Citation
Color polymorphism in nymphs of the southern green stink bug, Nezara viridula (Hemiptera: Pentatomidae)

Material Information

Title:
Color polymorphism in nymphs of the southern green stink bug, Nezara viridula (Hemiptera: Pentatomidae)
Creator:
Johnson, Carol Brown, 1953-
Copyright Date:
1984
Language:
English

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Carol Brown Johnson. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
11941117 ( OCLC )
ACQ9640 ( LTUF )
0030585955 ( ALEPH )

Downloads

This item has the following downloads:


Full Text













COLOR POLYMORPHISM IN NYMPHS
OF THE SOUTHERN GREEN STINK BUG,
Nezara viridula (HEMIPTERA: PENTATOMIDAE)









By

CAROL BROWN JOHNSON


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




UNIVERSITY OF FLORIDA


1984




































To my parents
Bryce and Lillian Brown
for their endless love and encouragement




















ACKNOWLEDGEMENTS


I am grateful to my supervisory chairman, Dr. Thomas C. Emmel, for his unflagging encouragement, helpful advice, and assistance with many practical matters in all phases of this research.

Each member of my committee has spent hours discussing with me the various aspects of my work. I thank them all for their valuable contributions. Dr. Pauline Lawrence was especially stimulating with her knowledge and ideas about insect behavior and physiology. Her incisive questions and her positive attitude were always motivating. Dr. Reece Sailer was invaluable for his familiarity with Nezara and its natural history. He helped me gain access to fields and garden space and showed a constant enthusiasm for the problem. Dr. Carmine Lanciani gave me advice on experimental design and data analysis. He also provided me

with several environmental chambers.

I thank Steve Polasky, Harold Magazine, and Chris Victoria for their assistance in the care and maintenance of bugs in fall 1981.

I could not have completed this dissertation without the love and constant support of my husband, Doug. He helped me in every way through all of the difficult times.

Finally, I give thanks to God, whom I serve through the grace of Jesus Christ.


iii




















TABLE OF CONTENTS


PAGE


ACKNOWLEDGEMENTS. . . . . . . .


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


LIST OF FIGURES . . . . .


. . .iii


ABSTRACT. . . . . . . . . . . . . . . Ix

CHAPTER

ONE INTRODUCTION. . . . . . . . . . . . 1


Evolution and Maintenance of Color Polymorphisms. . .
Color Patterns in Immature Insects. . . . . . .
Coloration of the Southern Green Stink Bug. . . . .
Objectives of This Study. . . . . . . . .

TWO DESCRIPTION AND BIOLOGICAL COMPARISON OF NYMPHAL MORPHS .

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

THREE PARENTAL EFFECTS ON NYMPHAL COLOR MORPHS. . . . .

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


FOUR EFFECT OF NYMPH DIET ON COLOR MORPH RATIOS. .

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

FIVE EFFECT OF TEMPERATURE ON COLOR MORPH RATIOS .

Introduction. . . . . . . . .
Materials and Methods . . . . . .


. 1 .3
.4
.5


.7
.8
13
20


28

28
29 35
54


. . 61

. . 61 . . 63 . . 70 . . 77

. . 82

. . 82
. .84


iv


. . iii












Results . . . . . . . . . . . . . 86
Discussion. . . . . . . . . . . . . 94

SIX CONCLUSION. . . . . . . . . . . . 107

Control of Nezara Nymph Polymorphism. . . . . . 107
The Adaptive Significance of Nezara Nymph Polymorphism. . 108 Areas of Future Research. . . . . . . . . 109

LITERATURE CITED. . . . . . . . . . . . . 111

BIOGRAPHICAL SKETCH . . . . . . . . . . . . 118


v


















LIST OF TABLES


TABLE PAGE 2-1. Sex determination of 4th instar nymphs
with respect to color morph. . . . . . . . . 17

2-2. Sex determination of green and black 4th instar
color morphs sampled randomly from a single egg mass . . 17 2-3. Sex of 4th instar nymphs collected from
field populations . . . . . . . . . . . 18

2-4. ANOVA of 4th instar dry weights of black and green
morphs reared together from the same egg mass. . . . 18 2-5. Differences in morph frequencies between 4th instars
which appeared earlier and later in the molting period . 19 2-6. Duration of 4th instar in relation to color morph. . . 21 2-7. Comparison of fertility of adult females which were
different color morphs in the 4th instar . . . . . 22 3-1. Comparison of consecutive egg masses in which the
proportion of black morphs sharply increased . . . . 47 3-2. ANOVA of egg diameter on two factors: the resulting
frequency of black 4th instar morphs and the parents
of the egg mass. . . . . . . . . . . . 49

3-3. Variation in frequency of green morphs in the 4th instar,
with respect to parentage. Summer 1981 data set . . . 3-4. Variation in frequency of green morphs in the 4th instar,
with respect to parentage. Fall 1981 data set . . . 51 3-5. Influence of parental color morph in the 4th instar
on the color morph frequencies of their offspring. . . 52 3-6. Comparison of egg production of parents which had
been reared and maintained on different diets. . . . 53 4-1. The effect of diet on morph ratios . . . . . . 71


vi


I













4-2. Effect of rearing nymphs in a single dish ("crowded")
or in groups of less than 15 ("uncrowded") with a diet
of solely green beans. . . . . . . . . . 72

4-3. Deaths of nymphs reared on different diets . . . . 75

4-4. Development time from egg-hatch to 4th instar
for nymphs reared on different diets . . . . . . 76

5-1. Color differences in 3rd instar nymphs reared at 280C. . 91 5-2. Development time of nymphs reared at different
temperatures . . . . . . . . . . . . 92

5-3. Deaths of nymphs reared at different temperatures. . . 93 5-4. The effect of cold shock . . . . . . . . . 95

5-5. Comparison of live weights and body temperatures
of green and black morphs used to construct
heating curves . . . . . . . . . . . 97


vii



















LIST OF FIGURES


FIGURE

2-1. Melanic patterns of thoracic cuticles of Nezara
viridula 4th instar nymphs . . . . . . .

3-1. Fourth instar color morph frequencies of successive
broods (Summer 1981 data set).. . . . . . .

3-2. Fourth instar color morph frequencies of successive
broods reared on green beans and peanuts at 230C
(1983 data set). . . . . . . . . .

3-3. Fourth instar color morph frequencies of successive
broods reared on peanuts and water at 23 C (1983
data set). . . . . . . . . . . .

3-4. Effect of parental diet on 4th instar color morph
frequencies of their offspring . . . . . .

4-1. Experimental design for testing the effects of diet
on color morph ratio . . . . . . . .

4-2. Experimental design for determining the effect
of a diet switch in the 2nd or 3rd instar on the
color morph ratios in the 4th instar . . . .


PAGE


. . 15


. . 41


. . 55 . . 66 . . 67


4-3. Effect of diet change on 4th instar coloration . . .


. 74


5-1. Experimental design for testing the effect of
rearing temperature on color morph ratio . . . . . 85 5-2. Effect of rearing temperature on 4th instar color
morph frequencies under two diet regimes . . . . . 89 5-3. Body temperature of 4th instar nymphs after
8 minutes of illumination in relation to live weight . . 96 5-4. Heating curves of 4th instar nymphs. . . . . . . 98


viii


















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



COLOR POLYMORPHISM IN NYMPHS
OF THE SOUTHERN GREEN STINK BUG, Nezara viridula (HEMIPTERA: PENTATOMIDAE)


By

Carol Brown Johnson


December 1984


Chairman: Thomas C. Emmel
Major Department: Zoology


Coloration of Nezara viridula (Hemiptera: Pentatomidae) nymphs

varies between and within instars. All nymphs transform from black to green in one or two molts. Adults are green. Polymorphism within the 3rd, 4th, and 5th instars can be explained by variation in the timing of color transformation. This study includes a biological comparison of 4th instar morphs and an evaluation of parental and environmental effects on nymph coloration.

Black and green 4th instar nymphs did not differ significantly with respect to sex, dry weight, development time through the 3rd instar, or reproductive success. Black morph 4th instar duration was slightly longer.

Color morph ratios varied among successive broods from single-pair matings. Last egg masses laid by senile females usually had higher



ix











proportions of black morphs than previous broods. Major shifts from green-dominated broods to black-dominated broods corresponded to lower survivorship through the 3rd instar but not to egg size or hatch success. Parental coloration during the 4th instar influenced the ratio of offspring color morphs, but did not restrict the range of variation. Differences in diet (green beans, peanuts, or both) affected adult fertility but did not have a clear effect on offspring coloration.

The effects of nymph diet and temperature were tested by dividing egg masses and rearing halves on different diets or at different temperatures. At 23 C, nymphs reared on single-item diets (green beans or peanuts) had slower growth rates, higher mortality, and higher occurrence of melanism than nymphs fed both food items. A diet switch during the 2nd or 3rd instar affected coloration of morphs in the 4th instar. Lower rearing temperatures increased melanization among 3rd and 4th instar nymphs. Diet and temperature affected coloration and development time to different degrees.

Under incandescent lighting, black morph 4th instar nymphs heated

more rapidly and reached higher equilibrium body temperatures than green morphs. Flexible melanism in nymphs may be an adaptation to improve basking in cool seasons and crypsis in summer. Control of nymph color expression is probably hormonal. The maintenance of within-instar polymorphism may be related to predator deception.


X


















CHAPTER ONE
INTRODUCTION




Evolution and Maintenance of Color Polymorphisms



E. B. Ford (1945, p. 73) defined polymorphism as "the occurrence together in the same habitat of two or more distinct forms of a species in such proportions that the rarest of them cannot be maintained by recurrent mutation." He recognized two categories of polymorphisms: transient and balanced.

Transient polymorphism is found when a certain gene becomes

advantageous and begins to spread through the population (Ford 1945). The classic example is industrial melanism among European Lepidoptera, where changing backgrounds have favored increased selection for melanic forms (Kettlewell 1973).

Balanced polymorphisms can exist only when selective pressures against different forms are balanced against each other (Ford 1945). For example, the many color forms of some butterflies have been explained as multiple mimicry, where each form imitates a different distasteful model (e.g., Sheppard 1962). Balanced polymorphisms in some cryptic species, such as land snails (Jones et al. 1977) and spittlebugs (Owen and Wiegert 1962), have been explained by apostatic selection (Ayala and Campbell 1974). That is, predators search for the most common prey form and the rare, or apostate, form has a selective


I








2


advantage. Hailman (1977) appropriately called this situation "deceptive polymorphism" and suggested that many examples of extreme intraspecific variability fit in this category.

Endler (1978) proposed that polymorphisms could be selectively

neutral if each color pattern morph resembles a different random sample of the environment. A similar argument has been made for polymorphic snails to account for higher morph diversity in more diverse habitats: habitat diversity allows a greater number of cryptic associations (Cain and Sheppard 1950, Rex 1972). An alternative hypothesis involves disruptive selection. In patchy environments, selection may act in different directions on different parts of the population. Gene flow between the subpopulations can then maintain a polymorphism (Jones et al. 1977).

The strictest definition of polymorphism does not include seasonal variation or continuous variation that falls within a normal distribution (Ford 1945). However, the term "phase polymorphism" has been used to describe the different color forms of the solitary and migratory phases of migratory locusts (Gillet 1978, Sasakawa 1967). In addition, some authors now refer to "seasonal polymorphism" when two distinct forms occur at different times of the year (Ishii and Hidaka 1979). In such cases, different morphs are not determined by different genes, but by environmental cues affecting gene expression (Wigglesworth 1959). Photoperiod, temperature, crowding, and humidity are most often cited as responsible for morph changes (Watt 1969, Hamilton 1973, Douglas and Grula 1978, Myers 1977).








3


Color Patterns in Immature Insects



The variations in form and color that occur during insect

metamorphosis led Wigglesworth (1959) to suggest that metamorphosis could be viewed as a special kind of polymorphism, "successive polymorphism." Indeed, this viewpoint may help to elucidate the selective forces that have led to dramatic transformations during larval development. As an example, cecropia moth larvae are jet black in the 1st instar, while succeeding instars are yellow with black tubercles, bluish green with red, blue, and yellow tubercles, and bright green with red and yellow tubercles (Williams 1961). Since there will always be temporal overlap in egg deposition, the population of cecropia larvae will be functionally polymorphic, varying both in size and coloration.

What could be the adaptive value of a successive polymorphism? Our understanding of balanced polymorphisms can supply some hypotheses. Since larval stages suffer predation from visual predators (de Ruiter 1952, Heinrich 1979), cryptic and mimetic forms should be selected for. If larvae change their cryptic color pattern with every molt, an individual's chance of being in a predator's search image would be greatly reduced. Changes in size with normal larval development may require a color or pattern change in order to remain cryptic in the same habitat (Endler 1978). Or, behavior and habitat choice may change with development, requiring a concomitant change in coloration to remain cryptic. Hamilton (1973) suggested that many caterpillars are black in the first instars because the additional thermal gain helps them to grow and attain a larger size more quickly. As the body size increases, rapid growth becomes less important and cryptic color becomes more

important.









4


Coloration of the Southern Green Stink Bug



The southern green stink bug, Nezara viridula (L.) (Hemiptera: Pentatomidae), is a very successful cosmopolitan pest of field crops, particularly legumes. Its life cycle and natural history have been recounted by numerous authors (Jones 1918, Drake 1920, Corpuz 1969, Singh 1973, Todd and Herzog 1980) and DeWitt and Godfrey (1972) give a comprehensive bibliography of the literature. In the Indo-Pacific region, several genetically distinct adult color forms coexist (Yukawa and Kiritani 1965, Singh 1973), but only the typical form, smaragdula, is in the western hemisphere (Drake 1920, Yukawa and Kiritani 1965). This form is uniformly light green in color.

Nezara has a relatively long adult life span, usually a month and more (Kiritani et al. 1963, Mitchell and Mau 1969, Corpuz 1969). Females lay large batches of eggs (60-120) at approximately one week intervals (Drake 1920, Corpuz 1969, Kiritani et al. 1963). After the reddish brown 1st instar nymphs hatch, they tend to cluster tightly on or near the egg mass. Second instar nymphs are black, with various white, red, and yellow spots. Second and 3rd instar nymphs also aggregate, especially before and during molting. Nymphs in the 3rd, 4th, and 5th instars are variable in color, from black to green. Third instar nymphs are usually black, most 5th instar nymphs are green, and the 4th instar is the most diverse in coloration. All nymphal instars and adults share the same habitat and food source. Nymphs fall victim to predatory hemipterans, ants, robber flies, parasitic wasps, spiders, and frogs (Stam 1978), and they are also acceptable food for the lizard








5


Anolis carolinensis Voight (personal observation). Adults are preyed upon by birds as well (Genung and Green 1974). Parasitoids are a major threat to eggs, large nymphs, and adults (Drake 1920, Stam 1978, Buschman and Whitcomb 1980).

The coloration variability in Nezara, therefore, falls both in the category of successive polymorphism (Wigglesworth 1959) and polymorphism in the sense defined by Ford (1945). Coloration changes as the insect matures, but variation between nymphs of the same age class also occurs. Since the species is subject to predation by visual predators, the color patterns are most likely the best compromise between protective adaptation and adaptation to opposing selective forces.



Objectives of This Study



The broad objective of this study of N. viridula nymph polymorphism is to gain sufficient understanding of nymph color determination to formulate an appropriate hypothesis of the adaptive significance of the polymorphism. The aspects that will be investigated include

(1) Changes in color pattern that occur during the different

stages of development.

(2) Diversity of color pattern in each instar.

(3) Differences between color morphs with respect to size,

development rate, and reproductive success in the laboratory.

(4) Consistency of color morph frequencies in sibling broods.

(5) Relation of color morph of offspring to parental color morph

history.








6

(6) Maternal contribution (egg size, hatchability) to nymph color

morph.

(7) Effect of parental diet on the outcome of nymph color.

(8) Effect of nymph diet on the outcome of nymph color.

(9) Effect of rearing temperature on the outcome of nymph color.

(10) Thermal advantages of black coloration in nymphs.

I will conclude with a discussion of possible ecological functions of Nezara nymph polymorphism that would be consistent with this information.


















CHAPTER TWO
DESCRIPTION AND BIOLOGICAL COMPARISON OF NYMPHAL MORPHS





Introduction



Nymphs of the southern green stink bug, Nezara viridula (Hemiptera: Pentatomidae) undergo dramatic color changes as they mature. Early instars are mostly black, while late instars are mostly light green; adults are light green. In describing light and dark forms of the late instars, Jones (1918) added the brief comment that most 5th instar nymphs were of the light type and 4th instar nymphs had about equal numbers of both. He also mentioned the existence of intermediate forms.

To the human eye, a population of Nezara nymphs presents a diverse array of size and color patterns. The diversity of color pattern in the 4th instar nymphs is as striking as polymorphisms described in Cepaea land snails (Jones et al. 1977), ladybird beetles (Houston and Hales 1980), and a variety of moths (Kettlewell 1973). Those polymorphisms, however, differ in significant ways from what appears in Nezara. Most polymorphisms that have been studied have involved adult phenotypes or characters which are constant throughout the entire life. In the case of Nezara nymphs, both size and color pattern can change within a few days or weeks. Yet, adult N. viridula in the United States are monomorphic. When variable phenotypes occur in the adult stage, direct correlations can be made between phenotype and reproductive success.









8


However, any survival advantage of a particular nymphal color morph would be separated in time from actual reproductive success.

If a certain nymphal color morph has greater reproductive success as an adult, it does not necessarily follow that the coloration itself is advantageous. One must first show that color variation is unrelated to inherent variations in other biological features which might affect reproductive success. To that end, I present comparisons of 4th instar color morphs with respect to sex, size, development rate, and reproductive success under laboratory conditions. Prior to these comparisons, I provide descriptions of the changes in color pattern that occur during the different stages of development.






