for plants of the families Leguminosae and Solanaceae, the extensive

host lists of wild and cultivated plants, including both graminaceous

and malvaceous plants, indicate that they are polyphagous (Barber, 1937;

Brazzel et al., 1953; Snow and Brazzel, 1965; Green and Thurston, 1971).

Reports also indicate that wild hosts, principally wild geranium and

winter legumes, support the first larval generations with subsequent

generations developing on cultivated crops (Brazzel et al., 1953; Snow

and Brazzel, 1965).

Losses resulting from damage to cultivated crops in the United

States by H. zea alone were estimated in the hundreds of millions of

dollars by Hardwick (1965). If H. virescens is included and one con-

siders the high levels of pesticide resistance which have developed in

recent years (Harris et al., 1972; Clower et al., 1975) the losses may

now have doubled.

Although the principal and preferred food plant of H. zea is corn

(Brazzel et al., 1953; Hardwick, 1965; Snow and Brazzel, 1965) extensive

damage is also incurred by cotton, tomato, and leguminous plants such as

peas. Damage to cotton and tomato crops is, in fact, of such magnitude

that on these crops H. zea is referred to as the cotton bollworm and

tomato fruitworm, respectively. Although H. virescens has a somewhat

reduced host range [Brazzel et al. (1953) indicated its complete

absence from corn over a 2-year study], it is nonetheless of immense

importance as a pest of both cotton and tobacco, and is very likely of

more importance on cotton than H. zea due to its genetic propensity for

the development of insecticide resistance (Harris et al., 1972; Clower

et al., 1975). It is probably safe to consider H. zea and H. virescens

combined as the most serious pest threat to cotton in the United States


Heliothis subflexa has apparently not been recorded as a pest

species, and unlike its close relatives has a more confined diet,

feeding solely on species of the genus Physalis (Brazzel et al., 1953;

Roach, 1975).

Reproductive Biology: Multiple Mating

As among other noctuid species (Byers, 1978) these 3 Heliothis

species commonly engage in multiple mating (Callahan, 1958a; Hardwick,

1965; Hendricks et al., 1970), the propensity for which increases with suc-

cessive generations in laboratory stocks (Proshold and Bartell, 1972;

Raulston et al., 1975). Levels of multiple mating in natural populations

of H. zea have been found to be 41.7% (Callahan, 1958a), and 33.8%

(Hendricks et al., 1970), while a slightly higher value of 45.9% has been

reported for H. virescens (Hendricks et al., 1970). Although no data are

available regarding natural supernumerary matings in H. subflexa, labora-

tory tests conducted by Proshold and LaChance (1974) and Pair et al. (1977)

indicate that such are common in this species also. Of the several

benefits attributed to multiple mating in Lepidoptera (see Byers, 1978)

2 appear important in these heliothids. First, as females are apparently

capable of assessing the genetic adequacy of an initial mating (Proshold

and LaChance, 1974; Pair et al., 1977) and as mated females are less

receptive than are virgins to additional mating for a period of ca. 3

days after an initial mating (Raulston et al., 1975), multiple mating

ensures that a high percentage of eggs are fertilized, thereby limiting

the waste of gametes. Second, multiple mating increases the genetic

diversity of the offspring of a single female thereby increasing the

phenotypic variation and potentially increasing the progeny's fitness.

Interspecific Hybridization

Although broadly sympatric with both H. virescens and H. subflexa,

H. zea is effectively isolated from both by structural differences in

the genitalia (see Callahan 1958a; Callahan and Chapin, 1960; Hardwick,

1965, 1970). It should not, however, be interpreted that attempted

mating between these species does not occur, because Callahan (unpub-

lished but cited by Hardwick, 1965) has found one pair of H. zea and H.

virescens locked in copula in the field, and Shorey et al. (1965)

reported interspecific copulations with similar results in laboratory

cross-mating studies under no-choice conditions. Further, unlike H.

subflexa and H. virescens (see below) H. zea is mechanically isolated

from its much closer relatives, Helicoverpa (Heliothis) armigera (Hbn)

and Helicoverpa (Heliothis) punctigera (Wllgn) (Hardwick, 1965). In

fact, of the 115 no-choice hybridization tests carried out between H.

zea and the other 2 helicoverpids, only a single pair (H. armigera

females X H. zea males) produced fertile eggs and viable offspring and

of these the great majority were infertile (Hardwick, 1965).

Unlike H. zea the genitalia of H. virescens and H. subflexa are

sufficiently similar to pose no mechanical barrier to interspecific

hybridization. Although early attempts to hybridize these species

failed (Brazzel et al., 1953), Laster (1972) was able to obtain off-

spring from H. subflexa female X H. virescens male crosses. Subse-

quently Proshold and LaChance (1974) succeeded in obtaining progeny from

both H. subflexa female X H. virescens male and H. virescens female X H.

subflexa male crosses. Hybrid features of the former cross include: a

slight sexual dimorphism in larval development times (Laster, 1972), the

entry of ca. 40% of the female pupae into an extended diapause (Proshold

and LaChance, 1974), reduced mating, production of fewer eggs, and of

major importance, the production of essentially but not fully sterile

males (Laster, 1972; Proshold and LaChance, 1974). Interestingly,

although male sterility is maintained in successive backcross gener-

ations with H. virescens males (Laster et al., 1976), backcrossing with

H. subflexa males restores fertility within 2 generations (Karpenko and

Proshold, 1977). Distinct features of the reciprocal cross (H. virescens

female X H. subflexa male) are much less obvious as none enter diapause

and mating occurred as often as among parental stocks in studies carried

out by Proshold and LaChance (1974). Further, although infertility

occurred in both sexes it was not complete and fertility could be

restored with successive backcrossing to either parent (Proshold and

LaChance, 1974; Karpenko and Proshold, 1977). In both hybrid stocks the

number of chromosomes is reduced from the 31 pairs found in the parental

species to 20-28 bivalents (Proshold and LaChance, 1974).

Proshold and LaChance (1974) contended that the prominent male

sterility within both hybrid groups was the result of an inability to

transfer eupyrene sperm, perhaps due to chromosome desynapsis. Sub-

sequently, it was found that the number of eupyrene sperm bundles

transferred was similar to those transferred in conspecific matings by

the parent species, and that the sterility resulted from the inability

of the sperm bundles to break up in the spermatophore and from abnor-

malities in the axial filaments of the sperm (Proshold et al., 1975;

Richard et al., 1975). Further, as males from backcross generation 35

do not show chromosome desynapsis, it is unlikely that chromosome

deviations are the cause of sterility (Karpenko and Proshold, 1977).

Interspecific Variations

Although Sluss et al. (1978) found low interpopulation variability

in allozyme allele frequencies of H. virescens, there is probably con-

siderable interspecific variation, particularly in behavioral characters,

due to the broad range and high adaptability of this species. The best

evidence for this comes from differences observed between laboratory

stocks (Laster et al., 1977), laboratory and wild strains (Raulston et al.,

1975; Raulston et al., 1979), and hybrid laboratory X wild stocks (Young

et al., 1975). Major characters involved include: mating propensity,

oviposition periods and reproductive periods. In fact the 2 h difference

in mating periods between Texas laboratory-reared and Virgin Island wild

stocks was of sufficient magnitude to effectively isolate these 2


Evidence concerning populational differences in H. zea is more

concrete. Hardwick (1965) has noted definite morphological differences

between the Hawaiian H. zea and continental populations and there is

evidence of genetic variation among natural populations (Sell et al.,

1975) and laboratory and wild populations (Sluss et al., 1978). Al-

though Sluss et al. (1978) speculate that differences between laboratory

and natural populations result from genetic drift within a small iso-

lated laboratory colony, the observations of Raulston (1975) indicate

that such differences can occur within 2 generations and are more

probably due to environmental parameters.

Mating Behaviors and Reproductive Periods

As among other noctuid species (Shorey et al., 1968; Teal et al.,

1978), H. zea and H. virescens do not become sexually mature immediately

after emergence. Although no data are available for H. subflexa, the

same condition probably exists. In a behavioral study Callahan (1958b)

found that female H. zea began sexual activity on the 2nd night after

emergence, while Shorey et al. (1968) found that a substantial per-

centage mated on the 1st night. Both studies indicated that males were

capable of copulation on the 1st night postemergence. Data concerning

H. virescens is much more obscure. Although Shorey et al. (1968)

reported females mating on the 1st night after emergence, Hendricks

and Tumlinson (1974) found females actively calling on the 2nd night,

and Gaston and Shorey (1964) reported maximum sexual attraction occurred

on the 4th night, while Gentry et al. (1964) found females unattractive

until 5 days old. Although pheromone release is certainly not an

absolute indication of reproductive maturity and ability to mate,

pheromone release does not generally reach its maximum until gametes are

mature (Shorey, 1974). The differences reported by the above workers

may be the result of intraspecific differences resulting from laboratory

colonization. Male H. virescens reach reproductive maturity on the 1st

or 2nd night after emergence (Gentry et al., 1964; Shorey et al.,


The behaviors and postures assumed during pheromone release by H.

zea and H. virescens have been described by Callahan (1958b), Agee

(1969), Hendricks and Tumlinson (1974), and Mitchell et al. (1974).

Agee (1969) also indicated that females wiped their ovipositors on the

substrate, an act which has also been reported in the pink bollworm

(Leppla, 1972) whereby females deposit pheromone on the supporting

surface (Colwell et al., 1978a).

Male precopulatory behaviors have apparently been studied sequent-

ially only in H. zea, and even then only in small cage studies (Callahan,

1958b; Agee, 1969), although the criteria used in bioassay experiments

by Shorey and Gaston (1965), McDonough et al. (1970), and Hendricks and

Tumlinson (1974) are obviously involved in the ethology of mating. In

H. zea the sequence involves: antennal movement and wing fluttering,

circular ambulation, flight, extension of the claspers, landing and an

ambulatory approach close to the side of the female, continued wing

vibration, and a grab for the female ovipositor which terminates in

copulation (see Callahan, 1958b; Agee, 1969).

The diel mating periodicity of the 3 species has been studied both

in the laboratory (Callahan, 1958b; Gentry et al., 1964; Shorey and

Gaston, 1965; Hendricks and Tumlinson, 1974), and field (Goodenough and

Snow, 1974; Hendricks and Tumlinson, 1974; Mitchell et al., 1974;

Raulston et al., 1975, 1979; Tingle et al., 1978). These studies indi-

cate the diel periodicities as follows: H. zea 1-10 h in the scotophase

(peak at 4 h), H. virescens 3-9 h (peak at 6 h), H. subflexa 1-8 h (peak

at 4 h). In addition, studies by Raulston et al. (1975) indicate that

mated females containing sperm have a peak period of mating that occurs

1 h later than that of virgin females. A further study (Raulston et

al., 1976) has also shown a difference of 2 h between a wild strain from

the Virgin Islands and a laboratory strain from Texas.

Sex Pheromone Glands and Production

The sex pheromone gland of both H. zea and H. virescens has been

described as a complete ring of collumnar and cuboidal class I epidermal

gland cells (see Noirot and Quennedey, 1974) situated in the interseg-

mental membrane between abdominal segments 8 and 9 (Jefferson et al.,

1968). Unlike H. zea, which shows a rapid production of pheromone prior

to emergence, H. virescens emerges before any appreciable amount of

pheromone is produced (Shorey et al., 1968). The rise in the pheromone

concentration is, however, of such magnitude that by the Ist night after

emergence both H. zea and H. virescens produce about the same amounts

(Shorey et al., 1968).

The first evidence of the production of a volatile sex pheromone by

female H. virescens was reported by Gentry et al. (1964) using extracts

of 5-day-old females and abdomen tips. Similarly, crude abdomen tip

extracts of both H. virescens and H. zea elicited male ambulatory and

flight behaviors in laboratory bioassays. Berger et al. (1965), failed

to obtain male responses from either crude extracts or airborne col-

lections, and suggested that, as male responses were obtained from gas

chromatographic (GC) fractions, crude extracts were masked by the

presence of inactive contaminants within the gland. This masking was

subsequently negated by Shorey and Gaston (1967) in tests using both

crude extracts and GC fractions. Interestingly, the extracts used in

Shorey and Gaston's study failed to elicit close-range copulatory

behaviors, while excised ovipositors did, indicating that perhaps the

complete pheromone blend was not extracted.

Sex Pheromone Chemistry

McDonough et al. (1970) were first to report the isolation and

identification of behavior-influencing chemicals from H. zea. Utilizing

the same bioassay techniques used in Shorey's studies these workers

concluded that (E)-7-tetradecen-l-ol acetate and (E)-7-tetradecen-l-ol

were active components and indicated that several 14-carbon alcohols and

acetates could also be involved. As discussed by Mayer and McLaughlin

(1975), however, evidence for their actual role as sex pheromones is

tenuous. (Z)-9-Tetradecen-l-ol format, a compound subsequently found

to be an effective minic of one of the actual components of the phero-

mones of both species [(Z)-ll-hexadecenal] (Mitchell et al., 1975,

1978), was postulated as a pheromone of H. zea and, in combination with

(E)-9-tetradecen-l-ol, the pheromone of H. virescens (Jacobson et al.,

1972). Subsequent field testing (Hendricks, unpublished but cited by

Tumlinson et al., 1975; Mitchell et al., 1978), however, indicated that

neither component was as effective as crude abdominal washes. Sub-

sequently, Roelofs et al. (1974) isolated and identified (Z)-ll-hexa-

decenal and (Z)-9-tetradecenal as pheromone components of H. virescens

and (Z)-ll-hexadecenal from H. zea. Tumlinson et al. (1975) confirmed

the presence of (Z)-ll-hexadecenal and (Z)-9-tetradecenal in H. virescens,

finding a ratio of 16:1 within the insects. However, because synthesized

mixtures of the H. virescens pheromone were not as effective as crude

abdominal extracts in attracting males in field cage tests these workers

speculated that other chemicals were also produced.

The presence of other chemicals in the gland of both H. zea and H.

virescens has indeed been reported recently (Klun et al., 1980a,b).

According to Klun et al., in addition to (Z)-ll-hexadecenal traces of

(Z)-9-hexadecenal, (Z)-7-hexadecenal, and hexadecanal are present in the

pheromone of H. zea, while in addition to these chemicals the pheromone

of H. virescens contains minute amounts of tetradecanal, (Z)-9-tetra-

decenal, and (Z)-ll-hexadecen-l-ol. As trap catch is significantly

increased by the presence of these compounds in baits (Klun et al.,

1980a,b), certain of these compounds are undoubtedly principal com-

ponents of the pheromones of these species. Bioassays used by both

Roelofs et al. (1974) and Tumlinson et al. (1975) relied heavily upon

sustained flight and other cues attributed to primary pheromone com-

ponents (see Roelofs and Carde, 1977); thus, it seems probable that the

newly identified compounds are secondary, or close-range components.

Unfortunately, it is unknown if all or just certain of these chemicals

are necessary for maximum attraction and reproductive isolation between

these 2 species. To date none of the pheromone components of H. subflexa

have been reported.

Interspecific Attraction and Stimulation

Early evidence based on behavioral studies indicated that con-

siderable cross-attraction occurred between H. virescens and H. zea

(Shorey et al., 1965). Cross-attraction was so prominent in the above

study that the authors considered mechanisms other than pheromone to be

involved in reproductive isolation. All studies on the pheromone

chemistry of these species have shown the presence of at least one

common component. Interestingly, studies on the oldworld bollworm, H.

armigera, also indicate the presence of (Z)-ll-hexadecenal and (Z)-9-

hexadecenal (Nesbitt et al., 1979, 1980) common to the pheromones of

both H. zea and H. virescens and perhaps (Z)-9-tetradecenal in common

with H. virescens (Gothilf et al., 1978). The use of common chemical

components by closely related species is well known (Roelofs and Carde,

1974; Carde et al., 1977), and appears to be of considerable value in

phylogenetic studies (Roelofs and Comeau, 1969). Among tortricid moths,

for which considerable amounts of data on the component makeup of many

species are available, many species use the same chemical compounds but

rely on component ratios as a mechanism of isolation (Carde et al.,

1977). As indicated by the work of Klun et al. (1979), specificity

within these heliothids does not rely solely upon component ratios, but

also on the sequential addition of minor components to the pheromone

blend. Because males invest considerable amounts of energy during

upwind orientation, one draws the conclusion that certain of the trace

chemicals in the H. virescens pheromone might inhibit upwind flight by

male H. zea. Further, as the presence of virgin female H. zea and H.

virescens in the same electric grid trap significantly reduces male

attraction in both species (Haile et al., 1973), it is apparent that,

when in close association, the pheromones of both species or appro-

priately combined ratios thereof are inhibitory to males. Similar cases

of interspecific inhibition among other noctuids have been described by

Steck et al. (1977).

Semiochemicals for the Control of H. zea and H. virescens

Surprisingly, in view of our limited knowledge of the mechanics of

pheromone communication in these species, semiochemicals play major

roles, both directly and indirectly, in many control programs.

The use of both virgin females and synthetic attractants in popu-

lation monitoring is common (see Goodenough and Snow, 1973; Roach,

1975). Although there is no literature available on extensive mass-

trapping using the pheromones of H. zea and H. virescens it has been

alluded to (Mitchell et al., 1974). Currently, the most promising use

of semiochemicals in the direct control of these species is in mating

disruption. In studies on multi-species disruption using (Z)-7-dodecen-

1-ol acetate (pheromone of Trichoplusia ni (HLbner)) and (Z)-7-hexadecen-

1-ol acetate (an attractant of the pink bollworm), Kaae et al. (1972)

achieved considerable success in the disruption of both H. zea and H.

virescens, particularly with looplure. Also, recent studies based on

electroantennogram data (Priesner et al., 1975; Mitchell et al., 1975,

1976, 1978) have shown that atmospheric permeation using (Z)-9-tetra-

decen-l-ol format is a highly effective disruptant for both species.

Further, Mitchell et al. (1978) have shown that (Z)-ll-hexadecenal is

equally effective in disruption and conclude that the format probably

substitutes for the aldehyde at the level of the antennal receptor.

Interestingly, (Z)-7-dodecen-l-ol format, a very similar chemical to

(Z)-9-tetradecenal, is not effective in disruption, and when H. virescens

females are used as bait in an area permeated with this format con-

siderable numbers of H. zea males are trapped (Mitchell et al., 1978).

Because (Z)-9-tetradecen-l-ol format has been used effectively with

mating disruptants of other species it promises to be of considerable

value in the future (Mitchell et al., 1976).

In addition to the direct use of semiochemicals in the control of

both H. zea and H. virescens, programs using the sterile-release tech-

nique (see Knipling, 1970) rely heavily upon the attraction and sub-

sequent mating of native and released insects. Recently, based on the

inherited sterility of backcross progeny resulting from H. subflexa

female X H. virescens male hybrids, Laster et al. (1976) and Parvin et

al. (1976) have proposed such a program for the control of H. virescens.

Although an impressive and potentially very effective control proposal,

the original Laster-Parvin model contained several inconsistencies that

have been discussed and corrected by Makela and Huettel (1979). From a

chemical communication standpoint several factors appear unresolved in

the Laster-Parvin model. Although an investigation by Laster et al.

(1978) indicated that backcross (BC) and H. virescens females were

equally effective in the long-distance attraction of H. virescens

males, no comparison was made with virgin wild females, nor were close-

range copulatory behaviors critical to mating studied (see Roelofs and

Carde, 1977). Under natural unrestricted conditions wild males may

exhibit considerable choice favoring wild virgin females due to: 1)

differences in the pheromonal bouquet of laboratory-reared stocks due to

dietary differences (see Hendricks et al., 1977), differences in the

place of origin of the laboratory stocks (see Roelofs and Carde, 1974;

Laster et al., 1977), or differences resulting from hybridization which

effect close-range copulatory behaviors; 2) the evolution of close-range

courtship behaviors in laboratory stocks that differ significantly from

those of wild stocks due to high levels of selection resulting from

laboratory colonization (see Raulston, 1975; Chambers, 1977; Sluss et

al., 1978). Although many of these factors appear to have been tested

by Raulston et al. (1979), these authors indicate that their tests were

not carried out in the peak period of virgin mating and, as mated

females are less active sexually for some time after mating, the con-

tinuously active males may have settled for the next best thing, the BC

females (a similar theory was proposed for the attraction of males to

pheromone traps by these ijd'hors). Raulston et al. (1979) also indi-

cated that BC males were not competitive with H. virescens males in

mating with wild females, a point which supports the above and indicates

that further research into pheromonal communication between these

insects is warranted.