Materials and Methods



Rearing Nymphs

I obtained Nezara viridula nymphs from the following sources: (1) natural populations in gardens and fields in the Gainesville, Florida, area; and, (2) eggs laid in the laboratory by wild females which had mated in the field, females reared from field-captured nymphs and mated in the laboratory, and first generation adults from lab-reared nymphs. Females and mating pairs were kept in 100 mm plastic petri dishes. Females usually laid eggs on the paper toweling which lined the dishes. Each egg mass laid in the laboratory was divided into two groups of sibling nymphs. Each group was reared in a 100 mm petri dish. I formed these groups by breaking the egg masses in two or by sorting newly molted 2nd instars into two dishes. The groups could then be assigned to separate treatments.








9


Both adults and nymphs were successfully maintained on diets of green beans (Phaseolus vulgaris) and shelled raw peanuts (Arachis hypogaea) purchased in local supermarkets. Some nymphs were reared on green beans alone or peanuts alone (with water provided by a piece of

sponge or cheesecloth fastened to the lid), but growth, survival, and color patterns were always affected. (See Chapter 4 for further discussion.) I provided fresh green beans, peanuts, and water every

other day or as needed.

In the summer and fall 1981, nymphs and adults were kept in an air-conditioned laboratory (25-27 0C) under 15L:9D incandescent and fluorescent illumination and with some indirect sunlight from a window. In all later experiments, I kept nymphs in constant temperature chambers at 23 or 28 C and with 12L:12D fluorescent lighting.



Color Morph Description and Categorization

In my preliminary observations of Nezara development, I looked at pattern, hue, and shade differences among individuals of each instar. I reared some nymphs in isolation in order to detect day-to-day changes. I kept others in small groups of like age and color to observe the diversity of changes that occurred with each molt. I reared entire egg masses or halves of egg masses together in a single dish to see the diversity among siblings.

Because the greatest diversity of color and pattern occurred in the 4th instar, I concentrated my study on this life stage. I established color morph categories by comparing many sequences of 4th instar nymphs arranged from lightest (green) to darkest (black). A set of sketches helped me to be consistent in categorizing nymphs. The extent of


I








10


cuticular melanization was verified by removing the dorsal cuticles of representative types, cleaning them of all epidermal tissue, and affixing them to glass slides for examination. Descriptions of the color patterns of the 4th instar and other stages follow in the results section.



Relation of Color Morph to Sex

To determine the relationship of 4th instar color morph and sex, I reared 4th instar nymphs of known color morphs to the adult stage. Since there is no suggestion in the literature that Nezara sex ratios at hatching are different from 1:1, I hypothesized that green and black

morphs from the same egg mass had equal chances of being male or female. I tested this by rearing all 53 polymorphic nymphs from a single egg mass of a wild female and comparing sex and color morph in a chi-square contingency table. As another verification, I randomly selected green and black nymphs reared from another, larger egg mass and tested the resultant sex ratios of each color morph with a binomial test (Siegel 1956). I also compared the sex ratios of small field collections of 4th instar nymphs in early fall (September October) 1981 and mid-summer (July) 1983.



Color Morph and Nymph Size

I tested for differences in 4th instar nymph size in relation to color morph by comparing dry weights of siblings of the same age. I selected 22 groups of polymorphic nymphs which had been reared on green beans and peanuts at 23 C in August September 1983. Each group consisted of nymphs from half of an egg mass reared together in the same









11

petri dish. Each egg mass was from a different set of parents. Nymphs in each group had been preserved together in a vial of Hood's solution (5 ml glycerol to 95 ml 85% ethanol). Therefore, within a sibling group, specimens received the same treatment until I removed them for drying. I used a random number generator to select a subsample of five green and five black nymphs out of all green and black nymphs in the vial. Nymphs were dried in individual aluminum cups at 600C until no further weight loss was measured. I tested the data with a two-way analysis of variance, in order to distinguish variation caused by the source of nymphs (i.e., egg mass) from variation attributable to coloration.



Color Morph and Development Rate

Since insect development rate can be influenced by many factors, I used a matched pair design to test for differences associated with color morph. I made observations of sibling groups (each from half of an egg mass) reared in petri dishes on diets of green beans and peanuts or peanuts and water at 23 C. As soon as the nymphs in each dish began to molt from the 3rd to 4th instar, I made daily records of the number and types of each morph in the dish. The number of 4th instar nymphs which had molted within the previous 24 h was found by subtraction of the consecutive daily totals. All nymphs molted within a two- to four-day period. I then compared the proportion of green, intermediate, and black morphs in earlier and later observations, using a sign test (Siegel 1956). If significant differences in development rate exist among morphs, then one morph should consistently appear more frequently among the earlier molting nymphs than among the later molting nymphs.









12


Development rate during the 4th instar was compared by continuing to rear selected single nymphs until the molt to the 5th instar. In 1982, I selected one set of 19 green and 7 black nymphs drawn from a single sibling group. I continued rearing them on a peanuts and water diet at 280C, recording molts every 24 h. In 1983, I reared a second set of 23 green and 22 black nymphs on green beans and peanuts at 230C. They were drawn from four different sibling groups of newly-molted 4th instars on the same day. I made observations of molting every 12-14 h. All nymphs were reared to adult stage to determine sex. I used a binomial test to determine whether sex ratio within a color morph sample differed from 1:1. Differences in the average duration of the 4th instar for green and black morphs were tested with a one-way analysis of variance.



Color Morph and Reproductive Success

Nymphs of known color type were collected in the field in September

- October 1981, reared to adults, and separated into mating pairs in 100 mm plastic petri dishes. Most of the matings matched adults which had similar color types. Additional mating pairs were formed from siblings which were reared from eggs in the laboratory. The adults in each of these pairs had identical patterns in the 4th instar. I collected and recorded all egg masses. By allowing eggs to hatch, I distinguished fertilized, viable eggs from unfertilized or otherwise inviable eggs. One measure of reproductive success is the production of fertile, viable eggs. If no differences exist between color morphs, the relative proportion of females laying fertile eggs should be the same. I tested this hypothesis with the chi-square statistic.








13

I also compared the size of egg masses of parents which were green or black in the 4th instar. To reduce the possible effects of genetic variability, I compared egg masses of sibling pairs derived from a single egg mass. I found the average number of eggs per mass for each pair and tested the differences with the Mann-Whitney U-test (Siegel 1956).






Results



Color Morph Descriptions

First instars were reddish brown in hue, while 2nd instars were black. Appearance changed as the cuticle stretched around the growing nymph. A pattern of red and white abdominal spots common to all stages (see Jones 1918) was very inconspicuous in recently molted nymphs. These spots seemed to grow in size as the abdomen distended. Thus, 2nd instars appeared solid black on the first day of the stage, but they became conspicuously spotted on subsequent days. However, all 1st and 2nd instar nymphs of the same age looked the same.

In contrast, there was great variation among nymphs of the same age during the 3rd, 4th, and 5th instars. The darkest individuals had a coloration pattern identical to the 2nd instar: black, except for the characteristic white and red abdominal spots and two golden-yellow spots on each lateral margin of the thorax. The lightest individuals were almost completely green. In all stages, blackest nymphs left black exuviae upon molting; the exuviae of greenest nymphs were transparent, with only a faint tan hue. Third instar nymphs were usually black; 5th









14


instar nymphs were usually green. The 4th instar was the stage most variable in color and pattern.

Fourth instar nymphs could be arranged in a continuous sequence

from lightest to darkest according to the degree of melanization of the thorax. When the thorax was least melanized, the general color of the head, thorax, abdomen, and legs was green. As melanization increased, the head became yellowish or black, the unmelanized part of the thorax became more yellowish in hue, the abdomen became dark green to brownish black, and the legs became black. The underside of the abdomen also varied from light green to pink to red.

After ecdysis and sclerotization was completed, the only aspect of a nymph's coloration that changed during the 4th instar was the shade of its abdomen. As the abdomen distended with feeding, dark green and greenish brown abdomens got distinctly lighter. On the other hand, the thoracic pattern could be confidently determined soon after the molt, did not change during the instar, and did not fade in preservatives. I therefore decided to categorize the 4th instar nymphs on the basis of thoracic patterns of melanization. I established eight categories, from lightest (I) to darkest (VIII) (Figure 2-1).

All of the nymphs that I observed individually changed from the black coloration characteristic of the 2nd instar to the green hue characteristic of the adult in one or two steps. In some cases, the black nymphs (2nd, 3rd, or 4th instars) changed to green nymphs in one molt, and they were green in all subsequent molts. In other cases, black nymphs molted to intermediate forms, which always melted to green forms in the following molt. I never observed a green or intermediate nymph of any age molt to a darker color.













. V I V


-4 40
0I


/cr~(N


V


Vill


Figure 2-1. Melanic patterns of thoracic cuticles of Nezara viridula 4th instar nymphs.
Drawings were made directly from specimens representing each of the 8 color morph categories (i-Viii).


III


VI Vil








16


Relation of Color Morph to Sex

Both males and females were reared from all 4th instar color

morphs. Within a brood, sex ratio did not differ significantly among the different color morphs (Table 2-1), and sex ratios of black and green morphs did not differ from 1:1 (Table 2-2).

The samples of 4th instar nymphs collected in the field did not include enough darker morphs to test sex ratio in a contingency table (Table 2-3). Both sexes were represented in each color category in the fall sample. No black females and only two black males were collected in the summer sample.



Color Morph and Nymph Size

The average dry weights of 110 green and 110 black 4th instar siblings from 22 egg masses were 11.33 mg and 10.60 mg, respectively (pooled standard deviation = 3.28). In 8 out of 22 sibling groups, black morphs were larger than green morphs. Analysis of variance of the dry weights showed that the difference between morphs is not significant (Table 2-4). The different sibling groups had a significant effect on the variance of the data.



Color Morph and Development Rate

No particular color morph could be significantly associated with faster development from hatch to the 4th instar (Table 2-5). However, in the majority of sibling groups observed, there was a higher frequency of green morphs earlier in the molting period than later.








17



Table 2-1. Sex determination of 4th instar nymphs with respect to color morph. All reared from a single egg mass from wild parents.


Color Morph Female Male


Green (I-1I) 11 4 Intermediate (III-V) 6 8 Black (VI-VI1I) 11 13 X = 3.56

df 2

p = 0.169















Table 2-2. Sex determination of green and black 4th instar color morphs sampled randomly from a single egg mass.


4th Instar Total No. in Sample
Color Morph Egg Mass Size Female Male p* Green (I-II) 27 19 11 8 0.65 Black (VIII) 25 10 5 5 0.62


*Binomial test (2-tail).








18


Table 2-3. Sex of 4th instar nymphs collected from field populations.


Sept./Oct. 1981 July 1983 4th Instar
Color Morph Female Male Female Male


Green (I-1I) 7 16 51 35 Intermediate (III-V) 2 1 2 4 Black (VI-VIII) 5 4 0 2














Table 2-4. ANOVA of 4th instar dry weights (mg) of black (VI-VIII) and green (I-II) morphs reared together from the same egg mass. All egg masses were from different parents. All nymphs were reared at 23 C on green beans and peanuts.


Variable Sum of Squares d.f. Mean Square F p Egg Mass 1581.323 21 75.301 6.978 <0.01 Color Morph 29.588 1 29.588 2.742 >0.05 Interaction 288.734 21 13.749 1.274 >0.05 Error 1899.237 176 10.791








19


Table 2-5. Differences in morph frequencies between 4th instar nymphs which appeared earlier and later in the molting period. Values are the number of sibling groups in which differences in frequency occurred.


Relative Morph Frequency


Total No. Color Earlier% Later% > Later% =
Diet Groups Morphs > Later% Earlier% Earlier% p*


Green Beans, 18
Peanuts

Green 9 4 5 0.13 (I-II)
Intermediate 3 12 3 0.02 (III-V)
Black 7 8 3 0.50 (V-VIII)


Peanuts, 16
Water

Green 10 4 2 0.09 (I-II)
Intermediate 7 8 1 0.50 (III-V)
Black 2 8 6 0.06 (VI-VIII)


*Sign test (1-tail).








20


All color morphs were represented among the first and last nymphs to molt in the sibling groups. On the whole, the duration of the 4th instar was slightly longer for black morphs than for their green counterparts (Table 2-6).



Color Morph and Reproductive Success

Nymphs of all color morphs were equally likely to successfully reproduce as adults (Table 2-7). There were also no significant differences in the fecundity of sibling females which had been different morphs in the 4th instar. In 11 sibling mating pairs reared from one egg mass, the 7 green (I-II) pairs had an average of 98.5 eggs per mass per mating pair, compared to 104.0 for the 4 black (VIII) pairs (U = 13, p = 0.928, 2-tail). The largest egg masses of green pairs ranged from 85 to 129 eggs; egg masses of black pairs ranged from 117 to 123. The average hatch success of eggs laid by green mating pairs was 81.4% (56.0

- 98.4%), compared to 86.9% (68.3 96.8%) for eggs of black pairs.






Discussion



Polymorphism in Nezara

The basic color pattern of Nezara viridula nymphs changes only at the time of the molt. The transition from black to green occurs in one or two stages between the 2nd instar and the adult stage. The polymorphism among 3rd, 4th, and 5th instar nymphs can be explained by variability in the timing of the color transition. Under normal









21


Table 2-6. Duration of 4th instar in relation to color morph. Differences tested with one-way analysis of variance. n = number of nymphs.


Rearing Conditions


Morphs Compared


No. of Days n (Mean + S.E.)


Peanuts, Water*
280C



Green Beans,**
Peanuts,
23 C


Green (I-II) Black (V-VI)



Males:
Green (I-II)
Black (V-VIII)


Females: Green (I-II) Black (VI-VII)


19
7





7 13


16
9


6.6 + 0.31 10.0 + 1.15





5.9 + 0.13 6.4 + 0.33



6.8 + 0.33 7.4 + 0.41


15.41 <0.001 1.14 >0.05


1.02 >0.05


*Observations made every 24 h. All nymphs drawn from one egg mass. Sex ratio in each color category not significantly different from 1:1.

**Observations made every 12-14 h. Nymphs drawn from 4 egg masses. Sexes analyzed separately since sex ratio for green category is different from 1:1.


F


p









22


Table 2-7. Comparison of fertility of adult females which were different color morphs in the 4th instar.


Field-Captured* Lab-Reared** 4th Instar No. No. Laying No. No. Laying
Color Morph Mated Fertile Eggs Mated Fertile Eggs


Green (I-I) 15 10 9 8 Intermediate (III-V)*** 2 2 --- --Black (VI-VIII) 5 3 9 8 X2 = 0.016 X2 = 0 df = 1 df = 1 p > 0.90 p > 0.99

*As 2nd, 3rd, or 4th instar nymphs.

**Sibling crosses.

***Values pooled with Black category for chi-square analysis.








23


conditions, most nymphs change color at the 3rd or 4th nymphal molt, making the 4th instar the most diverse stage. An individual may be transformed in one molt from a black bug to a green bug, or it may go through one instar with intermediate coloration. Once the green coloration is attained, subsequent stages are also green.

Kobayashi (1959) provides descriptions and drawings of 3rd, 4th, and 5th instar polymorphism in N. viridula in Japan. The range of variation in 4th and 5th instar color patterns is the same as I have observed, but my intermediate classifications (III, IV, and V in Figure 2-1) are not represented. His drawings of 3rd instar nymphs show only dark types (VI-VIII), whereas I have collected green (II) 3rd instar nymphs in a Gainesville garden. Other published descriptions of nymphal stages do not mention 3rd instar polymorphism; they describe a monomorphic black pattern identical to the 2nd instar (Jones 1918, Drake 1920, Singh 1973). Singh (1973) did not find black morphs of the 5th instar in Jabalpur, India, and the darkest 4th instar nymph he described had green, unmelanized areas on the thorax. Even though several adult color forms (none of which involve melanism) have been described in the Indo-Pacific region (Yukawa and Kiritani 1965), local researchers have not made distinctions in the appearance of their nymphs (Kobayashi 1959, Singh 1973). At least one other Japanese Nezara species, N. antennata Scott, has polymorphic nymphs like N. viridula (Kobayashi 1959, Kariya 1961). Light and dark color morphs have also been described for nymphs of the closely related Acrosternum hilare (Say) in the United States (Drake 1920).








24


Similar Polymorphic Systems

Nymphal polymorphism has been described in another pentatomid, Perillus bioculatus Fabricius, which is also polymorphic as an adult (Knight 1924). Adult Perillus have complex striped patterns of melanization that vary in extent on the pronotum, scutellum, and corium. Unmelanized areas vary in hue front white to yellow to red; red accompanies increasing melanization. The darkest nymphal morphs are red with black thoraxes and wingpads, while the lightest are white with some unmelanized areas on the thorax. Under normal conditions, nymph color forms coincide closely with adult color form. Knight concluded from his rearing experiments that temperature and humidity were the major factors accounting for the polymorphism.

A diversity of dorsal black cuticular pigmentation patterns appears among nymphs of the lime aphid Eucallipterus tiliae (L.) (Kidd 1979).

First generation nymphs are all yellow and unmelanized, but later nymphs develop black markings on the head, thorax, and abdomen in 2nd instar. Unlike Nezara patterns, these remain constant through remaining nymphal stages. Adult coloration is not variable. Kidd (1979) associated the appearance and greater frequency of melanics with crowding and leaf maturity.

Another larval polymorphism very similar to the Nezara system is found in the saturniid moth Saturnia (Eudia) pavonia L. First instars are black, while the last instars are usually solid pale green. In the 4th and 5th instar, larvae develop black patterns on each segment which can vary from thin black markings around the dorsal tubercles to a solid transverse band of black (Long 1953). Hintze-Podufal (1974) described the morphs extensively and, using larva] exuviae, showed that the black








25

markings were incorporated in the cuticle. The frequency of occurrence of melanic forms in the laboratory was influenced by crowding, light, humidity, temperature, and the freshness of the food (Long 1953, Hintze-Podufal 1977).



Differences Between Morphs

Sex. There is no evidence that color morph expression in the 4th instar is related to the sex of the individual (Tables 2-1, 2-2). Since males are heterogametic (Kiritani et al. 1962), any genetic control over the polymorphism is probably autosomal. More field sampling (Table 2-3) is needed to determine whether there are differences in the sex ratio of morphs in the field. Since males develop slightly faster than females and are smaller as adults (Kiritani 1964), adaptive value of a color morph may differ between the sexes.

Size and development rate. A distinction should be made between growth rate (rate of increase in mass or size) and development rate (rate of maturation). Reaching a larger size faster may confer a number of advantages, including escape from predators (Hamilton 1973) and improved thermal stability (Casey 1981, May 1976). Larger male Nezara live longer, have better mating success and are more fertile (McLain 1981). Female fecundity and longevity increases with body weight (Kiritani and Kimura 1965). On the other hand, reaching reproductive maturity at an earlier age, regardless of size, can increase an individual's relative genetic contribution to succeeding generations (Hamilton 1973).








26

Since there are no significant differences in dry weights of green and black 4th instar morphs reared in the laboratory (Table 2-4), color morph is probably not inherently linked with growth rates. However, green morphs appear to develop at a slightly faster rate. In most of the broods tested, there were proportionally more green morphs among the first nymphs to molt to the 4th instar than among the later-molting nymphs (Table 2-5). Black morphs average significantly longer 4th instar periods when equal numbers of each sex were reared on peanuts and water at 28 C (Table 2-6). At 230 and with the green beans and peanuts diet, black morphs averaged about half a day longer than green morphs of the same sex. Although the differences were not significant, the trends in these samples justify further experimentation. Inherent differences in black and green morph development rates may be magnified at lower temperatures. Since sex influences development rate, development of black and green morphs of both sexes should be monitored to the adult stage.

Reproductive success. Differences in 4th instar coloration were not associated with differences in general reproductive success in the laboratory. Equal proportions of fully reproductive adult females were reared from representatives of all 4th instar morphs (Table 2-7). Sibling matings of like types (green-green and black-black) demonstrated equal fertility and fecundity of males and females derived from both morphs.



Conclusions

Fourth instar color pattern is not clearly linked to other genetically or maternally inherited characters that could confer









27

selective advantage. There is some evidence, however, that development rate may be slightly faster in green morphs than in black morphs. Since the differences overall appear minor, I must conclude that nymphal color polymorphism (1) is linked to some other adaptive feature that I have not measured, (2) is selectively neutral (ecologically irrelevant), or

(3) has its own ecological function, relevant only in the natural habitat.

















CHAPTER THREE
PARENTAL EFFECTS ON NYMPHAL COLOR MORPHS



Introduction



The ecological function of any color polymorphism must be examined in the context of the control of color pattern expression. For example, mimetic polymorphisms can only be maintained when there is strict genetic inheritance of a particular pattern (Ford 1945). Apostatic selection can maintain a cryptic polymorphism only if the rare morph is able to transmit its deviant alleles to its offspring (Ayala and Campbell 1974). However, if color patterns are determined exclusively by environmental factors, the specific color pattern of an individual will be independent of the color pattern of its parent. This situation is found when environment changes radically from one generation to the next and the adaptive value of a parental color pattern changes with it (Rowell 1971). For example, migratory locusts transform in one generation from cryptic, solitary forms to densely aggregating, conspicuously colored forms (Goodwin 1949).

The late instar nymphs of the southern green stink bug, Nezara

viridula, are polymorphic (Jones 1918). The most diverse stage, the 4th instar, typically has forms ranging from black to light green. The darker color morphs have greater cuticular melanization in addition to darker subcuticular pigmentation. The blackest forms are identical in color pattern to the typically monomorphic 2nd and 3rd instars. The


28








29

greenest forms are identical to the common green morph of the 5th instar and are the same hue as the green adult. (See Chapter 2 for complete descriptions.)

If the control of color pattern in Nezara nymphs is purely genetic, then the ratio of morphs in the broods of any mating should remain the same regardless of environment. Black-black and green-green matings should yield offspring morph ratios at opposite extremes. If the variation in color pattern is a purely environmental effect, then all broods in the same environment should have the same variety and frequency of color patterns, regardless of parentage. Maternal effects would be indicated if the mother's environmental history or present health corresponds with changes in the color patterns of her offspring.

In this paper, I present color morph ratios for mutiple broods from single-pair matings of Nezara viridula and the results of crossing like morphs. In addition, I examine possible maternal effects by comparing the color morph ratios of broods from parents reared on different diets and by comparing egg sizes where black color morph frequency changes between successive broods of the same parents. The relative importance of genetic, environmental, and maternal effects will be discussed.





Materials and Methods



General Approach

Single-pair matings of virgin adult Nezara viridula provided series of egg masses which had common parentage. After rearing broods under identical conditions, I compared the frequencies of 4th instar color









30


morphs in broods from the same parents and from different parents. In addition, I made comparisons between offspring of (1) pairs which had been different color morphs in the 4th instar and (2) pairs which had been reared and maintained on different diets. I also looked for correlations between several measurable characteristics of the egg mass and the frequency of black color morphs that resulted.



Source of Specimens

The eggs and nymphs used in this study came from parental stocks of Nezara viridula formed in June 1981, September-October 1981, and July-August 1983. Data sets originating from these stocks will be referred to as "summer 1981," "fall 1981," and "1983," respectively. I made all- field collections of Nezara in gardens and soybean fields in Gainesville, Florida.

I formed the summer 1981 parental stock from the offspring of a pair of adults which had been collected the previous October and had overwintered in the laboratory. These siblings originated from the sameegg mass and were paired together according to their coloration during the 4th instar.

The fall 1981 stock came from (1) 3rd, 4th, and 5th instar nymphs

collected in the field in September and (2) nymphs reared from eggs laid by two gravid females captured in early September. Matings from the latter source were sibling crosses. Wherever possible, I mated adults which had had the same coloration in the 4th instar.

All of the 1983 mating pairs were made up of adults reared from

2nd, 3rd, 4th, and 5th instar nymphs collected in mid-July 1983. All of the 4th instars were green morphs (I-II, Figure 2-1). With the








31


exception of a special group of sibling matings, all matings were between bugs that had been collected on different days and on different plants. I formed a separate group of sibling mating pairs from a single dense aggregation of 2nd instars collected in the field. On the basis of my prior observations of nymph development and behavior, I concluded that these nymphs had hatched from the same egg mass. This group of bugs was subdivided and reared on different diets, as described below.



Maintenance of Parental Stock

Single mating pairs were kept in 100 mm plastic petri dishes lined with paper toweling. All mating pairs remained in a laboratory room (25-270C) under 15L:9D incandescent and fluorescent lighting. Some indirect sunlight came through a window. Unless specified otherwise, all adults received a diet of fresh green beans (Phaseolus vulgaris) and shelled raw peanuts (Arachis hypogaea) purchased in local supermarkets. I used garden-grown green beans for a few weeks in June and September 1981. I provided fresh green beans every other day in 1981 and every day in 1983. I replaced peanuts when they became soft or moldy. I cleaned dishes and replaced paper toweling as needed. Each day I collected egg masses, which were usually laid directly on the paper toweling. If egg masses were laid elsewhere, I used white glue to affix them to a piece of toweling.



Nymph Rearings

In 1981, I kept all egg masses and nymphs with the parental stocks in the laboratory at 25-27 C, 15L:9D. In 1983, developing egg masses were moved to a 23 0C constant temperature chamber on the 4th day of









32


incubation. Nymphs hatched and developed in the chamber, which had 12L:12D fluorescent illumination.

The nymphs from each egg mass were reared in two 100 mm plastic

petri dishes. The two groups were formed by (1) allowing the entire egg mass to hatch and later dividing newly molted 2nd instars (1981 only) or

(2) breaking the egg mass in two before hatch (1981 and 1983). The different techniques appeared to have no effect on survival or color morphs. In 1981, half of the nymphs of each egg mass were reared on green beans and peanuts while the other half received peanuts and water from a moistened piece of cheesecloth. In 1983, both halves of each brood of three mating pairs were reared on peanuts and water. For 12 other mating pairs, only one half of each of the first three broods received green beans and peanuts; the other half received treatments in conjunction with other experiments. For the remaining broods, both halves received green beans and peanuts. I provided fresh food and clean dishes every other day in 1981 and each day in 1983.



Determination of Color Morph Frequencies

When the nymphs in a petri dish reached the 4th instar, I assigned each nymph to one of eight color morph categories based upon the degree of melanization of the thorax (Chapter 2, Figure 2-1). For analysis, I combined the two lightest categories (green), the three intermediate categories (intermediate), and the three darkest categories (black). I calculated green, intermediate, and black color morph frequencies by dividing the number of nymphs in each category by the total number of nymphs reared in the dish. When both halves of a brood received identical diets, I combined the data from both dishes.









33


Variation in Color Morph Frequencies Among Broods

In summer 1981 and in 1983, I reared continuous series of egg

masses from different mating pairs. All series included the first egg masses produced by the mating pair. In 1983 I continued rearings of fertile egg masses until the female died. I made graphical comparisons of the frequencies of black, intermediate, and green 4th instar color morphs resulting from each egg mass or half egg mass for each mating pair.



Egg Mass Characteristics and Resulting Color Morphs

When dramatic changes in color morph frequencies occur between two broods from the same mating pair, a nongenetic factor must be responsible. Such changes occurred in my rearings of successive egg masses in 1983, despite my efforts to maintain constant rearing conditions. If the changes were associated with changes in the mother's reproductive state, we might expect to see changes in egg size, number of eggs per egg mass, and egg viability.

I selected seven mating pairs for which the frequency of black

morphs in their broods increased 33 to 78 percentage points from one brood to the next. For each egg mass, I counted the eggs and calculated the percent successfully hatched. I selected a random sample of 35 eggs and measured the diameter of the operculum using an ocular micrometer and a dissecting microscope. I ran a two-way ANOVA (Sokal and Rohlf 1973) of egg diameter on the two factors: (1) low vs. high black morph frequency and (2) the parents of the egg mass.








34


Effect of Parentage on Color Morph Frequencies

Variation in color morph frequency can occur both among broods from the same parents and between broods of different parents. If parentage has an effect on offspring color morphs, one would expect that the variation among sibling broods would be less than the variation among all broods in the population. I tested this hypothesis with a Kruskal-Wallis one-way analysis of variance by ranks (Siegel 1956) on green morph frequencies in broods from the summer 1981 and fall 1981 data sets. I restricted analysis to mating pairs for which I had data from three or more broods.

If 4th instar coloration is heritable, one would expect a strong correlation between the parents' 4th instar color morphs and the frequency of those morphs among their offspring. For example, parents which were completely black in the 4th instar ought to have a higher proportion of black morph offspring than parents which were green in the 4th instar. I tested this idea using the summer 1981 and fall 1981 mating pairs which had been either black or green in the 4th instar. Since I had reared only one, two, or three egg masses of many of the mating pairs, I used only the first three egg masses of the other mating pairs in my analysis. For each mating pair, I found the average black and green color morph frequencies. I then tested the values obtained for green morph parents against those for black morph parents using a Mann-Whitney U-Test (Siegel 1956).



Effect of Parental Diet on Color Morph Frequencies

Because embryos are packaged in an egg shell with a supply of nutrients and hormones provided by the mother (de Wilde and de Loof









35


1973a), the physiological state of the mother can play a role in the success of the offspring. Since nymph color morphs are affected by their own diet (Chapter 4), parental diet may have an effect as well. I tested this hypothesis in 1983 by rearing three sets of sibling 2nd instars on three diets: (1) green beans and peanuts, (2) peanuts and water, and (3) green beans alone. I matched pairs of virgin males and females which had been reared on the same diet and continued the diets until the death of the female. I divided each of the first two or three viable egg masses from these matings and reared half of the nymphs on green beans and peanuts and half on peanuts and water at 23 C. I compared color morph frequencies, hatch success, and egg mass size.






Results



Variation in Color Morph Frequencies Among Broods

Summer 1981 data set. Frequencies of green, intermediate, and

black morphs in the 4th instar varied between successive broods from the same parents (Figure 3-1). For most mating pairs (B, C, E, F), the first three half-broods reared on the green beans and peanuts diet were more similar to each other than to the last two or three broods. In those cases, the proportion of green nymphs declined in the later egg masses. The ratios of color morphs of half-broods reared on peanuts and water also varied. Again, in all cases the later broods had higher proportions of black morphs and lower proportions of green morphs than earlier broods.

In addition to the variation between broods from the same parents, egg masses laid on the same day by different mating pairs often yielded






































Figure 3-1. Fourth instar color morph frequencies of successive broods (Summer 1981 data set). Half of each brood was reared on green beans (G) and peanuts (P), and half on peanuts and water (W) at 25-27 C. The egg masses of 6 different mating pairs are represented (A-F). All parents are siblings from one egg mass. Number of 4th instars indicated above bars.









37


A


green
*Aintermediate


1008060-


22


43

57


53 59











6/11 6/17 &22 &27 7/4
P, W


100] B


23

21


334








6/11 6/16 6/22 6/27 7/2 7/9 6/11 6/16 6/22 6/27 7/2 7/9
G, P P, W 57 C 5
49 42




55
48
51 .. 54 47
51






6
6/11 6/16 6/22 6/2 7 7/4 6/11 6/16 6/22 6/27 7/4


G. P


P, W


DATE EGGS LAID


b lac 61



50
56 43 -0

45








6/11 6/17 6/22 6/27 7/4
G, P


40--I


20-


0-I-


80

z
W 60

0

400



0.



1008060


4020] 0


I


I


37


I


'
4













100]


Figure 3-1--Con tinued


38


80


60.


4020


0



100> 80w 60

D
LU 40 L2
20C0c 00


100-


D 42 1 42


47

24 _39 47









6/15 6/18 6/24 7/4 7/11 6/15 6/18 6/24 7/4 7/11
G, P P, W


E
_ 36 31 1



31 32 41
34 37 _33 36







6/16 6/22 6/27 7/4 7/9 6/16 6/22 6/27 7/4 7/9
G, P P, W


F 34 26
32
26
14
29

34 4 40 286/16 6/20 6/24 7/1 7/7 6/16 6/20 6/24 7/1 7/7
G, P P, W

DATE EGGS LAID


80. 60


4020


0








39


very different ratios of 4th instar morphs. For example, the June 16 brood of mating pair B had about 25% black morphs while broods of mating pairs C, E, and F had about 5% or fewer blacks on the diet of green beans and peanuts.

1983 data set. The color morphs of broods reared on green beans and peanuts in 1983 (Figure 3-2) varied less than the 1981 broods. For many mating pairs, the color morph frequencies remained consistent (many greens, few intermediates and blacks) for six or more successive broods (Figure 3-2 A-F, H, I). Sometimes the first brood (Figure 3-2 A), the second brood (Figure 3-2 D, H), or both (Figure 3-2 G) had distinctly higher proportions of darker morphs than the other broods preceding or following. But in all cases except mating pair A, the last broods produced by the mating pairs had higher proportions of black morphs and lower proportions of green morphs than earlier broods.

The results of rearing broods on peanuts and water (Figure 3-3) were very different. Two mating pairs (Figure 3-3 A, B) had five successive broods in which color morph frequencies were identical, with black morphs dominating. But in several subsequent broods, green and intermediate morphs dominated. The broods of the third mating pair (Figure 3-3 C) had lower proportions of black morphs than the broods of the other mating pairs. The second and third broods showed a shift to

dominating green morphs.



Egg 'Mass Characteristics and Resulting Color Morphs

There was no consistent trend in either egg mass size or hatch success when consecutive broods with low and high black frequencies occurred (Table 3-1). Egg mass size and hatch success decreased in





































Figure 3-2. Fourth instar color morph frequencies of successive broods reared on green beans and peanuts at 23 C (1983 data set). Each set of bar graphs incudes rearings of all fertile egg masses laid during the lifetime of one female mated to one male. The broods of 12 different mating pairs (A-L) are represented. Number of 4th instars indicated above bars. only one half of egg mass represented in this rearing.

