Research Aims

Obviously there are inconsistencies and gaps in our knowledge of

the biology of H. virescens, H. subflexa, and H. zea. Authors, even

within the same research group, report conflicting results, but never-

theless studies appear to take for granted basic facts that have not

been proven conclusively. This is all too evident in studies on phero-

mone communication among these species. For example, a total of 7

compounds have been identified as behavioral-modifying chemicals for H.

zea by 4 groups of researchers, 9 compounds for H. virescens, and none

for H. subflexa. Additionally, for those which have been identified

from the pheromone glands we have no idea which are involved in upwind

amenotaxis and which are important in close-range precopulatory behav-

iors, or in fact which are actually released as volatiles from the gland

surface. Further, we have no conception as to the effect of hybridi-

zation on pheromone-induced behavior within the group, and, as control

programs are fast moving toward the use of sterile hybrids in control,

such knowledge is a necessity.

The present study was undertaken to evaluate several aspects of the

sex pheromone mediated biology and chemistry of H. virescens and H.

subflexa in an attempt to provide the basics necessary for the develop-

ment of effective semiochemical control programs and to assess the

effects of sex pheromones on the reproductive isolation of these species.

In order to accomplish these goals the following studies were

undertaken: 1) an analysis of the reproductive behavior of H. virescens

under laboratory conditions; 2) an investigation of the sites of sex

pheromone production in the ovipositor of H. virescens; 3) a behavioral

and electrophysiological analysis of extracts obtained from these sites;

4) a behavioral analysis of the semiochemically induced interactions

occurring between H. virescens and H. subflexa; and 5) the identifi-

cation of a sex pheromone of H. subflexa and assessment of different

blends of its components in field trapping studies.



Due to its genetic propensity for developing resistance to in-

secticides (Harris et al., 1972; Clower et al., 1976), H. virescens has

become the subject of numerous studies using potential alternative

methods of pest suppression. One such method, population reduction by

infusion of hybrid sterility into natural popualtions, shows consider-

able promise, being both environmentally safe and effective over ex-

tended periods of time (Laster et al., 1976). The success of this

approach depends upon random mating between native populations of H.

virescens and introudced backcross individuals obtained through labora-

tory hybridization of H. virescens and its sibling species, H. subflexa.

Among sympatric, closely related species of Lepidoptera, repro-

ductive isolation occurs prior to gamete investment, being controlled to

a great extent by sex pheromones and related behaviors (Roelofs and

Carde, 1974). Although early studies suggested the use of a single

chemical component pheromone (Butenandt et al., 1959; Roelofs and Arn,

1968), recent work has indicated that few, if any, Lepidoptera rely upon

such simple systems. In addition to the species specificity imparted by

multicomponent pheromone blends and ratios acting at a distance (Carde

et al., 1977), it appears that certain minor components acting either

alone or in concert with other components are responsible for maximizing

the behavioral events comprising courtship (Roelofs and Carde, 1977).

Hence it is of utmost importance that the sequence of behavioral events

required for successful intraspecific mating be documented prior to the

ascribing of roles to the pheromone components or attempting to enhance'-

interspecific mating success via pheromonal means.

Although courtship studies have been made on H. zea (Callahan,

1958a,b; Agee, 1969), these studies were not primarily concerned with

the role of sex pheromones, and as shown elsewhere (Grant and Brady,

1975; Grant et al., 1975), behaviors required for successful mating by

one species cannot be extended to other closely related species. This

study reports the results of laboratory studies on the precopulatory

behaviors of H. virescens as a prelude to work on the reproductive

interactions between H. virescens and H. subflexa.

Methods and Materials

Rearing and Adult Holding

Heliothis virescens used for all studies were obtained as pupae

from laboratory stocks maintained at USDA facilities in both Oxford, NC,

and Stoneville, MS. All insects were allowed to emerge under test con-

ditions and in isolation from members of the opposite sex. Newly

emerged adults, collected daily, were housed in 30 X 30 X 30 cm plexi-

glass cages for at least 2 days prior to testing to ensure reproductive

maturity. All cages were provided with a 10% sucrose solution for

nutrient. The upper limit for testing was set at 8 days postemergence.

Tests were conducted during the dark phase of a reverse 16:8 light:dark

cycle at a temperature of 20(+1)C, and relative humidities of 54(+2)%

during the day and 60(+2)% during the night. Test individuals of both

sexes were selected on the basis of good appearance and physiological

vigor, as indicated by sustained flight after being startled during the

photophase preceding each test.


Two assay systems were employed to provide data on both precourt-

ship and close-range courtship behaviors. The 1st, used in the analysis

of male activation, orientation, and initial analysis of courtship

interactions, consisted of a 1.5 X 0.5 X 0.5-m plexiglass wind tunnel

through which air was pulled at a constant rate. Groups of 5 females

were placed on a tobacco plant in the upwind end of the tunnel during

the premating period of the scotophase (Tingle et al., 1978). Female

behavior was monitored throughout the mating period. Individual males

were placed in release cages positioned above the body of the tunnel and

10 cm from the downwind end at least 15 min prior to being lowered into

the tunnel. The largest possible pheromone plumal area was estimated

with titanium tetrachloride fumes emitted from cotton wicks positioned

to approximate the outermost edges of the tobacco plant. A light level

of 2 lux, to which the insects had been entrained during holding, was

found to be sufficient for observations when augmented with a diffuse

red-filtered flashlight during close-range studies. The comments of an

observer were recorded on audio-cassette tapes and later transcribed.

The 2nd assay system was used to observe courtship interactions and

consisted of a 20 X 10 X 10-cm plexiglass cage having a 4-cm2 door at

one end to admit insects. Groups of 3-4 females were placed in the

chamber during the photophase preceding each test, and males were

released individually when females had been observed calling for a 5-min

period. Behaviors were recorded on video-tape using a JVC cassette

recorder and a Panasonic WV-10500 camera with spectral sensitivity

between 400-800 nm (minimum light intensity for camera sensitivity = 0.5

lux), a neutral density remote-control zoom lens and a remote-control

pan-tilt device. Tapes were transcribed from a television monitor using

both stop action and standard speed modes.

Statistical Anlysis

Frequencies of the observed behaviors were tabulated in Ist order,

preceding-following, behavior transition matrices, and comprehensive

ethograms were devised. Chi-square values of all cells with probabil-

ities >0 were calculated according to common techniques (Stevenson and

Poole, 1976). Since a successful mating included only a single approach

and clasp, and since self-perpetuated acts were considered single

events, X2 comparisons were based on 56 degrees of freedom. Individual

transitions with observed frequencies greater than the expected value

were considered significantly greater than chance if (observed ex-

pected) (expected).5 values were greater than (x20.05 56df).5 562

(Bishop et al., 1975; Fagen and Young, 1978). Standard normal deviates

were calculated for successful courtship sequences and applied to a

binomial test for individual transitions (Siegel, 1956). Transitions

with deviates yielding probabilities of p <0.05 were used in construc-

tion of an ethogram to indicate the most probable courtship interaction

sequences resulting in successful mating.

Observations and Discussion

Female Precourtship Behavior (N = 75)

In common with other species of Lepidoptera (Agee, 1969; Fatzinger

and Asher, 1971; Barrer and Hill, 1977; Teal and Byers, 1980), distinct

phases of calling (sex pheromone release) were obvious in our study.

One phase, indicating the initiation of female sexual activity, was

marked by short bouts of calling separated by periods of ambulation,

wing fanning, and flight. Although the glandular surface was normally

retracted during interruptions, some females were seen with their ovi-

positors extended during periods of flight and ambulation. The 2nd

phase of calling was preceded by a period of scent marking (p = .73)

during which time females dragged the ventral portion of the ovipositor

on the supporting substrate. As males will land and search a scent-

marked area, often exposing their hairpencils, after removal of the

female, such pheromone deposits appear to aid in the chemical elici-

tation of male reproductive behaviors. Following this, females entered

a period of quiescent calling during which they remained relatively

stationary and held their wings in a "V" at ca. 450 to the body. Unlike

female H. zea (Agee, 1969), wing vibration by female H. virescens during

this time was sporadic and of short duration (p = .08) in the small

populations tested.

Interestingly, these 2 phases of calling are fairly well correlated

with sexual receptivity. Females calling for short sporadic bouts

actively reject males that are committing appropriate behaviors early in

the courtship sequence. However, females calling for prolonged bouts

tend to reject males because of incorrectly committed male behaviors in

the latter part of the sequence (Table 1). Because a male bias in the

operational sex ratio probably exists under natural conditions (Teal et'-

al., 1978), rejections during the initial phase of calling would pro-

bably not affect an individual female's chances for successful mating on

a given night. Further, as the mating periods of H. zea and H. subflexa

[both known to enter into interspecific copulation with H. virescens in

the laboratory (Shorey et al., 1965; Laster, 1972; and see Hardwick,

1965 for a naturally occurring case)], do marginally intersect the early

part of the H. virescens calling period (Tingle et al., 1978), the

reduced sexual receptivity during early discontinuous calling bouts may

serve to isolate female H. virescens behaviorally during chance en-

counters with related species.

As a result of sporadic calling during the early bouts, effective

chemical signaling occurs primarily during the prolonged quiescent

calling period. This is quite different from that of many other noctuid

species, which exhibit discontinuous patterns of calling throughout the

mating period (Sower et al., 1971; Marks, 1976; Swier et al., 1977).

Since constant shifting by calling females coupled with stationary scent

marks would tend to confuse orienting males, it would seem advantageous

for females to remain within a restricted area during calling and

enhance the orientation signal by both calling and depositing a rela-

tively concentrated scent mark. Support for this hypothesis comes from

the calling behaviors of 2 other species commonly occurring at high

population densities, H. zea and Pectinophora gossypiella (Saunders),


o u

o a)



0 ..
r-- O
C 5-


E 0-
cl E


J 0

4- r-
E 0



4- (fl

00 r-

(a a) a) a) a)

.E I E E

.. L.. 2: L.. U.. U..