Screen


97


erme a e black
-5* 51* 102 108 5 58
109




753*






7/28 8/3 8/8 8/13 8/18 8/23 8/28 9/2


1008060



40200



100. > 80
u
z
60


4020



01


54*


30*


101


K


108 93


B


60


8/6 8/11 8/16 8/25 9/3 9/11


61* 98

47*
102












8/9 8/14 8/19 8/24 8/29 9/4


91 -83


C


72

74








9/9 9/14 9/19 9/30


DATE EGGS LAID


41


84A


















9/8


_49' 8/2


10080-


60-j


4020-


.L-









42




100-47 69 7875 73 D


80 824 40*

60- 35*

65
40


20


0 L
8/13 8/18 8/23 8/29 9/4 9/10 9/16 9/22 9/28 10/4


# 100
E
103 113
80
46 43 78 Z 60 42
w

0 40 61


L- 20



8/13 8/17 8/22 8/28 9/3 9/8 9/13
0

100. 58 54 108 92
_52' -88 _96 -73 F
801 40


60


40


20


01 - -
8,14 8/19 8/25 8/31 9/6 9/12 9/18 9/26 10/3 DATE EGGS LAID
F. -Lre 3-2--Continued


I









43


100- G

53
80- 84
48*
91
60. 93
29
87
40 43


20


0
8/5 8/14 8/22 W30 9/6 9/13 9/20 9/28 1C/8


ON 100.
0H 15 %N 35* > 80 61 112 u 109 92 101
z
Wj 60

73 81 WL 40 LL
20



8/9 8/13 8/18 8/23 8/28 9/3 9/10 9/15 9/20 9/25 10/2


100 56 90


80 47 33 44 52

60- -7 31 20 77
60


40


20


0 L
8 15 8,20 8/25 8/31 9/6 9/13 9/18 9/24 10/1 108

DATE EGGS LAID

Fire 3-2--Continued








44


100- 11034


8() _43'

55 61* 52
60 105 41
40 100
74
40


20



8,6 8/14 8/22 8/30 9/6 9/13 9/20 9/26 10/4 10/11 10/18


O 100 61K 0
89
> 0-57' 48*
o 80
z 65 WU 60-5

0
w
40. 85 5


20







100
65*
80- 32* 99


60 53* 87 62 63 51
40


20


0
8/14 8/20 8/24 8/30 9/5 9/11 9/16 9/22 9/28


DATE EGGS LAID
Fi,,ure 3-2--Continued


|




































Figure 3-3. Fourth instar color morph frequencies of successive broods reared on peanuts and water at 230C (1983 data set). Each set of bar graphs includes rearings of all fertile egg masses laid during the lifetime of one female mated to one male. The broods of 3 different mating pairs (A-C) are represented. Number of 4th instars indicated above bars. Two broods reared at 21 0C indicated by *.









46


100 A ~green
100- intermediate
Black
80- 74 94 78
60- / / 8





08





__100- B


o3 75 40- 8












20'
7/25' 8/5188/6 8/12 8/1798/239813099/4











100 C


80

69
60
64 46 40 76







8/1 8/9 8/17 8/23 9/2 9/11 9/20 9/28


DATE EGGS LAID












Table 3-1. Comparison of consecutive egg masses in which the proportion of black morphs sharply increased. Mating pairs correspond to those in Figure 3-2. Egg operculum diameters are means + I S.E. for a sample of 35 eggs.


Date
Mating Eggs Pair Laid


D 9/28
10/4

F 9/26
10/3

G 9/20 9/28

I 9/24
10/1

J 9/20 9/26

K 10/2
10/9

L 9/5 9/11


% Black 4th Instars


0.0 38.5

0.0 77.5

16.1 78.6

0.0 70.0

20.3 56.1

26.6 96.7

12.6
46.0


Total No. Eggs


66 73

96 62

88
107

89 72

110
94

103 93

103 103


Surviving
% To 4th Hatched Instar


83.3 100.0

97.9 100.0

98.9 96.3

68.5 50.0

89.1 88.3

96.1 90.3

96.1 99.0


85.5
47.9

77.7
64.5

100.0 81.6

85.2 86.1

75.5
49.4

79.8
71.4

86.1 61.8


Egg
Di ame ter


0.572 + 0.0021 0.579 + 0.0023

0.557 + 0.0028 0.565 + 0.0023

0.589 + 0.0027 0.588 + 0.0020

0.593 + 0.0020 0.587 + 0.0021

0.545 + 0.0027 0.552 + 0.0027

0.583 + 0.0024 0.589 + 0.0029

0.588 + 0.0030 0.585 + 0.0027









48

only four out of seven comparisons. However, there was significantly lower survivorship in the broods with the higher proportion of black morphs (Wilcoxon T = 1, N = 7, p < 0.025, 1-tail). Egg size varied significantly with respect to parentage, but there was no effect attributable to the differences in color morphs (Table 3-2).



Effect of Parente on ColorMop Frequencies

Parentage had an influence on offspring 4th instar color morph frequencies (Table 3-3, 3-4). That is, broods from the same parents were more similar in coloration to each other than to broods from the general population. Differences between broods of sibling parents of the summer 1981 data set (Table 3-3) were not as strong (0.05 < p <

0.10) as those found in -the fall 1981 data set (Table 3-4, p < 0.01), where most parents were unrelated.

In addition, parental influence was significantly related to the

4th instar coloration history of the parents (Table 3-5). Parents which had been green during the 4th instar had broods with higher green morph frequencies and lower black morph frequencies than parents which had been black in the 4th instar. However, one pair of black morph parents had two egg masses which yielded more than 90% green morphs and no black morphs. One egg mass of one green morph parent resulted in 82% black morphs and only 6% green.



Effect of Parental Diet on Color Morph Frequencies

Five out of nine mating pairs reared on peanuts and water failed to produce viable eggs, and only one pair produced normal-looking egg masses (Table 3-6). In contrast, all mating pairs reared on the other








49


Table 3-2. ANOVA of egg diameter (mm) on two factors: the resulting frequency of black 4th instar morphs (low or high) and the parents of the egg mass.


Source of Variation Sum of Squares df Mean Square F p Frequency of Black Morphs 0.000817 1 0.000817 3.69 >0.05 Parentage 0.106533 6 0.017756 80.25 (0.01 Interaction 0.003345 6 0.000558 2.52 <0.05 Error 0.105308 476 0.000221


I








50


Table 3-3. Variation in frequency (%) of green morphs in the 4th instar, with respect to parentage. Summer 1981 data set. Frequencies are listed in the order egg masses were laid, but some broods are missing from the sequences. Nymphs reared on green beans and peanuts. The effect of parentage is strong but not significant (Kruskal-Wallis H = 12.8, df = 7, 0.5 < p < 0.10). Fourth instar color morph of parents (male, female) are indicated by G (= green), I (= intermediate), and B (= black).


Parental Parents Morphs Ti-1 (F)* G,G T2-1 (C)* G,G


T5-1


B,B


T7-1 (A)* B,B


T5-3


T4-2 (E)*


B,B I, I


T7-2 (B)* B,B T7-3 (D)* B,B


% Green Morphs in Each Brood


88 92 100 45 93 86 95 25 26 64 94 79 52 38 51


50 40 74


81 52 56 27

21 71 57 71 12 54 23 11


*Indicates same set of parents found in Figure 3-1.


66

50


58


Mean


78.2 69.8 61.3 55.6

54.7 51.0

45.7 21.8


39


31 23


9








51


Table 3-4. Variation in frequency (%) of green morphs in the 4th instar, with respect to parentage. Fall 1981 data set. Frequencies are listed in the order egg masses were laid, but some broods are missing from the sequences. Nymphs reared on green beans and peanuts. The effect of parentage is significant (Kruskal-Wallis H = 33.9, df = 13, p < 0.01). Fourth instar color morphs of parents (female, male) indicated by G (= green), I (= intermediate), and B (= black).


Parental Parents Morphs


X7

G8 z 1 G9


ZlO

X5 X4


X6

G4


I,B

G,G

G,G

G,G G,G G,G


?7


G, I

B,B G,G

B,B B,B B,B


Zll

Z14


Z2


Z15


% Green Morphs in Each Brood


64 100 89 100 100 90 63 67 93 72 79 100 97 41 88 48 67 58 61 84 65 79 38 66 67 70 72 49 67 32 71 37 15 48 66 61 72 59 29 52 93 40 35 28 51 32 31 58 14 15 45 31 25 29 40


Z13 B,B


Mean


84.3 84.0 81.3 79.3 67.7 67.0 65.3 51.2 50.2

49.0 38.0 32.5 31.3


0 9 5


4.7








52


Table 3-5. Influence of parental color morph in the 4th instar on the color morph frequencies of their offspring. Nymphs were reared on green beans and peanuts. Values given are averages of the first egg masses produced by each mating pair. The number of egg masses used to determine each value is given in parentheses. Nymphs reared on green beans and peanuts. U = Mann-Whitney statistic.


Average Morph Frequency Among Offspring


Green Morph


Green Parents Black Parents


Black Morph


Green Parents Black Parents


SUMMER 1981


97.6 (2) 93.1 (3) 91.1 (3)


Grand Mean = 93.9


U = 0, p = 0.012


U = 0, p = 0.012


FALL 1981


98.6 96.8 88.2 82.9 79.3 55.8 55.0
49.6 49.2


(2)
(3)
(1)
(2)
(3)
(3)
(1)
(3)
(3)


Grand Mean = 72.8


93.6
43.1 41.5 34.6 34.1 25.0
4.4


(2)
(3)
(2)
(3)
(1)
(1)
(2)


36.4 32.3 30.0 28.3 10.8 9.5 1.6 1.5 0.0


39.5


(3)
(3)
(1)
(3)
(3)
(2)
(3)
(2)
(1)


16.7


66.4 42.7 38.9 36.6
34.2 34.0 0.0


36.11


U = 10.5, p < 0.025


2.3 (2) 0.7 (3) 0.0 (3)


61.3 (3) 56.1 (3) 55.7 (2) 54.6 (3) 49.7 (3) 29.9 (3)


51.2


43.1 27.0 23.6
18.4 17.3 13.6


(3)
(3)
(3)
(3)
(3)
(2)'


1. 0


23.8


(2)
(3)
(1)
(1)
(3)
(2)
(2)


U = 7, p < 0.01








53


Table 3-6. Comparison of egg production of parents which had been reared and maintained on different diets. Parents were reared from a single aggregation of 2nd instar nymphs collected in the field.


Parents' Diet


Green Beans
Peanuts


Peanuts Water


Green Beans
Only


No. mating pairs set up

No. pairs laying normal,
viable eggs

No. viable egg masses
produced in lifetime
(range)

Interval between egg masses
(days)

No. eggs per egg mas **
n
Average
Range


Hatch success (%) per egg mass** n 12 Average 97.5 Range 93.9 100.0


3
92.4
87.3 98.2


7
76.8
52.3 100.0


*Five pairs laid no viable eggs. Three other pairs laid 2 4 small, mostly inviable egg masses from which a few nymphs hatched.

**Includes only the egg masses reared through to 4th instar. See Figure 3-4.


6

6


6 10



5 -8



12
109.7 59 132


9


8



5 6



3
114.0 110 121


3

3


3 -6



8 18



7
86.4 41 111








54


diets reproduced normally. The number, size, and hatch success of egg masses laid by the one peanuts-and-water pair fell within the range of parents reared on green beans and peanuts. On the other hand, the three pairs reared on green beans alone produced fewer egg masses at longer intervals. These egg masses were slightly smaller and less viable on the average.

Despite the differences in laying interval and viability, the 4th instar color morphs of broods from parents reared on green beans alone did not differ from broods of parents reared on green beans and peanuts (Figure 3-4). The only significant difference was for the broods of the parents reared on peanuts and water: the three half-broods reared on green beans and peanuts had much higher proportions of green morphs.






Discussion



Variation between Successive Broods

Environmental variation. The striking differences in the ratios of color morphs that occur among broods of a single parental pair (Figures 3-1, 3-2, 3-3) indicate that nongenetic factors are important in the determination of 4th instar color. The greater variation of summer 1981 broods (Figure 3-1) may be because rearing conditions, particularly temperature and photoperiod, were less controlled than in 1983 (Figure 3-2). Cooler rearing temperatures bring about higher frequencies of black morphs in laboratory populations (Kariya 1961). Chapter 5 of this dissertation deals further with this effect. Short photoperiod stimulates increased melanism in a number of insects (e.g., Watt 1969,










green
-' intermediate
blac k


+


7 T


G,


80604020-


3


7


P, W G Nymph diet= P,W


Figure 3-4. Effect of parental diet on 4th instar color morph frequencies of their offspring. All parents were brood siblings reared and maintained from the 2nd instar on the diets indicated (1983 data set). Each egg mass was divided, and each half of the nymphs reared on different diets. G = green beans; P = peanuts; W = water. Number of egg masses indicated above bars. Vertical lines delimit + I S.E.


100-


be





U.

M
CL LL 0.
E.


F


[17


12


t.


Parent's Diet -+ G,P


P, W G Nymph diet= G, P


3


I


12








56


Ishii and Hidaka 1979), and nymphs in summer 1981 may have perceived changing day length because of a laboratory window.

Nutritional variation. The summer 1981 rearings (Figure 3-1)

clearly show that rearing nymphs on the peanuts and water diet results in increased incidence of melanism in the 4th instar. The effects of diet are examined further in Chapter 4. Purchased green beans differed in maturity and probably nutrition from week to week, and Nezara does show a preference for young fruit over mature in the field (Drake 1920). I tried to minimize this variation by purchasing enough beans at once to feed all cultures for several days. Peanuts varied less because a single bag could supply nymphs for weeks. However, in 1981 rapid growth of fungus on old peanuts sometimes occurred between cleaning periods. I often found nymphs feeding on those infested peanuts. In 1983, I successfully prevented fungus outbreaks by removing old peanuts sooner. The water source provided for nymphs on the peanuts and water diet might have retained salivary enzymes of drinking nymphs. Such a variable chemical cue might account for some variation between broods reared on peanuts and water.

Maternal effects. Despite the possible influences of rearing

conditions on brood color morph frequencies, there are indications that maternal factors affect the variation between successive broods. Broods from egg masses laid on about the same day by different parents should experience very similar rearing conditions. If conditions change to favor black morphs, then all broods should exhibit an increase in black morphs relative to the preceding broods. This was not always the case (Figures 3-1, 3-2, 3-3). In fall 1983 (Figure 3-2), for example, pair E's last brood on September 13 had a marked shift toward darker morphs,








57


while pair F's September 12 brood had a high green morph frequency, consistent with preceding and succeeding broods.

With the exception of pair A, all of the brood series reared on green beans and peanuts in 1983 (Figure 3-2) showed distinct shifts toward darker morphs in the last one or more viable egg masses laid before the female's death, independent of the date. A similar trend was observed in 1981 (Figure 3-1). Offspring of older Nezara viridula females tend to have lower larval survivorship, lower adult longevity, and lower fertility than offspring of younger females (Kiritani and Kimura 1967). My data (Figure 3-1, 3-2) provide evidence that female senility also affects coloration of offspring. Wellington (1965) showed that the quantity of yolk allocated to tent caterpillar eggs was closely related to the activity and vigor of the hatchlings. The last eggs laid by a female had less yolk and produced less vigorous hatchlings. However, Nezara egg mass size, egg size, and hatch success did not vary significantly when sudden increases in black morphs occurred in a brood series (Table 3-1, 3-2). On the other hand, nymph survivorship to the 4th instar decreased in six out of seven cases, and increased by only

0.9% in the one other case (Table 3-5).

Although increases in black morphs were not preceded by measurable changes in egg size, a maternal effect may still have been transmitted in the yolk. In addition to providing lipids, proteins, and carbohydrates, maternal cells of Oncopeltus (Lygaeidae) transfer RNA to developing oocytes (de Wilde and de Loof 1973a). Yolk deposition in Rhodnius (Reduviidae) and other insects is activated by secretions from the corpus allatum that seem identical to juvenile hormone (de Wilde and de Loof 1973b). Corpus allatum activity is reduced in diapausing








58

insects (de Wilde and de Loof 1973b) and may change with aging. Female cecropia moths (Saturniidae) supply significant quantities of juvenile hormone to their eggs, but the significance to embryonic and nymphal development is unknown (Gilbert 1964, Gilbert and King 1973).



Effect of Parentage on Average Color Morph Frequencies

In spite of the brood-to-brood variations, there was an overall

significant effect of parentage on the coloration of broods (Table 3-3, 3-4). The effect was not as strong among the sibling parents of summer 1981 (Table 3-3), where broods from half of the parents averaged 51-61% green morphs. In the fall 1981 data set, the range of green morph frequencies was wider and more evenly distributed (Table 3-4). Here there were two sets of sibling pairs (Zi, Z2 and Zl, Z13, Z14, Z15), and the other pairs were unrelated matings.

The parental effect on nymph color morph was definitely associated with the color morph of the parents (Table 3-5). On the average, the broods laid by parents which had been black nymphs themselves in the 4th instar had relatively more black morphs than the broods of parents which had been green morphs. However, the ratios of morphs obtained are not typical of other genetically controlled polymorphisms. For example, melanic patterns of Coelophora inaequalis (F.) (Coleoptera: Coccinellidae) are controlled by eight alleles of a single gene (Houston and Hales 1980). Numerous intermediate forms are heterozygotes, but the blackest morph is a simple homozygote. Nezara 4th instar black color cannot be a simple homozygotic trait because black morph crosses did not result in 100% black morph offspring (Table 3-5). Likewise, if inheritance of 4th instar color pattern were polygenic (Strickberger








59

1968), then black-black crosses would at the least produce a majority of black morphs. Instead, most crosses produced less than 50% black morphs and one cross produced no black morphs at all (Table 3-5).

Therefore, the parental effects observed are either the result of maternal inheritance or a complex system of modifier genes. Maternal transmission of environmentally influenced polymorphisms has been observed in other insects. Melanism induced by low temperature in the adult wasp Habrobracon and the lygaeid bug Oncopeltus can be passed on to the next generation (Wigglesworth 1965). On the other hand, there are examples of melanic polymorphisms which vary with environment but are also influenced by genetic inheritance. These include a noctuid moth larva (Long 1953), a tortricid moth larva (Baltensweiler 1977), a geometrid moth (Majerus 1981), Colias butterflies (Roland 1982), and a syrphid fly (Heal 1979).



Effect of Parental Diet

Parental diet had a definite effect on fertility (Table 3-6), but its relationship to offspring coloration is still unclear. Most adult pairs reared on peanuts and water were infertile. Yet, egg mass size and hatch success of masses from the onp fertile pair were similar to egg masses laid by parents reared on green beans and peanuts (Table 3-6). Egg masses laid by parents reared solely on green beans were smaller and had lower hatch success than eggs of other parents. Nevertheless, out of all of the half-broods reared on either diet, only the broods laid by the peanut-fed parents were different (Figure 3-4). Their half-broods reared on green beans and peanuts had twice the frequency of green morphs in comparison to nymphs from the other








60


parents. Unfortunately, it is impossible to separate genetic and maternal causes in this case. Further repetitions would be difficult to obtain because there is such high infertility among parents reared on peanuts and water. A better approach might be to rear nymphs on the standard diet and switch adults to different diets.



Conclusions

Nezara viridula adults contribute significantly to the distribution of color morphs among their 4th instar offspring. In particular, the parents' coloration during their own 4th instar is associated with higher proportions of that color type among their offspring. However, this research did not determine the exact mode of inheritance, whether genetic or maternal. A maternal effect is suggested by a consistent pattern of higher black morph frequencies in the broods of senile females. In addition, higher black morph frequencies in successive broods are associated with lower nymph survivorship.

Genetic and nongenetic parental effects are difficult to assess because of the flexible response of nymph coloration to environment. Nezara nymph diet (Figure 3-1; Chapter 4) and rearing temperature (Kariya 1961; Chapter 5) can cause as much variation between two halves of a single brood as can be found between broods of any two parental pairs. Knight (1924) looked for patterns of genetic inheritance of the

color polymorphism of nymphs and adults of another pentatomid, Perillus bioculatus Fabricius. After three years of breeding experiments, he finally concluded that color patterns of individual bugs were more strongly influenced by external conditions, such as temperature. than by inheritance from parents. The same appears to be true for Nezara viridula nymphs.


















CHAPTER FOUR
EFFECT OF NYMPH DIET ON COLOR MORPH RATIOS





Introduction



Nymphs of the southern green stink bug, Nezara viridula, undergo dramatic changes in coloration during postembryonic development. Although all instars retain the same basic patterns of white and red abdominal spots, dorsal ground color changes from reddish brown in the 1st instar to black in the 2nd instar, and ultimately to green in the adult. The green ground color may be attained as early as the 3rd instar and is accompanied by reduced melanization of the head capsule, thorax, legs, and abdomen (Chapter 2). The consequence of the variation in the timing of this color change is that 3rd, 4th, and 5th instar populations can be polymorphic. Third instar nymphs are usually black and 5th instar nymphs are usually green, but 4th instar nymphs often range from black to green with intermediate stages. Jones (1918) provides descriptions of Nezara nymphs, including dark and light forms of the 4th and 5th instars. A comparison of morphs is detailed in Chapter 2.

This investigation is based on my initial observation that the frequency of melanization among laboratory-reared 4th instar nymphs differed from that of nymphs reared in an outdoor green bean garden, even though both groups were from the same parental stock. I then


61


MENEW








62


discovered differences that seemed to be associated with the type of laboratory diet the nymphs had received.

Food plant species, plant quality, and amino acid content of diet have been associated with differences in the degree of and frequency of occurrence of melanism in a variety of insects. Under crowded conditions, larvae of the moth Plusia gamma L. (Lepidoptera: Plusiidae) range from light green to almost black, depending on the amount of diffuse melanin in the cuticle (Long 1953). Long found that larvae reared on broccoli were distinctly less melanized than larvae reared on five other plant species. In solitary cultures, 20% of Erinnyis ello L. larvae (Lepidoptera: Sphingiidae) reared on Poinsettia were brown and 80% were green. However, up to 90% of larvae reared on Euphorbia were brown (Schneider 1973). In contrast to these examples, the polymorphic melanic pattern of larvae of Papilio demodocus (Lepidoptera: Papilionidae) in South Africa is not determined by diet, even though the distribution of pattern type is closely associated with food plant species in the field (Clarke et al. 1963). In that case, the authors suggest that visual predators have selected for different genetically-determined cryptic patterns on the different plant backgrounds.

In general, poor food quality increases melanism in insects. E. ello larvae were more commonly brown when fed defoliated Poinsettia shoots instead of fresh shoots (Schneider 1973). Nymphs of the grasshopper Paulinia acuminata (de Geer), which feeds exclusively on a single aquatic fern species (Salvinia), changed from green to brown or black in successive instars when switched from fresh green fern to old brown fern (Meyer 1979). A higher proportion of lime aphid nymphs, Eucallipterus tiliae (L.), developed black cuticular bands when they were reared on








63


mature leaves instead of young leaves (Kidd 1979). On the other hand, Hintze-Podufal (1977) concluded indirectly from her experiments that Saturnia (Eudia) pavonia L. larvae (Lepidoptera: Saturniidae) feeding on drier food plants become less melanized than those feeding on fresh, moist food.

Goldberg and De Meillon's (1948) study on nutritional needs of mosquito larvae (Aedes egypti L.) produced direct evidence that nutritional components of diet affect melanization. Larvae reared on artificial diets free of the amino acids tyrosine and phenylalanine did not develop normal melanic larval pigmentation.

In this study, I compare the effects of a diet of fresh green

beans, shelled raw peanuts, or a combination of the two on the frequency of occurrence of color morphs in the 4th instar. In addition, I test diet changes during the 2nd and 3rd instars to determine whether a sensitive period for 4th instar color determination exists. The effect of diet on development time and mortality also gives insight to the mechanism by which 4th instar color pattern is determined.






Materials and Methods



General Approach

All experiments were based on matched comparisons of two groups of siblings from a single egg mass. This protocol controlled for genetic or maternal factors that might influence color pattern development of nymphs. Although the number of nymphs varied in each repetition of an experiment, matched experimental pairs had close to the same number of

nymphs.








64


Parental Stocks

Nymphs and adults of Nezara viridula from Gainesville, Florida, populations were reared in the laboratory on a diet of green beans (Phaseolus vulgaris) and shelled raw peanuts (Arachis hypogaea) purchased in local supermarkets. In 1981, adults were kept as pairs in 100 mm plastic petri dishes or in groups of five pairs in quart mason jars. Every other day, I provided fresh food and clean paper toweling on the bottoms of dishes and jars, cleaning containers whenever necessary. Direct illumination was provided by incandescent and fluorescent lights; some indirect sunlight came in through a laboratory window. The light regime was 15L:9D. Room temperature was 25-27 0C. The parental stock for 1983 experiments was composed exclusively of adults collected in the nymphal stages. Virgin adults were placed together as permanent mating pairs in separate petri dishes. They were maintained as in 1981, except that each pair received fresh green beans daily, instead of every other day.

Adults usually laid eggs directly on the paper toweling on the

bottom of the petri dish or on strips of paper hanging in the mason jar. Eggs laid elsewhere were easily pried off with a razor blade and glued to paper toweling with white glue. This procedure had no effect on hatching success. I collected eggs daily, cutting out the small square of toweling around each egg mass. I kept these egg masses in petri dishes which had a 2.5 x 2 x 0.3 mm piece of cellulose sponge or cotton cheesecloth taped to the lid and moistened to provide humidity. In 1981, egg masses were assigned arbitrarily to different experiments, without respect to parentage. In 1983, however, all repetitions of each








65


experiment used egg masses from different mating pairs. In addition, the egg masses in each experiment were all the same rank order. For example, the first egg masses laid by 22 females were assigned to the first experiment; the second egg masses were assigned to another.



Experimental Procedures

Rearing experiments in 1981 were started in two ways: (1) One to two days before hatch, egg masses were broken into two parts having about the same number of viable eggs, and the toweling was cut between the halves. Each half was arbitrarily assigned to one of two treatments (or the same treatment if a control) and placed in a 100 mm petri dish with the appropriate diet (Figure 4-1). (2) The entire egg mass was allowed to hatch in a petri dish having only moistened sponge or cheesecloth. First instar nymphs appeared to behave normally and usually aggregated on or next to the water source by the second day. Within 24 hours of the first molt, the 2nd instars were divided arbitrarily into separate dishes with the appropriate diet treatment (Figure 4-1). The data from rearings begun in either of the two ways were pooled, since there appeared to be no consistent effects attributable to the starting technique. The egg-splitting technique was faster and easier, and was used for all 1983 experiments. All 1983 experimental treatments were started on the 4th day after eggs were laid.

Figure 4-2 outlines the experimental design used to determine whether a diet difference during the 2nd or 3rd instar affects color morph ratios.

In 1981, experimental and control dishes were kept in the same lab area as the adults and developing eggs. Every day, I supplied distilled














EXPERIMENTAL GROUPS (1981)
--------- Half A:


egg mass ---<


--------- Half B:


CONTROL GROUPS (1981


egg mass ---K


--------- Half A:


--------- Half B:


egg mass ---K


--------- Half A:


DIET TREATMENT



green beans (2 peanuts (2) water

peanuts (3-4) water



green beans (2 peanuts (2) water

green beans (2 peanuts (2) water

peanuts (3-4) water


Half B: peanuts (3-4)
water


EXPERIMENTAL GROUPS


(1983)


egg mass ---K


Half A: green beans
peanuts (2)


(2 segments)


--------- Half B:


CONTROL GROUPS (1983)
--------- Half A:
egg mass ---<
--------- Half B:


egg mass ---K


--------- Half A:

--------- Half B:


green beans (3-4 segments)


green beans (3-4 segments) green beans (3-4 segments) green beans (3-4 segments) further divided at 2nd instar into groups of 15 or fewer nymphs per dish, each with 3-4 green bean segments.


Figure 4-1. Experimental design for testing the effects of diet on color morph ratio. Green bean segments were 4-7 mm long; raw peanuts weighed about 1 g each. Water was provided by a moistened piece of sponge or cheesecloth attached to the lid of the rearing dish.


66


segments)


segments)


segments)








67


DIET TREATMENTS DURING EACH INSTAR


Second Instar


/
green beans -----> green beans --< peanuts peanuts


egg mass --> peanuts ---------> peanuts ------<
water water


-(A) green beans ------> / peanuts egg mass --> water only --<


-(E) green beans ------>
peanuts


-(A) peanuts ---------->
/ water
egg mass --> water only --K


Third Instar


--(A) green beans
peanuts


--(B) peanuts
water


--(A) green beans
peanuts


--(B) peanuts
water


green beans peanuts


peanuts water


green beans peanuts


-(B) peanuts ---------->
water


-(A) water
/ only egg mass -<


-(E) water
only


green beans ------> peanuts


green beans ------> peanuts


green beans peanuts


peanuts water


Figure 4-2. Experimental design for determining the effect of a diet switch in the 2nd or 3rd instar on the color morph ratios in the 4th instar. See Figure 4-1 caption for details of food quantities. Last treatment executed in 1983; all others run in 1981.


First Instar


egg mass -->


peanuts water


-------- >








68

water to the sponge or cheesecloth as needed and replaced moldy peanuts and green beans. Every other day, I transferred nymphs to a clean petri bottom with new paper toweling, fresh green beans, and new peanuts. During the 1983 experiments, nymphs were reared in an environmental chamber at 230C, with a 12L:12D light regime. Regular care of nymphs was the same as in 1981, except that I provided fresh green beans each day instead of on alternate days. I also recorded nymph deaths and the date of appearance of the first 4th instars in the dish.

Because the molt to the 4th instar is asynchronous, I chose to terminate rearings when 85-100% of the nymphs in a dish had molted. Usually only three or fewer 3rd instars remained. I assigned 4th instars to eight color morph categories based on the degree of melanization of the dorsal cuticle of the thorax (Figure 2-1). Pattern types ranged from unmelanized (only a few minute black dots) to completely black (except at the lateral margins, which remained golden yellow). The unmelanized nymphs had a light green ground color on the thorax and abdomen. Nymphs with black thoraxes had very dark black-brown abdomens, although the only melanized areas were black margins around the medial scent gland openings and semicircular markings around the connexivum (margin of the abdomen). Nymphs in the middle color morph categories usually had a more golden-yellow color in the unmelanized part of the thorax, and their abdomens were a very dark, blackish green.



Data Analysis

I combined the eight original color morph categories into 3 major categories for analysis:








69


(1) green: the two lightest colored categories (I-II, Figure

2-1); the maximum extent of melanization is a central U-shaped

pattern of black lines, a few spots, and finger-like patterns

projecting inward from the lateral margins; thorax ground

color green; abdomen ground color light green to dark green.

(2) intermediate: the three middle categories (III-V, Figure

2-1); black markings on thorax extensive, but covering less

than 3/4 of area; thorax ground color green to golden yellow;

abdomen ground color dark green to black-brown.

(3) black: three darkest categories (VI-VII, Figure 2-1); more

than 3/4 of thorax black (except for yellow lateral margins);

unmelanized spots are golden-yellow; abdomen ground color

black-brown.

I calculated color morph frequency for each half of an egg mass by dividing the total number of 4th instar nymphs in a category by the total number of 4th instar nymphs counted. Frequencies are expressed here as percentages. Differences between treatments were tested with the Wilcoxon Matched Pairs Signed Ranks Test (Siegel 1956).

Percent mortality during each instar was calculated by dividing the number of deaths at the end of the instar by the number of nymphs present at the beginning of the instar. Differences were tested with the Wilcoxon Test.

Development time for an egg mass half was defined as the number of days between egg-hatch and the first appearance of 4th instar nymphs in the group. Differences between two halves were tested by the Sign Test (Siegel 1956).








70


Resu Its



Single-Item Diets

The frequency of melanism is greater among 4th instar nymphs reared on a single food item than on the combined green bean plus peanut diet (Table 4-1). Compared to the combination diet, the peanut diet resulted in fewer green morphs and more black morphs, but about the same frequency of intermediate morphs. The green bean diet also led to a shift toward darker morphs in comparison to their siblings reared on the combination diet. There were no significant differences between the morph frequencies of sibling groups reared on identical diets.

Nymphs reared on green beans alone produced liquid waste at a much faster rate than their siblings on green beans and peanuts. The 30-70 2nd or 3rd instar nymphs from one half of an egg mass usually produced enough waste in 24 hours to completely saturate three layers of paper toweling. Their siblings which were feeding on both green beans and peanuts only slightly dampened the paper linings. To check the effect of increased humidity and waste product, I ran an additional eight control masses on the green bean diet. For these controls, I divided the 2nd instars from one of the half-mass groups into separate dishes containing no more than 15 individuals. These "uncrowded" nymphs did not differ significantly from their "crowded" siblings with respect to morph frequency (Table 4-2). In fact, the crowded group had more green morphs and fewer black morphs in five out of the seven cases with differences.








71


Table 4-1. The effect of diet on morph ratios. Samples are matched by assigning different treatments to each half of an egg mass. Numbers 1
3 are 1981 experiments; 4 and 5 are from 1983. Values in parentheses are standard errors of the mean. GB = green beans, PN = peanuts, W = water, N = number of cases with differences, T = statistic for Wilcoxon Matched Pairs test. SS = p < 0.01, S = p < 0.05, NS = p > 0.05


Treatment


1. GB, PN, W
PN, W


No. of Mean No. Egg of 4th Masses Instars


36 41.9 + 1.8 38.4 + 2.1


Mean Percent of Each Morph Green Intermediate Black


68.4 + 4.1 24.4 + 4.4


14.5 + 1.7 17.2 + 1.8


N=36,T=0 N=34,T=252


SS


2. GB, PN, W
GB, PN, W


3. PN, W
PN, W


4. GB, PN
GB


5. GB
GB


32 41.8 + 1.6
41.3 + 1.4


26 37.7 + 2.2
37.4 + 2.1


22 46.9 + 3.6
42.7 + 2.8


15 47.4 + 2.8
46.5 + 2.7


67.1 + 5.0 64.8 + 5.1

N=32,T=210
NS

40.7 + 5.2 38.6 + 5.5

N=24 ,T= 18
NS

82.1 + 4.0 39.6 + 5.7

N=20,T=0
SS

32.9 + 5.4 31.1 + 6.3

N=14,T=47
NS


NS


14.4 + 1.7 15.1 + 2.0

N=32,T=259
NS

24.5 + 2.4 21.9 + 2.2

N=25 ,T=12 I
NS

12.2 + 2.6 21.9 + 2.2

N=20,T=44
S

24.5 + 2.2 24.2 + 2.8

N=15 ,T=64
NS


17.1 + 3.0 58.4 + 5.2

N=35 ,T=3
SS

20.8 + 4.1 17.2 + 4.2

N=28,T=127
NS

34.8 + 5.1 39.5 + 5.7

N=24,T=102
NS

5.8 + 1.8 38.5 + 5.3

N=19 ,T=0
SS

42.5 + 4.7 44.7 + 6.0

N=13,T=40
NS








72

Table 4-2. Effect of rearing nymphs in a single dish ("crowded") or in groups of less than 15 ("uncrowded") with a diet of solely green beans. Matched by dividing single egg mass into 2 treatment groups. N = number of cases with differences, T = Wilcoxon statistic, NS = p > 0.05


Treatment of Each Half


Crowded Uncrowded


No. of egg masses tested Mean no. of 4th instars


8


40.6 + 4.0


39.5 + 2.5


Mean % green morphs


Mean % intermediate morphs


9.6 + 5.4 5.4 + 2.7

N = 7, T = 7, NS


17.5 + 4.3 11.1 + 4.1

N = 7, T = 5, NS


Mean % black morphs 72.9 + 9.0 83.5 + 6.1 N = 7, T = 6, NS








73


Switching Diets

A diet switch during either the 2nd or 3rd instar affected the

outcome of morphs in the 4th instar (Figure 4-3). Nymph groups switched from peanuts and water to green beans and peanuts at the beginning of the 3rd instar had more green morphs and fewer black morphs than their sibling groups which remained on peanuts and water (Figure 4-3, Treatments 2 and 4). However, the reverse diet change (from green beans and peanuts to peanuts and water) only had an effect when 1st instars were unfed (Figure 4-3, Treatments 1 and 3).