) o
- 0
C; C:,

c- -

)3 a 4

S = C CF
0 a 0- (

0. > *r- >
0- 0 ad 0
a r

X 4
c) U
Sc S a

which call for extended periods near scent marks (Agee, 1969; Colwell et

al., 1978a,b).

Male precourtship behaviors (N = 76)

Of 76 males exhibiting reproductive behaviors in the 1.5-m flight

tunnel, 44 began from rest and committed common pheromonally mediated

activation behaviors (Fig. 1) (see Bartell, 1977). The remainder were

active prior to or become active while being lowered into the plume. A

possible reason for immediate activity in our tests was movement during

and prior to release. However, a constant basal activity level of ca.

20% has been noted in bioassay studies with synthetic and glandular

odorants (Teal et al., unpublished data). The existence of a naturally

occurring, non-mating activity period is common among many moth species

(Carde et al., 1975; Baker et al., 1976; Marks, 1976), and judging from

disparities between time intervals for significant trap captures using

blacklight and female-baited traps (Graham et al., 1964; Goodenough and

Snow, 1973), such a premating activity period is indicated for H.

virescens. Hence, it is probable that a major portion of the natural

male population encounters pheromone stimuli while in flight (Sparks et

al., 1979), and the significance of the various activation behaviors

listed below are therefore indeterminant. Although wing elevation,

vibration, and antennal movement were similar to those described for H.

zea (Agee, 1969), no H. virescens males were seen extending and wiping

their claspers on the substrate prior to flight. Antennal cleaning with

the epiphysis of the foreleg (Callahan and Carlysle, 1971) and extension

Figure 1. Precourtship reproductive behaviors of initially inactive
male H. virescens (n = 44). Males are not necessarily
successful during courtship. Values indicate probabilities.

Inactive on Release

Female Produced

Non-directed Flight Directed Flight
(More than 30 Sec.)--- up plume

Diminished Rate of
Forward Advance

.1 .I 375 .01 2
Partial Hairpencil


of the proboscis were also common (p = 0.5), as was a period of random

ambulation during preflight wing fanning (p = 0.64). Although tnese

activation behaviors are of benefit in analysis of interspecific phero-

mone communication, I have, for the reason above, not elaborated them

for analysis (Fig. 1).

Initially active moths underwent a period of erratic flight,

bumping into the walls, moving past calling females (sometimes within 4-

5 cm), and stopping frequently. Preceding upwind taxis, these moths

generally performed a series of horizontal and vertical sweeps through

the downwind 1/2 of the plumal area. A similar period of aerial search-

ing was also observed during 55% of the tests involving initially

inactive males. In common with other species (Kennedy and Marsh, 1974;

Farkas and Shorey, 1974), flight upwind toward calling females was

marked by a series of tightening zigzags approximating plume width. At

a distance of 5-15 cm from the calling female, the rate of forward

advance was distinctly slowed (p = 0.89), and in some instances halted

momentarily as males entered a "choice" situation. At this point males

made the "decision" to continue or to reorient from downwind. Continu-

ance in all cases was marked by a momentary hairpencil exposure in which

the hairpencils were extended to ca. 1/2-3/4 of their length and landing

from 2-4 cm to the lower left (p = 0.46) or right (p = 0.54) of the

female. As males undergoing random flight or aerial searching rarely

entered into courtship during encounters with receptive females (p =

.03), a period of close-range tactic behavior, during which males per-

ceive sufficient stimuli to release continued reproductive behaviors,

appears to be necessary. This close-range male orientation and dis-

crimination phase is apparently necessary for reproductive isolation

because, although male H. virescens will move toward calling H. subflex-

females from a distance, the majority are incapable of either finding

females when at close range or completing courtship.

Courtship Interactions

Although flight tunnel assays were useful in outlining the more

obvious steps in male courtship, a considerable number of male-female

interactions went unobserved, necessitating the use of video-tape

studies. Video analysis of 30 successful matings revealed a highly

variable series of interactions (Fig. 2, Table 2), in which only one

transition with an observed frequency greater than the expected result

(female wing fanning partial hairpenciling) was not significantly

greater than that expected by chance. However, results of a binomial

test of individual transitions indicated that a common trend was present

(Fig. 3). A male commonly began the sequence by approaching and moving

under a female's wing, but unlike H. zea (Agee, 1969) did not antennate

the ovipositor. At this point, the female would normally fan her wings.

The male continued his forward advance, partially exposing his hair-

pencils in the process, until his head was approximately parallel with

the female's. Having achieved this position, the male commonly turned

to face the female and, in 45% of the matings, the female also turned

toward the male. At this point, the male hairpenciled and clasped the

female's genitalia. Interestingly, male wing fanning, a mechanism used

Figure 2. Comprehensive ethogram of 30 successful matings. Values
indicate probabilities of performance of a pair of
behaviors sequentially.

-Female Calling




4 clasp
S +

LC C a a








o c


E o

*- -I










- a a a a a a adh

a no aaa aa a


O a a a a g I a

.~ a I II

'" --i g
aI a a -

"- "XF B S u S *

aL ID .C C
- 'Cx
O 0 i 'Ca z 'C
a 'C '0 C 0 '
CC IiC 0 C C C-'

C'Ftlol a




N Y)
01 II~~1INO


g O


"a a~
xN D (Y O ~ O O XN


"~ :O i
PX 'Z~; S r' 9

1 1 -

~SX E""

"" ""~""~~"

""" """
o~~o o~or







OO d r

O l l l O l l l

Figure 3. Most frequent behavioral paths in successful copulations
based on standard normal deviates having probabilities
<0.05 in a binomial test. Male wing fanning is offset
as it does not appear to interact directly with other
male behaviors. Values are probabilities given in
Fig. 2.

afApproach ?


oMoves under ? wing

27 cWing fan- ? Wing fan

crPartial hairpencilling

23 .33

aeMoves parallel with ?


eClasp .3

for scent dispersion of male Grapholitha molesta (Busck) (Baker and

Carde, 1979) and Plodia interpunctella (Hbbner) (Grant and Brady, 1975),

was most often the response to female fanning. Since fanning directs

air to the posterior of the male (Baker and Carde, 1979) and away from

female H. virescens, it does not appear to serve in scent dissemination

in this case. More probably, male fanning is merely the result of

female contact and the male's excited behavioral state.

Copulatory Behaviors

Among many lepidopteran species, successful genitalic clasping is

followed by an immediate shift to a "heads away" position (Agee, 1969;

Grant and Brady, 1975; Barrer and Hill, 1977). However, the immediate

shift (i.e., <10 sec) among engaged H. virescens pairs occurred in only

45% of the copulations and was obviously a result of attempted escape by

the female. Although both sexes appeared to effect the turn, the major

benefit is apparently to the male, as it enables him to extend his

hairpencils fully. This is the only time when the hairpencils are fully

exposed and pulsed in and out. Since this behavior continued until the

female acquiesced, I assume that the hairpencils release a secretion

which serves an arrestant function. Further, as females commonly

resumed calling and attracted males after courtship breakdowns in which

males partially or fully exposed their hairpencils (p = 0.67 both

assays), the effects of this secretion in inducing a cessation in

pheromone release (Hendricks and Shaver, 1975) are clearly short lived.

Courtship Breakdown and Reorientation

In both assay systems, a considerable number of the initial repro-

ductive encounters failed to terminate in successful mating (32/76

flight tunnel; 23/53 video studies), and although males in 72% of the

failures monitored in the flight tunnel reoriented, either from downwind

(p = 0.43) or via ambulation (p = 0.57), fewer than 20% were successful

in copulating on subsequent attempts. Although males may reorient

several times, only 54% of those undergoing ambulatory reorientation

after the 1st encounter attempted mating with the same female. Of those

reorienting from downwind, only 4 were successful in engaging a female

and only one succeeded on the next encounter. Video analysis of court-

ship failures involving 23 pairs of insects revealed that the majority

of breakdowns were due to the failure of a male to move parallel with

the female prior to attempted clasping and hairpencil exposure (Table

1), a failure which resulted in these behaviors being performed across

the female's abdomen. Females commonly responded by withdrawal of the

ovipositor, ambulation, and extensive wing fanning. Although failure on

the part of the male accounted for most breakdowns (Table 1), female

rejection during the approach, movement under the wing and movement

parallel phases was also common (p = 0.31). These rejections were the

result of encounters involving females which called for short, sporadic

bouts. From these data I suggest that the absolute prerequisites for

successful mating include: 1) female quiescence during male approach,

2) the male's ability to move parallel with the female prior to hair-

penciling and attempting to clasp, and 3) relative quiescence on the

part of the female during the clasp attempt.

Behavioral Releasers

The induction of particular reproductive behaviors results from

various species-specific cues that cause behavioral release (Grant and

Brady, 1975; Baker and Carde, 1979). These cues range from self-in-

duction by the preceding behaviors of the same individual to cases in

which both inter- and intraindividual cues moderate transitions (Grant

and Brady, 1975; Barrer and Hill, 1977; Baker and Carde, 1979;

Castrovillo and Carde, 1980). As among other species, the primary

release of male reproductive behaviors is the female's sex pheromone

blend which, at a minimum, induces orientation. The 1st male-produced

cue appears to be a prelanding hairpencil display, which may reinforce

female quiescence. Male approach is probably induced by both the

presence of the specific pheromone bouquet and the female form. How-

ever, preliminary results with a model system similar to that of Shorey

and Gaston (1970) suggest that the chemical cue strongly outweighs any

visual input. Exposure of the male hairpencils during this period

presumably reinforces female quiescence. Upon coming parallel and

turning toward the female, the male's forelegs come in close approxi-

mation with the female's face. This sequence is of interest in view of

Callahan's (1969) description of scent-like scales on the male forelegs

of H. zea, and although Callahan found foreleg-amputated males capable

of mating, a chemical messenger cannot be ruled out. The final hair-

pencil display, delivered during the clasp, possibly functions in the

maintenance of female quiescence, although since a number of females

attempt escape, it is of limited effect.

As discussed by Grant et al. (1975) and Grant and Brady (1975), it

is not unlikely that closely related sympatric species sharing common

periods of reproductive activity can functionally maintain species

isolation solely on the basis of long-distance pheromones. This is

particularly true among such sibling species as H. virescens and H.

subflexa, which are capable of producing viable, semi-sterile hybrid

progeny. Although the reproductive behaviors of H. virescens are in

many respects similar to those of H. zea (Agee, 1969), several distinct

points have become apparent. Apart from the distinct pheromones of the

2 (Roelofs et al., 1974; Tumlinson et al., 1975; Klun et al., 1979),

which undoubtedly impart reproductive isolation in distance communi-

cation, several other points in the sequence appear to be of importance

in delivering species-specific cues.



Sex pheromone glands of female Lepidoptera are commonly found in

the intersegmental membrane between abdominal segments 8 and 9 (Percy

and Weatherston, 1974). However, cases have been reported in which the

glands are found elsewhere (McFarlane and Earle, 1970; Chow et al.,


As indicated in the previous chapter, studies on the reproductive

behavior of H. virescens implicated several anatomical areas as possible

sites of sex pheromone production. The present work discusses histo-

logical studies on sites of pheromone production in the terminal ab-

dominal segments of female H. virescens and extends and clarifies the

work of Jefferson et al. (1968).

Methods and Materials

Female H. virescens were observed during the initial 3 scotophases

after emergence to ensure that calling (pheromone release behavior),

indicated by ovipositor extension, had developed. Actively calling

females were removed from the observation cages on the 4th scotophase

and placed in a freezer (-100C) for 10 min prior to removal of the

ovipositor for microscopic examination.

The terminal abdominal segments were extended by applying pressure

to the abdomen, removed and pricked several times with a minute pin to

facilitate fixative infiltration prior to being placed in a dissecting

dish flooded with 2% gluteraldehyde in 0.1 M PO4 buffer (pH 7.2) and a

small quantity (2-3 drops) of Photoflow to reduce surface tension.

After 1 h, the material was transferred to fresh cold fixative (6C),

soaked overnight, and postfixed in 2% osmium tetroxide in 0.1 M PO4

buffer (pH 7.2) for 2 h. The tissue was then dehydrated in ethanol and

embedded in Spurr's resin. Both longitudinal and cross sections between

0.5 and 1.5 pm in thickness were cut with a glass knife. Sections were

mounted serially on gelatin-coated slides and were stained on a hot

plate (900C) with methylene blue in 1% borax.

Ovipositors to be used in scanning electronmicroscopy (SEM) were

removed using the above method and immediately placed in buffered 4%

osmium tetroxide for 6 h. The tissues were dehydrated through both

ethanol and ethanol-Freon series prior to critical-point drying from

Freon. They were then affixed to SEM stubs with silver conductive paint

and were sputter coated with gold. SEM observations were made using a

Cambridge Stereoscan MKIIA.

Results and Discussion

Examination of semi-thin sections revealed 2 morphologically

distinct areas of glandular tissue. The most extensive area of gland-

ular epithelia (GI in Figs. 4, 5) was situated in the intersegmental

membrane (Ism) between abdominal segments 8 and 9 + 10. A 2nd area of

glandular tissue (GII in Fig. 4) was found throughout the dorsal valves

papillaee annales).

The glandular tissue within the Ism was found to form a chevron,

similar to that of the sex pheromone producing gland of Plodia inter-

punctella (Hlbner) (Smithwick and Brady, 1977a). It extended from a

ventral position midway through the Ism to the junction of the Ism and

9th sternite and dorsolaterally to the region of the postericr apophyses

(Figs. 4, 5, 6). The glandular tissue was large in surface area and

showed considerable folding (Fig. 3); when exposed during calling it

formed a visible bulge over the ventrolateral 2/3 of the Ism (Figs. 4,

5). Jefferson et al. (1968) found cells from this glandular area to be

continuous over the dorsum, but I cannot confirm this. Epidermal cells

that extended between the posterior apophyses above the hindgut were

flattened, and had indistinct lateral cellular membranes. No vacuoles

characteristic of pheromone gland cells were present in these epidermal

cells (Fig. 7).

Cells composing the glandular tissue were similar to those found in

other noctuids (Jefferson et al., 1966; Percy, 1979; Teal and Philogene,

1980). They had large central to basal nuclei and ranged from columnar

over the extensive central area to cuboidal in the periphery (Figs. 6,

8). The cytoplasm was packed with vacuoles (Fig. 9). Because extracts

of the Ism from both calling and noncalling females elicited male sexual

responses (Chapter IV) and because similar vacuoles have commonly been

associated with cells producing sex pheromones in other noctuid species

(Jefferson et al., 1966; Jefferson and Rubin, 1973) it is probable that

these cells are production and storage sites for at least one pheromone


Gland cells within the dorsal valves (GII in Fig. 4) were found to

be composed of 2 distinct types (Figs. 10, 11, 12). The major portion

of the glandular epithelium was composed of columnar or cuboidal cells

Figure 4. Ventrolateral aspect of the terminal abdominal segments of
female H. virescens. VIII Segment 8, IX + X segments
9 + 10, Ism intersegmental membrane. DV dorsal valves,
S. setae, G.I glandular area I, GII glandular area 2.

Figure 5. Longitudinal section through the ovipositor. HG hindgut;
OD oviduct; M muscle; Ret retractor of Gl; Ep -
undifferentiated epidermal cells; VIII abdominal segment
8; Ism intersegmental membrane; IX + X abdominal seg-
ments 9 + 10; DV dorsal valves; GI glandular area 1.


Figure 6. Cross-section through GI area of a partially extended ovi-
positor. PA posterior apophysis; L limit of GI cells,
EP undifferentiated epidermal cells. Note fold in GI.

Figure 7. Undifferentiated epidermal cells (EP) overlying the hindgut
of Fig. 6. Fb fat body.

Figure 8. Columnar (Co) and cuboidal (Cu) cells forming GI.




6 1
mm &e

rE I~lk~i

&*I' il

similar in size and structure to the gland cells found in the Ism (Fig.

11). The 2nd type of gland was composed of tormogen and trichogen

cells, arranged in setiform glands and associated with tubular setae

(Figs. 12, 13).

The columnar and cuboidal cells contained a considerable number of

vacuoles, reminiscent of other pheromone producing glands. The dorsal

valves papillaee annales) of female Lepidoptera are considered to be

remnants of the terminal 11th and 12th abdominal segments (Matsuda,

1976), therefore this glandular area is considered to be distinct from

that found in the 8th intersegment and adduce the differential evolution

of each site. Inasmuch as the cuticle overlying pheromone producing

tissue within the Ism of H. virescens and other Lepidoptera (Percy and

Weatherston, 1974) has been found to be composed of loosely packed

cuticle having little sclerotization, the dense, heavily sclerotized

cuticle surrounding the glandular tissue of the dorsal valves was found

to be quite distinct. However, it is reasonable to consider that a

mechanism for pheromone release, perhaps similar to the epicuticular

filaments hypothesized by Percy (1974) as being pheromone storage and

release sites in Choristoneura fumiferana (Clem.), could function in

this instance. Such densely packed cuticle probably precludes the

cuticular storage of considerable amounts of pheromone such as the

amount stored by ]. ni (Percy, 1979). In fact, this cuticle may provide

a mechanism for the slow release of a sex pheromone because the movement

of pheromone through its heavy sclerotization may be slower than release

through the intersegmental membrane. Extracts from both sites elicited

Figure 9. Columnar gland cells of GI. N nucleus; V vacuole;
Ib apparent infolds of apical cell membrane.

Figure 10. Longitudinal section through the dorsal valves indicating
2 glandular types (TI, TII). Hg hindgut.


Il I 1 ^ "

I ; ,...-* L- 4-
P j" Gi j ^ h r ~ w : '

Figure 11. Section through columnar epidermal cells of the dorsal
valves. Note the heavily sclerotized cuticle (C). N -
nuclei, V vacuoles, M muscle.

Figure 12. Section through trichogenous (TII) gland of dorsal valves.
Va large vacuole, Iv involuted apical cell membrane,
V vacuole.

Figure 13. Section through trichogenous gland. N lobulate nucleus,
Va large vacuole, L lipid droplet, So socket.



* Iv



a somewhat different series of male behavioral responses in bioassays

(Chapter IV) it is speculated that a slower release rate from the dorsal

valves or the production of a different blend of components from that

produced in the Ism may establish the precise blend of pheromone vola-

tiles necessary for maximum effect during sexual signaling.

The trichogenous glands were far less numerous than the columnar-

cuboidal gland cells and were distributed over the outer surfaces of the

dorsal valves. The basal area of the trichogen cell contained a large

lobulate nucleus and a considerable number of vacuoles. Apically there

was a less dense area, reminiscent of the infolded apical membranes

associated with other such glands, in particular male hairpencil glands

(Noirot and Quennedey, 1974; Wasserthal and Wasserthal, 1977). Also,

within the apical area of the trichogen cells there were variable

numbers of large vacuoles (Figs. 12, 13), similar in shape and position

to the lipid droplets found in the papillae annales of Euxoa species

(Teal and Philogene, 1980). Although the vacuoles were apparently

empty, certain sections showed traces of a slightly osmiophylic sub-

stance within the vacuoles. The contents of these large vacuoles were

probably extracted during the dehydration sequence. Unfortunately, I do

not known whether the contents of these vacuoles contribute to the

pheromone blend released from this site. However, based on the large

content of these vacuoles and the minute quantities of components

obtained in gas chromatographic analyses of dorsal valve extracts I

hypothesize that these secretions serve that some other function.

Although nerves occur near the basal areas of the trichogen cells,

it is not known if synaptic junctions with the trichogen cells or


dendritic connectives to the tubular setae are present. However,

because cases of intragland nervation are infrequent among insect

species (Noirot and Quennedey, 1974), it may be that closely associated

setae provide a sensory input.



The pheromone blend released by female H. virescens was assumed to

be produced by a single area of hypertrophied epidermal gland cells that

form a complete ring in the 8th abdominal intersegment (Jefferson et

al., 1968). However, in the previous chapter, 2 distinct areas of

glandular epiderm were described in the terminal abdominal segments.

The distinct morphological positions of the sites and structural dif-

ferences in the cuticle overlying each area suggested that each site

developed independently, perhaps contributing different component ratios

to the total pheromone blend. The present study was undertaken to

assess the effects of extracts from each site in the elicitation of

neural responses and male sexual behaviors using the electroantennogram

(EAG) technique and wind tunnel bioassay procedures.