In comparison to siblings which received green beans and peanuts during both the 2nd and 3rd instars, nymphs switched to peanuts and water during the 2nd instar showed a significant shift towards black morphs (Figure 4-3, Treatment 5).



Mortality and Growth Rate

Mortality in the 2nd and 3rd instar was higher among nymphs reared on peanuts and water compared to their siblings reared on green beans and peanuts (Table 4-3). Mortality among 2nd instar nymphs reared on green beans alone was significantly higher than their siblings reared on green beans and peanuts. Overall mortality in all experimental groups was extremely low.

Nymphs reared on single-item diets developed more slowly than those reared on the combination of green beans and peanuts (Table 4-4).









green intermediate black


100




75



Cr
"w 50

L
0

25






3rd
2nd 1st


(2)


GB, PN W, PN
W, PN W, PN n=8


+


FiLure 4-3. Effect of diet change on 4th instar coloration. GB = green beans, PN = peanuts, W = water. Significant differences between morph frequencies of treatment pairs determined by the Wilcoxon Test. *, **, and indicate p < 0.05, 0.025, and 0.005. n = number of egg masses tested. Treatment 5 was executed in 1983; all others were run in 1981.


+


+


(1)
GB, PN W, PN
GB, PN GB, PN
n=9


(3)


W n= 6


GB, PN GB, PN


W, PN GB, PN


(5) GB, PN
GB, PN
W


GB, PN W, PN
W


n=22


(4)
GB, PN W, PN W, PN W, PN
W
n=5


I











Table 4-3. treatments t


Deaths of nymphs reared on different diets. Samples are matched by assigning different o each half of an egg mass. Values in parentheses are standard errors of the mean. GB =


green beans, PN = peanuts, W = water, N Wilcoxon matched pairs test (one-tail).


= number of cases with differences, T = statistic for


No. of Egg Masses
Tested


20


22


Mean No. Nymphs Hatched


49.5 + 3.1 47.5 + 2.8


49.1 + 3.5 47.3 + 3.4


Mean % Death in Each Instar


1st 2nd


5.1 + 1.1 2.9 + 1.3

N=16,T=39 p > 0.05


1.5 + 0.7 1.2 + 0.7

N=8 ,T=20 p > 0.05


5.3 + 1.3 16.3 + 5.7

N=16 ,T=24 p K 0.01


0.5 + 0.3 1.5 + 0.4

N=12,T=1 7 p = 0.043


3rd


1.6 + 0.6 8.5 + 3.5

N=14 ,T=1 I p < 0.005


1.0 + 0.6 1.6 + 0.6

N=9,T=14 p > 0.05


Mean
% Deaths 1st 3rd


11.3 + 2.3 21.6 + 6.0

N=17 ,T=40
p = 0.045


2.9 + 1.0 2.0 + 0.5

N=16, T=48 p > 0.05


Treatment


GB, PN PN,W


GB, PN
GB








76

Table 4-4. Development time (number of days) from egg-hatch to 4th instar for nymphs reared on different diets. Observations were made once per day in late afternoon. Recorded time is appearance of the first 4th instar nymph(s) in the group. GB = green beans, PN = peanuts, W = water.


GB, PN W, PN diff.


17.9 -1.3

0.26 0.24 20 20


N = 15, x = 0, p < 0.004


GB, PN GB diff.


15.7 17.4


-1.7


0.18 0.25 0.24

22 22 22

N = 19, x = 0, p < 0.002


Mean S.E.


16.6 0.18


n


20


Sign Test








77


Discussion



Diet-Induced Melanism

These results differ from previous studies by showing that in

comparison to a single item diet, a combination of foods can reduce the occurrence of melanization (Table 4-1). In addition, Nezara nymphs are melanic in the earlier instars and make a transition to an unmelanized cuticle, whereas previous studies have been concerned with non-melanic immatures which develop melanization in later stages (Long 1953, Hintze-Podufal 1974, 1977, Kidd 1979).

Under the conditions tested, once a Nezara nymph had molted to an unmelanized cuticle in the 3rd, 4th, or 5th instar, successive molts were also unmelanized. I could not induce melanic adults. Meyer (1979) was able to induce dark grasshopper nymphs to molt into green nymphs and back again by changing the diet. He also obtained dark adults from dark nymphs. Hintze-Podufal (1977) reversed the direction of melanization between the 4th and 5th instar larvae of E. pavonia under some experimental conditions. However, melanic aphid nymphs (Kidd 1979) and 4th instar moth larvae (Schneider 1973) retained their developed melanism throughout the remaining immature stages.

The melanization effect does not seem to be caused by the increased humidity from the artificial water source or the green beans. The green bean and peanut diet resulted in fewer melanized 4th instar nymphs even when the water source was included as part of the treatment (Table 4-1, Treatment 1). Nymphs fed only green beans showed the same degree of melanization whether in a large group in one dish where moisture quickly accumulated or in small groups in dry containers (Table 4-2).








78


Timing of Color Determination

The color pattern expressed during the 4th instar seems to be determined by the length of time nymphs are allowed to feed on both green beans and peanuts (Figure 4-1). In the cases where significant differences in color morph frequency occurred, the group which had fed on both green beans and peanuts for the longer time had the higher proportion of non-melanics (green). Changes in food type during either the 2nd or 3rd instar affected the ratio of color morphs in the 4th instar. Thus, there is not a single "critical time" that determines the pattern. Hintze-Podufal (1977) found that changes in experimental conditions in the 2nd or 3rd instar affected the 4th instar color pattern of a saturnid larva, and conditions in the 3rd and 4th instar affected the 5th instar color pattern.



Nutritional Deficits and Excesses

The higher mortality among the 2nd and 3rd instars reared on

peanuts and the 2nd instars reared on green beans (Table 4-3) suggests that these single-item diets are nutritionally poorer than the combination diet. The longer development time of nymphs reared on the single-item diet (Table 4-4) also points to a nutritional deficit. Green beans are about 85% water compared to 5% water in peanuts and by weight contain only about 10% of the protein content of peanuts (Souci 1981). They also differ in amino acid and vitamin content by weight and relative proportion. Researchers who rear large laboratory colonies of Nezara viridula have best results with a regular diet of green beans and raw peanuts, with other items added from time to time (Harris and Todd








79


1981). Nezara reared on artificial diets take longer to mature and suffer higher mortality (Jensen and Gibbens 1973).

Nutrition has been linked to other kinds of polymorphisms in insects, such as castes in social insects (Wigglesworth 1954). In Drosophila, the penetrance of the mutant character tetraltera increased when flies were reared on food which prolonged development (Wigglesworth 1954). Kidd (1979) suggested that the decline in soluble nitrogen in mature leaves was responsible for increased melanization among lime aphid nymphs. The important role of the amino acid tyrosine in the production of normal dark pigmentation in Aedes mosquito larvae (Goldberg and De Meillon 1948) can be explained by the fact that tyrosine is a melanin precursor (Neville 1975, Kiguchi and Kimura 1981). Because quinones and phenols are used in the polymerization process that forms melanin (Bursell 1970), melanization may be a means of disposing

excess quantities of these potentially toxic by products of metabolism (Wigglesworth 1965).



Hormonal Role

The hormones responsible for insect growth and development also play a role in melanin production. Tyrosine is present in high concentrations in moth larval tissues immediately before ecdysis, and its release from the fat body seems to be caused by ecdysone (Kiguchi and Kimura 1981). Juvenile hormone does not affect this tyrosine release (Kiguchi and Kimura 1981) but does prevent melanization in at least two moth species (Kiguchi 1972, Hintze-Podufal 1976). The presence of the hormone bursicon is essential for the sclerotization of the cuticle, and its concentration may be correlated with the degree of melanization (Neville 1975).








80


Thus, diet may be affecting Nezara nymph color patterns directly, by providing chemical precursors for melanin production, or indirectly, by affecting the nymph's production of ecdysone, bursicon, or juvenile hormone. When nymphs are restricted to only one food item, they may have to take in excess amounts of certain amino acids or other melanin precursors in order to obtain minimum amounts of essential nutrients. Therefore, one would expect more melanism in nymphs which have been restricted to a single item diet for a longer time (Figure 4-1). Secretion of the hormone ecdysone is likely affected by the poorer diets, since development time is increased (Table 4-4). I have no data that links bursicon or juvenile hormone to the melanizing effects of diet. However, unmelanized cuticle is an adult characteristic of Nezara, and the completely melanized 3rd and 4th instar nymphs are identical in color and pattern to 2nd instar nymphs. Thus in this case, juvenile hormone may act to retain melanism, rather than prevent it.



Ecological Function of Diet-Induced Melanism

Does the melanization of the later instars provide any advantage to the nymph? The melanic cuticle may serve as a safe dumping site for toxic metabolites (Wigglesworth 1965). Melanin is known to add mechanical strength and to shield UV light (Neville 1975). It also strongly absorbs radiant energy in the visible spectrum (Watt 1968), and its possible role in thermoregulation is further discussed in Chapter 5. The melanized patterns may aid in avoidance and escape from visually-oriented predators. The most common escape response by a Nezara nymph is to simply let go and drop. A black or intermediate nymph would be cryptic if it dropped on to bare soil or leaf litter. A








81

green morph would more easily evade detection if it dropped on to green vegetation. This potential crypticity may be important, since some vertebrate predators do eat nymphs (Stam 1978).

In the typical seasonal pattern of Nezara, adults move to different host plants as they become available, and nymphs are free to feed on many plant parts (leaves, stems, flowers, and fruits; Todd and Herzog 1980). However, in late fall, suitable young host plants are no longer available. Since Nezara nymphs disperse very little (Panizzi et al. 1980), an egg mass deposited in a patch of senescent food plants would leave the nymphs with a limited diet of stems and mature pods. Since the nymphs are poikilothermic, additional solar energy absorbed by the melanized cuticle could help to counteract slower growth rates caused by either poor diet or cooler ambient temperatures. It is critically important for nymphs to reach maturity before winter begins because only adults are able to overwinter successfully (Drake 1920, Todd and Herzog 1980).



Conclusions

Diet has a significant effect on the expression of melanism in Nezara nymphs. This flexible color polymorphism provides a range of opportunities for further investigation into hormonal control of genetic expression and the ecological advantages of such flexibility. Rearing experiments on living plants are needed to test whether differences in availability of flowers, developing fruit, and/or mature fruit can produce differences in color morph ratios. Comparisons of field collections in young and senescent food patches at the same time of year should be made to confirm the diet effect in the field.

















CHAPTER FIVE
EFFECT OF TEMPERATURE ON COLOR MORPH RATIOS




Introduction



Pigmentation darkening in insects has often been associated with cooler temperatures (Wigglesworth 1965). Most often, the dark coloration is due to an irreversible increase in melanin pigment. The best studied examples are in the Lepidoptera. Early spring and late fall broods of Colias and Nathalis (Pieridae) have more melanic scales on the underside of the wings than mid-summer broods (Watt 1969, Douglas and Grula 1978, Hoffman 1978). Dark autumnal forms have also been described for a skipper (Hesperiidae) in Japan (Ishii and Hidaka 1979) and a noctuid moth in England (Myers 1977). The sycamore aphid has melanic pigmentation only in the spring and fall seasons (Dixon 1972). Melanic forms of some polymorphic caterpillars become more frequent at lower temperatures (Fye 1979, Baltensweiler 1977). In addition to seasonal dark forms, increased melanism has also been found among insects at higher latitudes and altitudes. For example, the frequency of melanics in alpine Colias (Roland 1982) and Scandanavian bumblebees (Pekkarinen 1979) is greater than in their congeners in lower regions. In the cases where temperature has been associated with within-species variation, short photoperiod and cold rearing temperature are the important factors that stimulate an increase in melanization. In one


82








83

saturnid moth caterpillar, however, melanics increased at higher rearing temperatures (Hintze-Podufal 1977).

Polymorphism in 4th and 5th instar nymphs of Nezara viridula is a result of variation in the amount of melanin in the cuticle and the shade and hue of subcuticular pigments in non-melanized areas. Coloration ranges from unmelanized forms, which have light green subcuticular pigmentation, to black forms, which have completely melanized thoraxes and black-brown subcuticular pigmentation. Jones (1918) provides complete descriptions of nymphs; diagrams of thoracic melanization patterns are provided in Chapter 2 of this thesis. The strong influence of diet on this polymorphism is discussed in Chapter 4.

In a brief experiment in October 1981, I compared laboratory

rearings on green beans and peanuts with caged outdoor rearings on green bean plants. All of the surviving 4th instar nymphs in the outdoor cages were black morphs, while many of their indoor siblings were green or intermediate in coloration. Earlier that summer, uncaged nymphs reared outdoors on green bean plants were usually less melanized than nymphs reared in the laboratory. Because night-time temperatures in October dropped below laboratory temperature, I hypothesized that temperature was a factor controlling melanization of Nezara nymphs.

This is a report of laboratory rearings of Nezara viridula nymphs at different temperatures and consequent shifts in color morph frequencies. In addition, I include a laboratory test of the influence of melanization on heat absorption during basking at low ambient temperature and a discussion of its possible adaptive significance.


|








84


Materials and Methods



Rearing Nymphs at Different Temperatures

A parental stock of Nezara viridula was formed from 4th and 5th instar nymphs from a single Gainesville, Florida, population collected in September 1982. I kept sexes in isolation until I united them as permanent mating pairs in separate 100 mm plastic petri dishes. Adults were maintained in the laboratory as described in Chapter 4.

I collected egg masses daily and assigned them by random number to one of four experimental treatment groups (Figure 5-1). After incubation in the laboratory for three days, I divided each egg mass in half along the longer axis, and randomly assigned each half to one of the two rearing temperatures. On the fourth day of development, I placed the eggs in 100 mm petri dishes with the appropriate diet and moved them to environmental chambers with 12L:12D lighting.

Nymphs hatched out after 1-2, 2-4, or 5-8 more days at 28, 23, and 18 0C, respectively. I monitored their progress daily, supplying fresh green beans at least every other day, and replacing peanuts when they became soft or moldy. I added fresh distilled water daily to the cheesecloth roll in the pearuts and water treatments. Every other day, I transferred nymphs to a clean petri dish bottom with fresh paper toweling. I recorded and removed dead nymphs daily and noted the date of first appearance of 4th instars in each dish.

I counted and preserved the nymphs in each dish as soon as (1) all nymphs reached the 4th instar, (2) any nymphs molted to the 5th instar, or (3) two days elapsed without further molting. I assigned the 4th instar nymphs to one of eight color morph categories according to the








85


Nymph Diet = Green Beans, Peanuts


Treatment Group 1


egg mass ---K


Treatment Group 2


egg mass ---<


--------- Half A

--------- Half B




--------- Half A

--------- Half B


280C 230C




230C 180C


Nymph Diet = Peanuts, Water


Treatment Group 3


egg mass ---<


Treatment Group 4


egg mass ---K


--------- Half A

--------- Half B




--------- Half A

--------- Half B


Figure 5-1. Experimental design for testing effect of rearing temperature on color morph ratio.


280C 230C




23 C 180C








86


degree of melanization of the thorax. (See Chapter 2 for complete descriptions.) For data analysis, I combined the eight original color type categories into the three major categories describe in Chapter 4: green, intermediate, and black. Color-type frequencies, percent mortality in each instar, and development time were calculated and statistically tested as in Chapter 4.



Cold Shock Experiment

From a second parental stock collected in July 1983, I obtained 20 egg masses. Each was the third egg mass laid by 20 different mating pairs. Three days after they were laid, I divided egg masses in half and randomly assigned each half to one of these treatments:

(1) constant temperature-- constant 23 0C; green beans and peanuts

diet; 12L:12D light regime.

(2) cold shock-- start out at 230C, but on first day of 2nd

instar, nymphs moved to 18 C; after 72 hours, nymphs moved

back to 230C; same diet and light regime as (1).

For an additional control, I took the fourth egg masses produced by the same set of mating pairs and randomly assigned each half to one of two groups (A or B). Both groups were reared at a constant 230C and with the green beans and peanuts diet.



Heating Curves

Using an indoor laboratory set-up, I compared heating rates of live 4th instar black and green morphs under artificial lighting. A horizontal platform was made of 0.5 mm thick styrofoam and covered completely with white paper. A 26 ga (0.46 mm) hypodermic needle








87


microprobe (Type MT-26/2, Bailey Instruments) was inserted through the styrofoam from below. Thus nymphs could be positioned horizontally on the platform surface and impaled by the probe from the underside. The light source was a 200 W incandescent lamp 22 cm directly above the platform surface. Nymph body temperature was measured with the microprobe and a digital readout instrument (Model BAT-12, Bailey Instruments), which has a 0.1 s time constant and a 0.1 C resolution. I measured shaded ambient temperature with a mercury thermometer shaded by aluminum foil and positioned 5 cm above the platform. The entire apparatus was kept in a constant temperature room, which maintained a constant 150C when the lamp was off.