Methods and Materials

Extraction of Glandular Sites

Ovipositors were removed from females during the 2nd 7th scoto-

periods following emergence. The cut end of the ovipositor was placed

on absorbent paper to remove haemolymph (Klun et al., 1980a) and the

respective glandular sites [8th intersegment (Ism) and posterior 2/3 of

the dorsal valves (DV)] were dissected from the remaining tissue. Ethyl

ether (Mallinckrodt, anhydrous reagent grade) extracts of whole ovi-

positors, and both glandular sites were stored at ca. -500C until use.

EAG Assays

Insects used in EAG studies were immobilized in modeling clay so

that the upper part of the head and antennae were free but capable of

little movement. Antennal responses, recorded between the distal tip of

the antennae and cranial vertex, were made using glass capillary Ag-AgC1

electrodes containing 3M NaC1. Impulses were amplified by a Grass P-16

preamplifier and monitored using a strip-chart recorder and a digital

voltmeter. The sample delivery system was similar to that used by

Roelofs and Comeau (1971). Compressed air, delivered at a constant rate

of 200 ml/min, was used as the carrier gas and the volatiles were

dispensed into the airstream through a 4-cm long glass tube (6-mm ID)

containing sample-impregnated 1 X 3-cm filter papers. The output of the

delivery system was positioned ca. 1 cm from the distal 2/3 of the

antenna. Test samples included 0.32 female equivalent (FE), 1.00 FE,

3.20 FE, and 10.00 FE concentrations of the tissue extracts, a 1 pg

standard of (Z)-ll-hexadecenal, and an ether blank. All test series

were begun with a blank and ended with a standard, and a 30-sec recovery

period was allowed after each response. Responses were corrected and

standardized by the method of Baker and Roelofs (1976).

Wind Tunnel Bioassays

A 1.5 X 0.5 X 0.5-m wind tunnel was employed to assess the relative

abilities of each test extract and the recombined Ism and DV extracts to

elicit flight, tactic behavior from a distance, and close-range repro-

ductive behaviors (Chapter II). Air speed through the tunnel was

regulated by a variable speed fan so that fumes emanating from a 1 X 3-,-

cm cotton wick impregnated with TiC14 positioned in the upwind end

formed a defined plume. All tests were conducted during the peak period

of male reproductive activity (Tingle et al., 1978). One x 3-cm filter

papers impregnated with 1 or 10 FE concentrations of each test extract,

the recombined Ism + DV extracts, or an ether blank were suspended

centrally in the upwind end and individual resting males were lowered

into the plume 10 cm from the downwind end. The behaviors exhibited by

each of the 20 individuals tested for each sample during a 5-min test

period were recorded by the observer on audio cassette tapes and later


Frequencies of the observed behaviors were tabulated in 1st order

transition matrices and ethograms were devised for each of the test

samples. Chi-square values and standard normal deviate values of all

transitions were calculated by common techniques (Stevenson and Poole,

1976) and were used to assess the differences in the behavioral patterns

resulting from different test extracts and concentrations. In addition,

populational means for each test extract were compared using Duncan's

multiple range for the following criteria: the number of taxes, time

prior to flight, time prior to entering the pheromone plume, time spent

in the plume during the 1st orientation, time spent searching on the

dispenser after the Ist landing, and the total time that males were

active (i.e., vigorous wing fanning, ambulation, or flight).

Results and Discussion

EAG Studies

EAG responses indicated that male H. virescens perceived volatiles

extracted from both the DV and Ism sites. However, responses to the DV

extracts were lower than those to whole ovipositor extracts (Fig. 14).

Interestingly, all concentrations of the whole ovipositor extract and

higher concentrations of each gland site elicited responses of equal or

greater magnitude than those obtained with the 1 pg (Z)-ll-hexadecenal

standard. At the 1 FE concentration the total amount of pheromone pre-

sented in a whole ovipositor sample represents 40X less than that of the

1 jg (Z)-ll-hexadecenal standard (based on data of Tumlinson et al.,

1975; Klun et al., 1979). Unfortunately, it is not possible to ascribe

this increased neuronal output to different receptor sites on a single

or several neurons without single cell analysis. However, the responses

elicited from a group of 10 females by all test odorants were not

significantly different from those of the blank; therefore, it is

believed that the responses to these extracts are sex specific.

Flight Tunnel Assays

Release of individual males in the flight tunnel not only provided

data on the ability of test extracts to elicit male activation and

upwind taxis, but also enabled the sequential analysis of an individ-

ual's behavior. As indicated in Figs. 15 and 16 the complete range of

precourtship behaviors exhibited by males during mating studies (Chapter

II) including: male activation, random flight, searching for the

Figure 14. Mean EAG responses of 10 males to glandular extracts.
Vertical bars indicate standard errors of the mean.
Responses are expressed as corrected values (test
response (mV) X 10/response (mV) to 1 pg (Z)-ll-
hexadecenal). Stippled area represents the standard
ized blank response. Dashed line represents standard
response to 1 pg of (Z)-ll-hexadecenal.

*-* Whole
---- Ism
A-...-A DV

I I jI
-.5 0.0 0.5 1.0

Log of Concentration (FE)

............................ .......


Figure 15. Behavioral sequences evoked by 1 FE concentration of
each extract and the solvent blank in the flight
tunnel. Probabilities of the individual transitions
indicated are the results of monitoring 20 individ-
ually released males.

9 I

4 5

I 9
4 4
u o a

a: in <

i, 01 < Q -
a i
4 4 3 2


4' 4 4 0

LL w g S;

44 is
4 e -15- c 0 4 ~
2 4 r- I-~- .-~Y4 21
4 U 4 4W\~/~'

4 Li

4 0.

41 h 4L 4 4) '-'4
o 4 Li 4

Figure 16. Behavioral sequences evoked by 10 FE concentrations of each
extract and the solvent blank in the flight tunnel.
Probabilities indicated are the results of monitoring 20
individually released males.

4 4'

4, 4

-~ LU

-^ -) a- a 0:

a 4 4 L

^ ~~~ i I I' I
o i4 4 a-
(0 a- "o

"- a " 5 "

= LU -^ L
F a "

LU W.U 4 4
4 4 4Z a-

H 4 E ~ A UI a-
1:: ~ C, a 4 a- 4

4 ~ ~ ,L. 4 4

4 0 a -
LU a- 4 4 a-
4 4 LU
a- 4 4 a- L

4 4

Y 2 a a-a a

LU ta 4 i i
4~"L~k-~ 4,, U 4
4< A" V V

Figure 17. Male hovering at ca. 3 cm from dispenser. Note
antennae are pointed toward the dispenser.

Figure 18. Same male after landing on dispenser. Note antennae

Figure 19. Male exposing his hairpencil (arrow) on dispenser.

Figure 20. Male searching dispenser.

Figure 21. Ventral view of male searching dispenser. Note extent
of hairpencil exposure (arrows).

Figure 22. Homosexual mating attempt near dispenser.


* /


odorant plume, taxis up the plume, and a period of arrested forward

advance at 5-10 cm from the odor source were observed in these assays.

However, the only behaviors paralleling those exhibited during close-

range male-female courtship interactions were landing on the dispenser,

hairpencil exposure, and searching the dispenser (Figs. 17-22). Con-

sequently, the effects of the test extracts in evoking such prerequisite

courtship behaviors as movement under the female wing and clasping

(Chapter II) remain obscure.

Extracts from each glandular site were effective in evoking more

numerous responses at both concentrations than was the blank (Tables

3, 4), and behavioral transitions attributable to pheromonal cues

beginning with an aerial search for the odor plume and taxis up the

plume occurred only when glandular extracts were tested (Figs. 15,

16). While all of the sample extracts were equally effective in induc-

ing males to complete the reproductive behavioral sequence assessed

at the 10 FE concentration (Table 4), neither glandular site alone was

capable of eliciting the complete behavioral array when 1 FE concen-

trations were presented (Table 3). However, 1 FE concentrations of both

the whole ovipositor and recombined Ism + DV extracts were effective in

inducing the complete array of behavioral events (Fig. 15). In fact,

the recombined sites induced a greater behavioral array than did the

whole ovipositor extract (Table 3). The difference between the whole

ovipositor and Ism + DV samples can be accounted for by an increase in

the number of males entering random flight and making the transitions

from flight commencement or random flight to searching for the plume and




0 ,
C 0 04 0 0
(4. S4 1 0 In ,- 04 0 04 *-
o 00404C 04O0404.- 04 04 0 0)


( 0 U I L

0 ~ ~ 0* 4 04 4 )0 0 04 04 O f m

o' 0 O 01 S -' m () ^ o 0~ 0o 0~ -
'~ S ttf -^ S g g D 0I o ?

o0 04- o ) 04 04 04 0 1 4
=- 04 0 ) 40 0 4 0

Li1 1 0 x0 1 1
4S. 0
.0 0- -f O 0 4 -
11 j | ss g g a 5
^ ~ ~~~ B d d W of o
o fc N ro o m
0 S '1

t0 I-.^ S i S ^ '^ 0 3
4 o 0 ^ o 0 0 04

o 000 04 0' 0 04 04 04 0
04r 000 0) 04 0- 04 0 04 04 04

S. 0

.0 0 0 0 = 0 i 04 0 04 0 04 04 o4 0 4

.^~ ~ o S go d d g

0 04 0= 04 0 L< L

0. '0 0 t i < |^ "; ^ ^ ^
Cr .04 04 04 0 2 0 0 4

0" .- 04 C- 04
o~ ;~"^ ~ r
0. U i
00 04 004 0
04 .0 C- 0 0 02 0 D0 C
0 04 04 0 0 0 0 0 0 0 4 0
04 C. 0 00 0 0 0 4
04 0 0 I C- 04 0 0 0
0 4 0 0 0
04" 0 ~ ~ 04 ~ 0 000 0
04~o .0 0 0n 04 4 04 0 4
0 0 4 0 0 020 0 0 0, 4
0r 0- 0) C- 04 0 4 04 0
.04 01 0 0 C 0 0 4 00 0
04 ~ ~ ~ ~ 0 40 C-0 0 0 0 0 4 C 4 4
04J 04 04 04 H- 0 0 0 0 0 04 0 0 0 4

0) "~
(0 X~"
I-" i


1o C o o C m Cn om C n
m-- ii r C. Do o 0o cn Ln ,l) o cn
N > C 0 CD CO cl c m CD
-- O oo o -- Ln
C) oD

O o CD CDo 0 O O O iO CID 4-
+ Ln > O O O LO O O O O D4-
4- > m CO o r- C :O Lo o L Cn *r-i
oo an i in ~ j o0 on n -
o E Co o -- C -

0 00 000000 0
U-".L O .O C
c +4-i i i L O O - CD C

2U 0

0 O > 0

4- 0 l N

o o
a + Lii L 10 on C a On on 10 o 4-'

0 0 0 0 i- O i O O 0

ai m

u u
0 0
,( 0 -i n - 2- i- r-"
LU 0 Li > D an Oi i O r-1 0 r-n r a0n
rLr" V L N 0 m LO L0 L

0 O O c- ,M cn cn r. 4
Co C C "

4 --

'r- 4-' 4-
) u +4- 10) D r'- in r- r'- r- i- aj
- r 10 i 0
o E LL O O m LO Ln LO Ln tO -' 0 0
o Lo D r- "N co co oo co oa 41 C

Lii in in a C-- an an C r--- -0
c- C 0- 0o E n
o *- 0- 0
S O r--- O en -

(U O- S- I-
.r o 4-1 w
4 1-i c a (
4- i 0 2-> C- L Co 1 0 2- 0
0 ii n i mc LO 1 1 U E
+ V) ca i 10 can r- ao co co n 2 -
Liir > O o 0 a 10 a n inn 1 a-ii
a 2 .a *. .... .. .... (n in
o L 3 > 0 0 0 > i in i i 0 n- -2 en
3 >C C CD Cm 0 ) N
-i0 2. 0

CD S C4- C-
0 a- E
-- *r- IQ fC3 0)

S- > 1 U4-

Xo en -U L + --i 0 4-

s- -0 cQ x a -
SN 2 1 Li 0C 01 Ci
o 4u j aj i a, I + o ^

4- C 2 O a 0 a -r-
E = 4 o U __ CD M
> C O u t in C a
(U C n C r- XL C. Li *- 0 O .C r2

.i z- -0 4- 0- (1 -0 S- F- E 0 2
2C 0 4 r x s.- C It ra ar-

(2 U 2 ( 0 S2.- 2 e C Lm CD 4- /
an ma e - -J = l) x L C< -

continuing the behavioral sequence from these points (Fig. 16). Al-

though these transitions are performed by males responding to calling

females the inability of males to enter taxis directly after activation

suggests that the recombined Ism + DV extract contains a slightly

imprecise blend of components. Hence, although flight is induced the

males often search for further or appropriate stimuli prior to con-

tinuing and completing the behavioral sequence. As indicated in studies

on other species (Roelofs, 1978), this may reflect an imprecise ratio of

components perhaps resulting from more rigorous extraction and inclusion

of other volatile components which may be somewhat disorienting

(Weatherston and Maclean, 1974) when the glands were excised and ex-

tracted individually.

In assessing the points which accounted for the reduced behavioral

repertoire observed during 1 FE tests of the individual gland sites

(Table 5), it appears that the terminal points in the sequence are not

elicited to any great extent when using the Ism extracts while the

failure of males to undergo taxis during DV studies results in a reduced

behavioral repertoire. Although the ability of DV extracts to evoke

close-range behaviors remains obscure, the ability of whole ovipositor

and recombined DV + Ism extracts to elicit the whole repertoire tends to

suggest that volatiles from the Ism are responsible for inducing tactic

behavior while DV extracts maximize the probability of completing

courtship behaviors. However, considerable increases in the numbers

undergoing taxis and subsequent behaviors during 10 FE tests of the DV

extracts and a significant increase in the number of males performing

Table 5.--x2 Comparisons of behaviors elicited b 1 and 10 FE
extract concentrations.



Begin flight

Random flight

Search for plume

Taxis to source

Arrested advance

Land on dispenser

Hairpencil exposure

Search dispenser












Ism + DV






























Values for each behavior are contributions to the total x2.
Significant differences at 0.01.


the terminal step when 10 FE samples of the Ism were presented (Table

4), indicates that neither site is absolutely necessary for the elici-

tation of any step in the behavioral sequence (Baker et al., 1979).

Results obtained from the analysis of the number of orientations

and means of the temporal criteria measured tend to support the hypoth-

eses established in sequential analysis studies (Table 6). All of the

gland extracts caused males to begin flight significantly earlier than

the blank samples suggesting that pheromonal cues responsible for the

initiation of flight are present within each gland site. Further,

although the 1 FE concentration of the DV extract resulted in a signi-

ficantly longer pretactic flight period than 1 FE concentrations of

either the whole or Ism + DV samples, a 10X increase in the DV con-

centration resulted in values for all criteria which were not signi-

ficantly different from those obtained using 10 FE of the whole ovi-

positor sample. Hence the DV gland site does appear to produce an

effective pheromone blend. The number of insects used in analysis of

times spent moving up the plume and searching the dispenser was variable

due to the relative ability of each extract to elicit both behaviors

(Figs. 15, 16). Hence, the observed results are of limited value.

However, the data do suggest that males spent about the same length of

time moving up the plume for all extracts evaluated. Because the

probability of entering the pheromone plume is both extract and con-

centration dependent (Figs. 15, 16) it may be that if a male perceives

sufficient stimulae to enter the plume then he will move upwind at a

constant rate provided the stimulus is not removed (see Kennedy and







e (30



o a


4- 1

4n u
4- a



- U







" "





a ~mui
~ ooo

o~ orolm



~ ~"o"o

" ae~

~ ~"o"
to o~roa



o, ",aL~:,

" %8~
~ NN


,"ov ,
U Tu ,

o L d o



co m

,8 *
3 *- E

^ I ^S-




















Marsh, 1974). However, significantly less time was spent in an active

state when 1 FE of either of the individual gland sites were presented

than when 1 FE of the whole ovipositor extract was used. Further the

whole ovipositor and Ism + DV extracts were always at least as effective

as either of the individual gland sites. Therefore, it would appear

that, here again, volatiles from both glandular areas act in concert to

minimize the time necessary for males to respond to and find sexually

responsive females.

In summary, results of the EAG, sequential behavioral analysis, and

temporal analysis studies all strongly support the hypothesis that

discrete epidermal glands in both the 8th intersegmental membrane (Ism)

and dorsal valves (DV) produce a volatile sex pheromone blend. Further,

both gland sites apparently produce volatiles which act in concert to

maximize the probability that a male will effectively seek a sexually

active female and enter into courtship with her. This optimization of

the pheromone blend has been shown to be of critical importance to

several species (Baker et al., 1976; Carde et al., 1977), and has been

hypothesized as being imperative for the maintenance of reproductive

isolation between some species (Baker and Carde, 1979). The development

of the sex pheromone gland within the dorsal valves of H. virescens may

have been a response to direct reproductive competition between it and

both H. zea (Boddie) and H. subflexa (Gn.) because all 3 have distinctly

similar sex pheromones, sympatric ranges, and intersecting mating

periods (Tingle et al., 1978; Klun et al., 1980b; Chapter VII). Such a

development would, perhaps, not only finely tune the pheromone blend of

H. virescens, thereby providing highly effective male behavioral

releasers, but could also impart a measure of semiochemical isolation

through the production of a blend which was unattractive to other

species. Although direct support for this hypothesis is unavailable,

critical behavioral evaluations of volatiles emitted from anatomical

areas which have been shown to have marginal sex pheromone activity in

other species, particularly those species living in sympatricity with

close relatives (Chow et al., 1976; Smithwick and Brady, 1977b) may help

to support or refute this hypothesis.

Possibility of Inter-gland Contamination

A distinct problem in conducting this study was the very close

proximity of the individual gland sites to one another. Therefore,

although care was taken to remove only the posterior 2/3 of the DV there

was a distinct possibility that pheromone components originating within

the Ism might contaminate the surface of the DV and give rise to dubious

results. However, gas chromatographic analysis of DV extracts using

both 50 m OV-101 and 66 m SP-2340@ capillary columns indicate the

presence of a (Z)-9-tetradecenal:(Z)-ll-hexadecenal ratio of ca. 4:1.

Ratios reported for these 2 compounds by Tumlinson et al. (1975) and

Klun et al. (1980b) were 1:16 and 1:25.4 [(Z)-9-tetradecenal: (Z)-ll-

hexadecenal], respectively. Hence, the ratio of these 2 compounds in

the DV extracts is reversed indicating that contamination was probably

not a factor in these biological studies.



Recently, considerable emphasis has been placed on the control of

the tobacco budworm, Heliothis virescens (F.), by genetic means. The

basis of this approach lies in the production of sterile male and fer-

tile female hybrid and backcross progeny resulting from matings between

H. virescens males and females of a related species, H. subflexa

(Gn.). However, such laboratory matings occur very infrequently and

commonly result in the production of inviable offspring (Brazzel et al.,

1953; Laster, 1972). Interspecific matings between H. subflexa males

and H. virescens females are even more difficult to obtain although the

resultant hybrids appear to suffer fewer genetic abnormalities than the

aforementioned hybrid group (Proshold and LaChance, 1974). Hence, it

would appear that this pair of species have developed effective mech-

anisms of both pre- and postmating reproductive isolation.

Among vagile, related species sharing common ranges, reproductive

isolation is commonly effected by mechanisms acting prior to copulation

(Mayr, 1970), and although several such mechanisms may be involved,

disparate sex pheromone blends appear to be of major importance for many

species of Lepidoptera (Roelofs and Carde, 1974; Carde et al., 1977)

including H. virescens and H. zea (Boddie) (Klun et al., 1979). Judging

from the sympatric distributions and overlapping mating periods reported

for H. subflexa and H. virescens (Tingle et al., 1978) a similar

situation may be responsible for their genetic isolation. However,

differences in the ease of obtaining the 2 hybrid stocks indicate that

males of each species may respond quite differently to the pheromone

produced by females of the other species. The following work discusses

the results of tests designed to assess the influence of semiochemicals

on reproductive isolation between H. subflexa and H. virescens.