Test animals were selected arbitrarily, with an effort to match sizes, from among black and green 4th instar nymphs reared from three unrelated egg masses. Each nymph was cooled on ice for at least 30 min, then quickly transferred to the platform. I inserted the probe through the underside at the midline, just at the posterior margin of the thorax and penetrating as far as possible (2-4 mm). I then secured the bug with a thread across the body, made taut by a weighted plastic ring. These steps took less than a minute, and the impaled bug was always several degrees below ambient temperature at this point. I allowed the nymph to warm up slowly to 14.70C; then I switched on the lamp to begin the run. Every 30 s, I recorded body temperature from the digital readout and shaded ambient temperature from the mercury thermometer. Most runs continued for 10 min, but some were stopped as early as 8 min if body temperature leveled off. All nymphs survived the

runs and behaved normally after they were released.








88


At least 10 min elapsed between runs to allow the apparatus to return to 15.1 0C. A dish of ice placed on the platform facilitated complete cooling. Black and green morphs were run alternately. I measured live weights on a Mettler balance after each run, rather than before, because the nymphs lost hemolymph when the probe was inserted. Since all factors were held constant in each run, except nymph color and size, I could make direct comparisons of nymph body temperature at any given time. The influence of size on the heating rate was determined by plotting body temperature at 8 min against live weight and was tested by the Spearman rank correlation coefficient (Siegel 1956). I used the Mann-Whitney U-test (Seigel 1956) to test for significant differences in body temperatures after 4 min and 8 min of exposure to light. I constructed heating curves by plotting body temperature against time.






Results



Lower rearing temperatures increased the degree of melanization among 4th instar nymphs (Figure 5-2). With a green bean and peanuts diet, 98% of nymphs were green at 28 0C while only 2% were green at 18 0C. Less than 1% were black at 28 C while almost 70% were black at 18 C. The average frequency of intermediates did not increase beyond about 30%, but greater frequencies did occur in individual cases. The same cooling effect occurred with the peanuts diet, except that at any given temperature, the proportion of melanics was greater. Also, the green, unmelanized morphs usually had a more yellowish green background color


than nymphs reared on green beans and peanuts.











green intermediate black


U U- I I- -


{


Green Beans and Peanuts Diet


Peanuts and Water Diet


Figure 5-2. Effect of rearing temperature on 4th instar color morph frequencies under two diet regimes. All differences between two temperatures (matched comparisons) are significant (p < 0.02, Wilcoxon Test), except for the intermediate morphs on the peanut and water diet at 28 and 23 C. n = number of egg masses reared.


100"


80



60.
C



40
CL
1L4

0


20




0


i1


+


280
n=16


230


230


180 n=10


0


23


180 n=8


280
n=15


0


23








90

While rearing these nymphs at the different temperatures, I noticed dramatic differences in 3rd instar color patterns between matched sibling groups. In addition to the typical black patterns, unmelanized patterns appeared. These were similar to the green and intermediate patterns of 4th instar nymphs, except that the background colors were usually more yellowish. I classified 3rd instar groups by overall aspect: all black, mostly black (some intermediates), mixed (few blacks or greens, many intermediates), and mostly light (few or no blacks, many greens and intermediates). Only nymphs reared at 280C fell into the mixed or mostly light categories, and their sibling counterparts at 230C were always classified as all black or mostly black. More groups with unmelanized 3rd instars occurred with the diet of green beans and peanuts (Table 5-1). In all but one of the unmelanized 3rd instar groups, more than 95% of the nymphs melted into green 4th instar morphs, regardless of diet (Table 5-1).

Development time increased with lower temperatures (Table 5-2).

The mean number of days from egg-hatch to the emergence of the first 4th instar approximately doubled for each 5 0C decrease. At any given temperature, nymphs reared on peanuts and water took longer to develop than nymphs reared on green beans and peanuts. These increases averaged only 0.85, 1.5, and 7.4 days at 28, 23, and 180C, respectively (Table 5-2). However, they are associated with major differences in color morph frequencies (Figure 5-2).

Average mortality from the 1st through 3rd instar increased

slightly between 280 and 230C, but only the 230 to 180C differences were significant (Table 5-3). Overall percent mortality at




Full Text

PAGE 1

COLOR POLYMORPHISM IN NYMPHS OF THE SOUTHERN GREEN STINK BUG, Nezara viridula (HEMIPTERA: PENTATOMIDAE) By CAROL BROWN JOHNSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OP' FLORIDA IN PAR1IAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1984

PAGE 2

To my parents Bryce and Lillian Brown for their endless love and encouragement

PAGE 3

ACKNOWLEDGEMENTS I am grateful to my supervisory chairman, Dr. Thomas C. Emmel, for his unflagging encouragement, helpful advice, and assistance with many practical matters in all phases of this research. Each member of my committee has spent hours discussing with me the various aspects of rny work. I thank them all for their valuable contributions. Dr. Pauline. Lawrence was especially stimulating with her knowledge and ideas about insect behavior and physiology. Her incisive questions and her positive attitude were always motivating. Dr. Reece Sailer was invaluable for his familiarity with Nezara and its natural history. He helped me gain access to fields and garden space and showed a constant enthusiasm for the problem. Dr. Carmine Lanciani gave me advice on experimental design and data analysis. He also provided me with several environmental chambers. I thank Steve Polasky, Harold Magazine, and Chris Victoria for their assistance in the care and maintenance of bugs in fall 1981. I could not have completed this dissertation without the love and constant support of my husband, Doug. He helped me in every way through all of the difficult times. Finally, I give thanks to God, whom I serve through the grace of Jesus Christ.

PAGE 4

TABLE OF CONTENTS PAGE ACKNOWLEDGEMENTS m LIST OF TABLES v i LIST OF FIGURES v m ABSTRACT ix CHAPTER ONE INTRODUCTION 1 Evolution and Maintenance of Color Polymorphisms 1 Color Patterns in Immature Insects 3 Coloration of the Southern Green Stink Bug 4 Objectives of This Study 5 TWO DESCRIPTION AND BIOLOGICAL COMPARISON OF NYMPHAL MORPHS ... 7 Introduction 7 Materials and Methods 8 Results 12 Discussion 20 THREE PARENTAL EFFECTS ON NYMPHAL COLOR MORPHS 28 Introduction 28 Materials and Methods 29 Results 35 Discussion 54 FOUR EFFECT OF NYMPH DIET ON COLOR MORPH RATIOS 61 Introduction 61 Materials and Methods 53 Results 70 Discussion 77 FIVE EFFECT OF TEMPERATURE ON COLOR MORPH RATIOS 8 2 Introduction g2 Materials and Methods g4

PAGE 5

Results 88 Discussion 94 SIX CONCLUSION 107 Control of Nezara Nymph Polymorphism 107 The Adaptive Significance of Nezara Nymph Polymorphism. . 108 Areas of Future Research 109 LITERATURE CITED HI BIOGRAPHICAL SKETCH 118

PAGE 6

LIST OF TABLES TABLE P AGE 2-1. Sex determination of 4th instar nymphs with respect to color morph 17 2-2. Sex determination of green and black 4th instar color morphs sampled randomly from a single egg mass .... 17 2-3. Sex of 4th instar nymphs collected from field populations 18 2-4. ANOVA of 4th instar dry weights of black and green morphs reared together from the same egg mass 18 2-5. Differences in morph frequencies between 4th instars which appeared earlier and later in the molting period ... 19 2-6. Duration of 4th instar in relation to color morph 21 2-7. Comparison of fertility of adult females which were different color morphs in the 4th instar 22 3-1. Comparison of consecutive egg masses in which the proportion of black morphs sharply increased 47 3-2. ANOVA of egg diameter on two factors: the resulting frequency of black 4th instar morphs and the parents of the egg mass 49 3-3. Variation in frequency of green morphs in the 4th instar, with respect to parentage. Summer 1981 data set 50 3-4. Variation in frequency of green morphs in the 4th instar, with respect to parentage. Fall 1981 data set 51 3-5. Influence of parental color morph in the 4th instar on the color morph frequencies of their offspring 52 3-6. Comparison of egg production of parents which had been reared and maintained on different diets 53 4-1. The effect of diet on morph ratios 71

PAGE 7

4-2. Effect of rearing nymphs in a single dish ("crowded") or in groups of less than 15 ( "uncrowded" ) with a diet of solely green beans 72 4-3. Deaths of nymphs reared on different diets 75 4-4. Development time from egg-hatch to 4th instar for nymphs reared on different diets 76 5-1. Color differences in 3rd instar nymphs reared at 28C. ... 91 5-2. Development time of nymphs reared at different temperatures 92 5-3. Deaths of nymphs reared at different temperatures 93 5-4. The effect of cold shock 95 5-5. Comparison of live weights and body temperatures of green and black morphs used to construct heating curves 97

PAGE 8

LIST OF FIGURES FIGURE pAG£ 2-1. Melanic patterns of thoracic cuticles of Nezara viridula 4th instar nymphs ~ ] ~, ...... 15 3-1. Fourth instar color morph frequencies of successive broods (Summer 1981 data set) 37 3-2. Fourth instar color morph frequencies of successive broods reared on green beans and peanuts at 23C (1983 data set) 4] 3-3. Fourth instar color morph frequencies of successive broods reared on peanuts and water at 23C (1983 data set) ,, 3-4. Effect of parental diet on 4th instar color morph frequencies of their offspring 55 4-1. Experimental design for testing the effects of diet on color morph ratio ,r 4-2. Experimental design for determining the effect of a diet switch in the 2nd or 3rd instar on the color morph ratios in the 4th instar 67 74 4-3. Effect of diet change on 4th instar coloration 5-1. Experimental design for testing the effect of rearing temperature on color morph ratio 85 5-2. Effect of rearing temperature on 4th instar color morph frequencies under two diet regimes 89 5-3. Body temperature of 4th instar nymphs after 8 minutes of illumination in relation to live weight .... 96 5-4. Heating curves of 4th instar nymphs 98

PAGE 9

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy COLOR POLYMORPHISM IN NYMPHS OF THE SOUTHERN GREEN STINK BUG, Nezara viridula (HEMIPTERA: PENTATOMIDAE ) By Carol Brown Johnson December 1984 Chairman: Thomas C. Emmel Major Department: Zoology Coloration of Nezara viridul a (Hemiptera: Pentatomidae) nymphs varies between and within instars. All nymphs transform from black to green in one or two molts. Adults are green. Polymorphism within the 3rd, 4th, and 5th instars can be explained by variation in the timing of color transformation. This study includes a biological comparison of 4th instar morphs and an evaluation of parental and environmental effects on nymph coloration. Black and green 4th instar nymphs did not differ significantly with respect to sex, dry weight, development time through the 3rd instar, or reproductive success. Black morph 4th instar duration was slightly longer. Color morph ratios varied among successive broods from single-pair matings. Last egg masses laid by senile females usually had higher

PAGE 10

proportions of black morphs than previous broods. Major shifts from green-dominated broods to black-dominated broods corresponded to lower survivorship through the 3rd instar but not to egg size or hatch success. Parental coloration during the 4th instar influenced the ratio of offspring color morphs, but did not restrict the range of variation. Differences in diet (green beans, peanuts, or both) affected adult fertility but did not have a clear effect on offspring coloration. The effects of nymph diet and temperature were tested by dividing egg masses and rearing halves on different diets or at different temperatures. At 23 C, nymphs reared on single-item diets (green beans or peanuts) had slower growth rates, higher mortality, and higher occurrence of melanism than nymphs fed both food items. A diet switch during the 2nd or 3rd instar affected coloration of morphs in the 4th instar. Lower rearing temperatures increased melanization among 3rd and 4th instar nymphs. Diet and temperature affected coloration and development time to different degrees. Under incandescent lighting, black morph 4th instar nymphs heated more rapidly and reached higher equilibrium body temperatures than green morphs. Flexible melanism in nymphs may be an adaptation to improve basking in cool seasons and crypsis in summer. Control of nymph color expression is probably hormonal. The maintenance of within-instar polymorphism may be related to predator deception.

PAGE 11

CHAPTER ONE INTRODUCTION Evolution and M ain ten a nce of C olor Polymorphisms E. B. Ford (1945, p. 73) defined polymorphism as "the occurrence together in the same habitat of two or more distinct forms of a species in such proportions that the rarest of them cannot be maintained by recurrent mutation." Ke recognized two categories of polymorphisms: transient and balanced. Transient polymorphism is found when a certain gene becomes advantageous and begins to spread through the population (Ford 1945). The classic example is industrial melanism among European Lepidoptera, where changing backgrounds have favored increased selection for melanic forms (Kettlewell 1973). Balanced polymorphisms can exist only when selective pressures against different forms are balanced against each other (Ford 1945). For example, the many color forms of some butterflies have been explained as multiple mimicry, where each form imitates a different distasteful model (e.g., Sheppard 1962). Balanced polymorphisms in some cryptic species, such as land snails (Jones et al. 1977) and spittlebugs (Owen and Wiegert 1962), have been explained by apostatic selection (Ayala and Campbell 1974). That is, predators search for the most common prey form and the rare, or apostate, form has a selective

PAGE 12

advantage. Hailman (1977) appropriately called this situation "deceptive polymorphism" and suggested that many examples of extreme intraspecif ic variability fit in this category. Endler (1978) proposed that polymorphisms could be selectively neutral if each color pattern morph resembles a different random sample of the environment. A similar argument has been made for polymorphic snails to account for higher morph diversity in more diverse habitats: habitat diversity allows a greater number of cryptic associations (Cain and Sheppard 1950, Rex 1972). An alternative hypothesis involves disruptive selection. In patchy environments, selection may act in different directions on different parts of the population. Gene flow between the subpopulations can then maintain a polymorphism (Jones et al. 1977). The strictest definition of polymorphism does not include seasonal variation or continuous variation that falls within a normal distribution (Ford 1945). However, the term "phase polymorphism" has been used to describe the different color forms of the solitary and migratory phases of migratory locusts (Gillet 1978, Sasakawa 1967). In addition, some authors now refer to "seasonal polymorphism" when two distinct forms occur at different times of the year (Ishii and Hidaka 1979). In such cases, different morphs are not determined by different genes, but by environmental cues affecting gene expression ( Wigglesworth 1959). Photoperiod, temperature, crowding, and humidity are most often cited as responsible for morph changes (Watt 1969, Hamilton 1973, Douglas and Grula 1978, Myers 1977).

PAGE 13

Color Pattern s in Imma ture_ Ins e c t s The variations in form and color that occur during insect metamorphosis led Wigglesworth (1959) to suggest that metamorphosis could be viewed as a special kind of polymorphism, "successive polymorphism." Indeed, this viewpoint may help to elucidate the selective forces that have led to dramatic transformations during larval development. As an example, cecropia moth larvae are jet black in the 1st instar, while succeeding instars are yellow with black tubercles, bluish green with red, blue, and yellow tubercles, and bright green with red and yellow tubercles (Williams 1961). Since there will always be temporal overlap in egg deposition, the population of cecropia larvae will be functionally polymorphic, varying both in size and coloration. What could be the adaptive value of a successive polymorphism? Our understanding of balanced polymorphisms can supply some hypotheses. Since larval stages suffer predation from visual predators (de Ruiter 1952, Heinrich 1979), cryptic and mimetic forms should be selected for. If larvae change their cryptic color pattern with every molt, an individual's chance of being in a predator's search image would be greatly reduced. Changes in size with normal larval development may require a color or pattern change in order to remain cryptic in the same habitat (Endler 1978). Or, behavior and habitat choice may change with development, requiring a concomitant change in coloration to remain cryptic. Hamilton (1973) suggested that many caterpillars are black in the first instars because the additional thermal gain helps them to grow and attain a larger size more quickly. As the body size increases, rapid growth becomes less important and cryptic color becomes more important

PAGE 14

Coloration of the Sou thern Green Stink Bug The southern green stink bug, Nezara viridu la (L.) (Hemiptera: Pentatomidae) is a very successful cosmopolitan pest of field crops, particularly legumes. Its life cycle and natural history have been recounted by numerous authors (Jones 1918, Drake 1920, Corpuz 1969, Singh 1973, Todd and Herzog 1980) and DeWitt and Godfrey (1972) give a comprehensive bibliography of the literature. In the Indo-Pacific region, several genetically distinct adult color forms coexist (Yukawa and Kiritani 1965, Singh 1973), but only the typical form, smaragdula is in the western hemisphere (Drake 1920, Yukawa and Kiritani 1965). This form is uniformly light green in color. Nezara has a relatively long adult life span, usually a month and more (Kiritani et al. 1963, Mitchell and Mau 1969, Corpuz 1969). Females lay large batches of eggs (60-120) at approximately one week intervals (Drake 1920, Corpuz 1969, Kiritani et al. 1963). After the reddish brown 1st instar nymphs hatch, they tend to cluster tightly on or near the egg mass. Second instar nymphs are black, with various white, red, and yellow spots. Second and 3rd instar nymphs also aggregate, especially before and during molting. Nymphs in the 3rd, 4th, and 5th instars are variable in color, from black to green. Third instar nymphs are usually black, most 5th instar nymphs are green, and the 4th instar is the most diverse in coloration. All nymphal instars and adults share the same habitat and food source. Nymphs fall victim to predatory hemipterans, ants, robber flies, parasitic wasps, spiders, and frogs (Stam 1978), and they are also acceptable food for the lizard

PAGE 15

5 Anolis carolin e nsi s Voight (personal observation). Adults are preyed upon by birds as well (Genung and Green 1974). Parasitoids are a major threat to eggs, large nymphs, and adults (Drake 1920, Stam 1978, Buschman and Whitcomb 1980). The coloration variability in Nezara therefore, falls both in the category of successive polymorphism (Wigglesworth 1959) and polymorphism in the sense defined by Ford (1945). Coloration changes as the insect matures, but variation between nymphs of the same age class also occurs. Since the species is subject to predation by visual predators, the color patterns are most likely the best compromise between protective adaptation and adaptation to opposing selective forces. Objective.s of This St udy The broad objective of this study of N^_ viridula nymph polymorphism is to gain sufficient understanding of nymph color determination to formulate an appropriate hypothesis of the adaptive significance of the polymorphism. The aspects that will be investigated include (1) Changes in color pattern that occur during the different stages of development. (2) Diversity of color pattern in each instar. (3) Differences between color morphs with respect to size, development rate, and reproductive success in the laboratory. (4) Consistency of color morph frequencies in sibling broods. (5) Relation of color morph of offspring to parental color morph history.