Methods and Materials

Heliothis subflexa were reared from eggs obtained from 1st to 3rd

generation laboratory stocks maintained at Gainesville, FL. Adults were

maintained as described in Chapter II, and held in isolation from members

of both the opposite sex and different species.

All bioassays were conducted in a 1.5 X 0.5 X 0.5-m wind tunnel and

during the period of overlapping mating activity described by Tingle et

al. (1978). In tests designed to assess the ability of actively calling

H. virescens females to elicit male sexual responses from male H.

subflexa, groups of 3-5 females were placed on a tobacco plant in the

upwind end of the flight tunnel during the prereproductive period.

Similarly, a Physallis plant provided a calling site for groups of 3-5

H. subflexa females during both intra- and interspecific mating studies.

Visual (Shorey and Gaston, 1970) and tactile cues (see Chapter II) have

been implicated as behavioral releasors for male reproductive behaviors

among several lepidopterous species so tests were also conducted using

ethyl ether gland extracts obtained from actively calling females of

each species. In each of these tests a 1 X 3-cm filter paper impreg-

nated with a 1 FE (female equivalent) concentration of the test extract

was suspended centrally in the upwind end of the flight tunnel. Males

used in all tests were released individually into the plume 1.4 m

downwind from the semiochemical source. The behavioral reactions of

each of the 20 males released for each assay were observed and recorded

on audio-cassette tapes by the observer during the 5-min test period.

Males that were active prior to being lowered into the plume were not

considered for analysis.

Frequencies of the observed behaviors were tabulated in Ist-order

transition matrices prior to formulating ethograms. Expected values of

each transition were calculated and compared using common techniques

(Chapter II).

Results and Discussion

Intraspecific Heliothis subflexa Mating

Although many aspects of both the male and.female reproductive be-

haviors of H. subflexa are similar to those performed by H. virescens

differences do exist.

The most prominent difference between the reproductive behaviors of

male H. subflexa and H. virescens is partial genital segment exposure by

H. subflexa males prior to and during tactic flight (p = 0.5). However,

male H. subflexa are capable of mating with both H. virescens (Proshold

and LaChance, 1974) and H. subflexa females without genital exposure

during precourtship behaviors. Therefore, although this behavior is

distinct and provides a good indication that males are sexually stimu-

lated it is impossible to infer that it will lead to further repro-

ductive behaviors or that a particular blend of synthetic pheromone

components which elicit this behavior will provide the stimuli necessary

for the completion of mating (Baker and Carde, 1979a).

Female H. subflexa are generally more passive than are H. virescens

during the protracted period of calling and can, in some instances, be

manually repositioned while calling. This passive behavior may explain

the absence from the genitalia of male H. subflexa of the long hair-

pencils present on the genitalia of both male H. virescens and H. zea.

This group of hairpencils appears to function in the dissemination of a

female arrestant during H. virescens mating and since H. subflexa

females seldom attempt escape when courted by H. subflexa (p = 0.05)

males, the evolution of these structures by both H. virescens and H. zea

males appears to be a species-specific event. Further, because 56% of

the interspecific courtships between H. virescens males and H. subflexa

females were terminated by female escape (Fig. 25), the development of a

species-specific arrestant pheromone by both H. virescens and H. zea

seems probable. Such species-specific "aphrodisiac" pheromones are

thought to function in the behavioral isolation of P. interpunctella

and Cadra cautella (Walker) (Grant et al., 1974) and in both cases

appear to provide a final premating mechanism of isolation.

Behavioral Interactions between Heliothis virescens and Heliothis


Studies on the semiochemically induced interactions and behaviors

between H. subflexa and H. virescens indicated that males of the 2

species respond quite differently to the naturally released sex phero-

mone of the other species.

Thirteen of the 20 male H. s:ub'exa released into the phercrone

plume created by calling H. virescens females exhibited activation

behaviors including wing fanning, ambulation, and genital exposure (Fig.

23). However, the mean time for these behaviors to occur was 3.37 min

(as opposed to ca. 1 min in conspecific studies), indicating that

although some semiochemical signal was being perceived it was not of

sufficient quality to induce an immediate behavioral response. Further,

of those males which committed activation behaviors, fewer were induced

to take flight, the only other behavior occurring with regularity

(binomial test, p = <0.05). Results of the study employing 1 FE of the

H. virescens pheromone bland extract were in general agreement with the

above results (x2, 6dF). However, there was a slight decrease in the

number of male H. subflexa becoming active and no males either entered

taxis or committed an aerial search in the upwind end of the tunnel

(Fig. 24).

Certain reproductive behaviors can be elicited using single phero-

mone components or even synthetic chemicals having some similarity to

known components (Roelofs, 1978), so it seems probable that pheromone

components shared by the 2 species are responsible for the elicitation

of the above responses. However, since only a single male was success-

ful in finding the H. virescens pheromone plume and entering taxis, and

because no males were successful in finding females when upwind, cues

necessary for the maximization and completion of reproductive behaviors

by H. subflexa males are absent from the naturally produced H. virescens

pheromone blend.

Figure 23. Behaviors performed by initially inactive H. subflexa
males released downwind from actively calling H.
virescens females. Values indicate the probabilities
that a specific behavioral transition will occur.

Male H. subflexo -- Female H, virescens

Inactive Remains Inactive

Activation Behaviors -Male stops

-- Nondirected Flight --- Male stops
T ""(- Continued Random Flight

Aerial Search Downwind--- (--Male stops

Upwind Male stops

Figure 24. Behaviors performed by initially inactive H. subflexa
males released downwind from a 1 FE sample of the H.
virescens pheromone gland extract. Values indicate
the probability that a specific behavioral transition
will occur. Stippled area includes activation

H. subflexa c--- I FE H. virescens Extract

Inactive cI-F --)- Remains Inactive

oF stops

-0d stops
c~ stops

Nondirected Flight --stops

Random Movement

The most effective blend of pheromone components identified from H.

subflexa female gland extracts has 2 components in common with H.

virescens, (Z)-9-hexadecenal and (Z)-ll-hexadecenal, and 3 unique

acetates, (Z)-7-hexadecen-l-ol acetate, (Z)-9-hexadecen-l-ol acetate,

and (Z)-ll-hexadecen-l-ol acetate (Chapter VI). Preliminary field

studies indicate that at the very least, the 2 C16 aldehydes plus (Z)-

ll-hexadecen-l-ol acetate are of importance to the capture of male H.

subflexa. Hence, the absence of these 3 acetate components from the H.

virescens pheromone (see Klun et al., 1979) and different ratios of (Z)-

9-hexadecenal:(Z)-ll-hexadecenal probably account for the semiochemical

isolation between H. subflexa males and H. virescens females. However,

the release of pheromone components by female H. virescens such as

tetradecanal, (Z)-9-tetradecenal, and (Z)-ll-hexadecanol which may be

disorienting to male H. subflexa and cannot be discarded as a possible

mechanism for reproductive isolation.

The behavioral repertoire exhibited by male H. virescens in re-

sponse to the pheromone blend produced by calling H. subflexa females

was quite distinct from that described above (Fig. 25). Only 10% of the

initially inactive male H. virescens failed to fly during these tests

and in fact, the probability of undergoing taxis toward calling 1H.

subflexa females (p = 0.80) was not significantly different from that

found in conspecific mating studies (Chapter II). Therefore, chemical

cues responsible for the elicitation of flight and tactic behavior by

male H. virescens are released by calling H. subflexa females. However,

while only 7% of the males failed to land, or approach, calling H.

virescens females in conspecific mating studies (Chapter II), 25% were

incapable of completing these behaviors in response to calling H.

subflexa females. This reduction in the number of males making contact

with females obviously contributes substantially to the very low pro-

bability of interspecific mating (p = 0.15) (p = 0.56, conspecific H.

virescens matings). Hence, it appears that semiochemically imparted

reproductive isolation between H. virescens males and H. subflexa

females results from the release of a pheromone blend that does not

provide a stimulus of sufficient magnitude to induce a high percentage

of the males to land and subsequently enter into courtship. Chemically,

this may result from the distinct ratio of (Z)-9-hexadecenal:(Z)-11-

hexadecenal produced by female H. subflexa (Chapter VII), and the

probable absence of both tetradecanal and (Z)-9-tetradecenal from the H.

subflexa blend. However, because a number of males do perform courtship

behaviors after contacting the female, disparities between the pheromone

blends of H. virescens and H. subflexa are not solely responsible for

reproductive isolation. Rather, it seems that both the pheromone blend

and the females' ability to escape from courting males (p = 0.25)

function as major inputs to reproductive isolation.

Cross-attraction between H. subflexa females and H. virescens males

has been shown to occur under field conditions (Tingle et al., 1978),

although the number of males captured was much lower than the number

landing near calling H. subflexa females in our studies. This reduction

in cross-attraction under natural conditions was expected because

species usually maintain premating mechanisms of reproductive isolation

more effectively in the wilo (Dobzhansky, 1951; Smith, 1953), and under

no-choice situations such as those usea in :his study, other closely

related noctuids have been shown to overcome barriers to reproductive

isolation (Byers and Hinks, 1978). Hence, it would appear that dif-

ferences in pheromone composition are the major factors contributing to

reproductive isolation under natural conditions and that female escape

behavior functions in a backup capacity.

Results of experiments employing 1 FE of the H. subflexa sex

pheromone gland extract were considerably different from tests in which

calling females were used (Fig. 26). In fact the range of behaviors

performed was so reduced that behaviors involved in the male activation

phase were assessed individually. This indicates that the pheromone

blend released from the H. subflexa extract was quite different from

that released by calling females. This is not an uncommon feature

encountered when using pheromone gland extracts and, in fact, the blend

of chemicals identified from H. subflexa gland extracts contained a

concentration of at least 1 mono-unsaturated C16 alcohol which was a

conspecific landing inhibitor (Chapter VI). The disparity between the

blends may involve a very slight difference in the component ratios

(Baker and Carde, 1979a), but it effectively shuts down long-distance

semiochemical communication when presented at a 1 FE concentration

level. Interestingly, small cage bioassays using the small chamber

described in Chapter II indicate that males do perceive sufficient

chemical stimulae from a 1 FE sample to perform such close-range court-

ship behaviors as searching the dispenser and hairpencil exposure.

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

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