PAGE 16

6 (6) Maternal contribution (egg size, hatchabi li ty ) to nymph color morph (7) Effect of parental diet on the outcome of nymph color. (8) Effect of nymph diet on the outcome of nymph color. (9) Effect of rearing temperature on the outcome of nymph color. (10) Thermal advantages of black coloration in nymphs. I will conclude with a discussion of possible ecological functions of Nezara nymph polymorphism that would be consistent with this informat ion

PAGE 17

CHAPTER TWO DESCRIPTION AND BIOLOGICAL COMPARISON OF NYMPHAL MORPHS Introduction Nymphs of the southern green stink bug, Nezara viridula (Hemiptera: Pentatomidae ) undergo dramatic color changes as they mature. Early instars are mostly black, while late instars are mostly light green; adults are light green. In describing light and dark forms of the late instars, Jones (1918) added the brief comment that most 5th instar nymphs were of the light type and 4th instar nymphs had about equal numbers of both. Ke also mentioned the existence of intermediate forms. To the human eye, a population of Nezara nymphs presents a diverse array of size and color patterns. The diversity of color pattern in the 4th instar nymphs is as striking as polymorphisms described in Cepaea land snails (Jones et al. 1977), ladybird beetles (Houston and Hales 1980), and a variety of moths (Kettlewell 1973). Those polymorphisms, however, differ in significant ways from what appears in Nezara. Most polymorphisms that have been studied have involved adult phenotypes or characters which are constant throughout the entire life. In the case of Nezara nymphs, both size and color pattern can change within a few days or weeks. Yet, adult N. viridula in the United States are monomorphic When variable phenotypes occur in the adult stage, direct correlations can be made between phenotype and reproductive success.

PAGE 18

However, any survival advantage of a particular nymphal color morph would be separated in time from actual reproductive success. If a certain nymphal color morph has greater reproductive success as an adult, it does not necessarily follow that the coloration itself is advantageous. One must first show that color variation is unrelated to inherent variations in other biological features which might affect reproductive success. To that end, I present comparisons of 4th instar color morphs with respect to sex, size, development rate, and reproductive success under laboratory conditions. Prior to these comparisons, I provide descriptions of the changes in color pattern that occur during the different stages of development. Materials a nd Methods Rearing Nymph s I obtained Nezara v iridula nymphs from the following sources: (1) natural populations in gardens and fields in the Gainesville, Florida, area; and, (2) eggs laid in the laboratory by wild females which had mated in the field, females reared from field-captured nymphs and mated in the laboratory, and first generation adults from lab-reared nymphs. Females and mating pairs were kept in 100 mm plastic petri dishes. Females usually laid eggs on the paper toweling which lined the dishes. Each egg mass laid in the laboratory was divided into two groups of sibling nymphs. Each group was reared in a 100 mm petri dish. I formed these groups by breaking the egg masses in two or by sorting newly molted 2nd instars into two dishes. The groups could then be assigned to separate treatments.

PAGE 19

9 Both adults and nymphs were successfully maintained on diets of green beans ( Phaseolus vulgaris) and shelled raw peanuts (Arachis hypogaea ) purchased in local supermarkets. Some nymphs were reared on green beans alone or peanuts alone (with water provided by a piece of sponge or cheesecloth fastened to the lid), but growth, survival, and color patterns were always affected. (See Chapter 4 for further discussion.) I provided fresh green beans, peanuts, and water every other day or as needed. In the summer and fall 1981, nymphs and adults were kept in an air-conditioned laboratory (25-27C) under 15L:9D incandescent and fluorescent illumination and with some indirect sunlight from a window. In all later experiments, I kept nymphs in constant temperature chambers at 23 or 28 C and with 12L:12D fluorescent lighting. Color Morph Des criptj.on_and_ Categorization In my preliminary observations of Nezara development, I looked at pattern, hue, and shade differences among individuals of each instar. I reared some nymphs in isolation in order to detect day-to-day changes. I kept others in small groups of like age and color to observe the diversity of changes that occurred with each molt. I reared entire egg masses or halves of egg masses together in a single dish to see the diversity among siblings. Because the greatest diversity of color and pattern occurred in the 4th instar, I concentrated my study on this life stage. I established color morph categories by comparing many sequences of 4th instar nymphs arranged from lightest (green) to darkest (black). A set of sketches helped me to be consistent in categorizing nymphs. The extent of

PAGE 20

10 cuticular melanization was verified by removing the dorsal cuticles of representative types, cleaning them of all epidermal tissue, and affixing them to glass slides for examination. Descriptions of the color patterns of the 4th instar and other stages follow in the results section Relation of Co lor_Morph_ t o_ Sex To determine the relationship of 4th instar color morph and sex, I reared 4th instar nymphs of known color morphs to the adult stage. Since there is no suggestion in the literature that Nezara sex ratios at hatching are different from 1:1, I hypothesized that green and black morphs from the same egg mass had equal chances of being male or female. I tested this by rearing all 53 polymorphic nymphs from a single egg mass of a wild female and comparing sex and color morph in a chi-square contingency table. As another verification, I randomly selected green and black nymphs reared from another, larger egg mass and tested the resultant sex ratios of each color morph with a binomial test (Siegel 1956). I also compared the sex ratios of small field collections of 4th instar nymphs in early fall (September October) 1981 and mid-summer (July) 1983. Color Morph and Nymph Size I tested for differences in 4th instar nymph size in relation to color morph by comparing dry weights of siblings of the same age. I selected 22 groups of polymorphic nymphs which had been reared on green beans and peanuts at 23 C in August September 1983. Each group consisted of nymphs from half of an egg mass reared together in the same

PAGE 21

11 petri dish. Each egg mass was from a different set of parents. Nymphs in each group had been preserved together in a vial of Hood's solution (5 ml glycerol to 95 ml 85% ethanol). Therefore, within a sibling group, specimens received the same treatment until I removed them for drying. I used a random number generator to select a subsample of five green and five black nymphs out of all green and black nymphs in the vial. Nymphs were dried in individual aluminum cups at 60C until no further weight loss was measured. I tested the data with a two-way analysis of variance, in order to distinguish variation caused by the source of nymphs (i.e., egg mass) from variation attributable to coloration. Color Morph and Development Rat e Since insect development rate can be influenced by many factors, I used a matched pair design to test for differences associated with color morph. I made observations of sibling groups (each from half of an egg mass) reared in petri dishes on diets of green. beans and peanuts or peanuts and water at 23 C. As soon as the nymphs in each dish began to molt from the 3rd to 4th instar, I made daily records of the number and types of each morph in the dish. The number of 4th instar nymphs which had molted within the previous 24 h was found by subtraction of the consecutive daily totals. All nymphs molted within a twoto four-day period. I then compared the proportion of green, intermediate, and black morphs in earlier and later observations, using a sign test (Siegel 1956). If significant differences in development rate exist among morphs, then one morph should consistently appear more frequently among the earlier molting nymphs than among the later molting nymphs.

PAGE 22

12 Development rate during the 4th instar was compared by continuing to rear selected single nymphs until the molt to the 5th instar. In 1982, I selected one set of 19 green and 7 black nymphs drawn from a single sibling group. I continued rearing them on a peanuts and water diet at 28 C, recording molts every 24 h. In 1983, I reared a second set of 23 green and 22 black nymphs on green beans and peanuts at 23C. They were drawn from four different sibling groups of newly-molted 4th instars on the same day. I made observations of molting every 12-14 h. All nymphs were reared to adult stage to determine sex. I used a binomial test to determine whether sex ratio within a color morph sample differed from 1:1. Differences in the average duration of the 4th instar for green and black morphs were tested with a one-way analysis of variance Color Morph and R eproductive Sue cess Nymphs of known color type were collected in the field in September October 1981, reared to adults, and separated into mating pairs in 100 mm plastic petri dishes. Most of the matings matched adults which had similar color types. Additional mating pairs were formed from siblings which were reared from eggs in the laboratory. The adults in each of these pairs had identical patterns in the 4th instar. I collected and recorded all egg masses. By allowing eggs to hatch, I distinguished fertilized, viable eggs from unfertilized or otherwise inviable eggs. One measure of reproductive success is the production of fertile, viable eggs. If no differences exist between color morphs, the relative proportion of females laying fertile eggs should be the same. I tested this hypothesis with the chi-square statistic.

PAGE 23

13 I also compared the size of egg masses of parents which were green or black in the 4th instar. To reduce the possible effects of genetic variability, I compared egg masses of sibling pairs derived from a single egg mass. I found the average number of eggs per mass for each pair and tested the differences with the Mann-Whitney U-test (Siegel 1956). Results Color Morph De scriptions First instars were reddish brown in hue, while 2nd instars were black. Appearance changed as the cuticle stretched around the growing nymph. A pattern of red and white abdominal spots common to all stages (see Jones 1918) was very inconspicuous in recently molted nymphs. These spots seemed to grow in size as the abdomen distended. Thus, 2nd instars appeared solid black on the first day of the stage, but they became conspicuously spotted on subsequent days. However, all 1st and 2nd instar nymphs of the same age looked the same. In contrast, there was great variation among nymphs of the same age during the 3rd, 4th, and 5th instars. The darkest individuals had a coloration pattern identical to the 2nd instar: black, except for the characteristic white and red abdominal spots and two golden-yellow spots on each lateral margin of the thorax. The lightest individuals were almost completely green. In all stages, blackest nymphs left black exuviae upon molting; the exuviae of greenest nymphs were transparent, with only a faint tan hue. Third instar nymphs were usually black; 5th

PAGE 24

14 instar nymphs were usually green. The 4th instar was the stage most variable in color and pattern. Fourth instar nymphs could be arranged in a continuous sequence from lightest to darkest according to the degree of melanization of the thorax. When the thorax was least melanized, the general color of the head, thorax, abdomen, and legs was green. As melanization increased, the head became yellowish or black, the unmelamzed part of the thorax became more yellowish in hue, the abdomen became dark green to brownish black, and the legs became black. The underside of the abdomen also varied from light green to pink to red. After ecdysis and sclerot izat ion was completed, the only aspect of a nymph's coloration that changed during the 4th instar was the shade of its abdomen. As the abdomen distended with feeding, dark green and greenish brown abdomens got distinctly lighter. On the other hand, the thoracic pattern could be confidently determined soon after the molt, did not change during the instar, and did not fade in preservatives. I therefore decided to categorize the 4th instar nymphs on the basis of thoracic patterns of melanization. I established eight categories, from lightest (I) to darkest (VIII) (Figure 2-1). All of the nymphs that I observed individually changed from the black coloration characteristic of the 2nd instar to the green hue characteristic of the adult in one or two steps. In some cases, the black nymphs (2nd, 3rd, or 4th instars) changed to green nymphs in one molt, and they were green in all subsequent molts. In other cases, black nymphs molted to intermediate forms, which always molted to green forms in the following molt. I never observed a green or intermediate nymph of any age molt to a darker color.

PAGE 25

15 a. j_i E >i O c •H (4J 0) u u o ai 4= a. o -a C 0) to -a .-. qj cu E 2 :n en 00 M men 3 5 > to cd i

PAGE 26

16 Relation of Col or Morph Co S e x Both males and females were reared from all 4th instar color morphs. Within a brood, sex ratio did not differ significantly among the different color morphs (Table 2-1), and sex ratios of black and green morphs did not differ from 1:1 (Table 2-2). The samples of 4th instar nymphs collected in the field did not include enough darker morphs to test sex ratio in a contingency table (Table 2-3). Both sexes were represented in each color category in the fall sample. No black females and only two black males were collected in the summer sample. Color Morph and Nymph Size The average dry weights of 110 green and 110 black 4th instar siblings from 22 egg masses were 11.33 mg and 10.60 mg respectively (pooled standard deviation = 3.28). In 8 out of 22 sibling groups, black morphs were larger than green morphs. Analysis of variance of the dry weights showed that the difference between morphs is not significant (Table 2-4). The different sibling groups had a significant effect on the variance of the data. Color Morph and Development Rate No particular color morph could be significantly associated with faster development from hatch to the 4th instar (Table 2-5). However, in the majority of sibling groups observed, there was a higher frequency of green morphs earlier in the molting period than later.

PAGE 27

17 Table 2-1. Sex determination of 4th instar nymphs with respect to cole morph. All reared from a single egg mass from wild parents. Color Morph Female Male Green (I-II) n 4 Intermediate (III-V) 6 8 Black (VI-VIII) 11 13 X 2 = 3.56 df = 2 p = 0.169 Table 2-2. Sex determination of green and black 4th instar color morphs sampled randomly from a single egg mass. 4th Instar Total No. in Sample Color Morph Egg Mass Size Female Male p* Green (I-II) 27 19 11 8 0.65 Black (VIII) 25 10 5 5 0.62 '"Binomial test (2-tail)

PAGE 28

18 Table 2-3. Sex of 4th instar nymphs collected from field populations. Sept. /Oct. 1981 July 1983 4th Instar Color Morph Female Male Female Male Green (I-II) Intermediate (III-V) Black (VI-VIII) 7

PAGE 29

19 Table 2-5. Differences in morph frequencies between 4th instar nymphs which appeared earlier and later in the molting period. Values are the number of sibling groups in which differences in frequency occurred. Relative Morph Frequency Total No. Color Earlier% Later% > Later% = Diet Groups Morphs > Later% Earlier% Earlier% Green Beans Peanuts Green

PAGE 30

20 All color morphs were represented among the first and last nymphs to molt in the sibling groups. On the whole, the duration of the 4th instar was slightly longer for black morphs than for their green counterparts (Table 2-6). Color Morp h and Repro duct ive Succesjs Nymphs of all color morphs were equally likely to successfully reproduce as adults (Table 2-7). There were also no significant differences in the fecundity of sibling females which had been different morphs in the 4th instar. In 1 1 sibling mating pairs reared from one egg mass, the 7 green (I-II) pairs had an average of 98.5 eggs per mass per mating pair, compared to 104.0 for the 4 black (VIII) pairs (U = 13, p = 0.928, 2-tail). The largest egg masses of green pairs ranged from 85 to 129 eggs; egg masses of black pairs ranged from 117 to 123. The average hatch success of eggs laid by green mating pairs was 81.4% (56.0 98.4%), compared to 86.9% (68.3 96.8%) for eggs of black pairs. Discussion Polymorphism in Nezara The basic color pattern of Nezara viri dula nymphs changes only at the time of the molt. The transition from black to green occurs in one or two stages between the 2nd instar and the adult stage. The polymorphism among 3rd, 4th, and 5th instar nymphs can be explained by variability in the timing of the color transition. Under normal

PAGE 31

21 Table 2-6. Duration of 4th instar in relation to color morph. Differences tested with one-way analysis of variance. n = number of nymphs Rearing Morphs No. of Days Conditions Compared n (Mean + S.E.) Peanuts, Water* Green (I-II) 19 6.6 + 0.31 15.41 <0.001 28C Black (V-VI) 7 10.0 + 1.15 Green Beans ,** Males: Peanuts, Green (I-II) 7 5.9+0.13 1.14 >0.05 23 C Black (V-VIII) 13 6.4 + 0.33 Females : Green (I-II) 16 6.8 + 0.33 1.02 >0.05 Black (VI-VII) 9 7.4 + 0.41 '•^Observations made every 24 h. All nymphs drawn from one egg mass. Sex ratio in each color category not significantly different from 1:1, ""-'•'Observations made every 12-14 h. Nymphs drawn from 4 egg masses. Sexes analyzed separately since sex ratio for green category is different from 1:1.

PAGE 32

22 Table 2-7. Comparison of fertility of adult females which were different color morphs in the 4th instar. Fie Id -Captured" Lab-Reared** 4th Instar No. No. Laying No. No. Laying Color Morph Mated Fertile Eggs Mated Fertile Eggs Green (I-Il) 15 Intermediate (III-V)*** 2 Black (VI-VIII) 5

PAGE 33

23 conditions, most nymphs change color at the 3rd or 4th nymphal molt, making the 4th instar the most diverse stage. An individual may be transformed in one molt from a black bug to a green bug, or it may go through one instar with intermediate coloration. Once the green coloration is attained, subsequent stages are also green. Kobayashi (1959) provides descriptions and drawings of 3rd, 4th, and 5th instar polymorphism in N. viridula in Japan. The range of variation in 4th and 5th instar color patterns is the same as I have observed, but my intermediate classifications (III, IV, and V in Figure 2-1) are not represented. His drawings of 3rd instar nymphs show only dark types (VI-VIII), whereas I have collected green (II) 3rd instar nymphs in a Gainesville garden. Other published descriptions of nymphal stages do not mention 3rd instar polymorphism; they describe a monomorphic black pattern identical to the 2nd instar (Jones 1918, Drake 1920, Singh 1973). Singh (1973) did not find black morphs of the 5th instar in Jabalpur, India, and the darkest 4th instar nymph he described had green, unmelanized areas on the thorax. Even though several adult color forms (none of which involve melanism) have been described in the Indo-Pacific region (Yukawa and Kiritani 1965), local researchers have not made distinctions in the appearance of their nymphs (Kobayashi 1959, Singh 1973). At least one other Japanese Nezara species, N. antennata Scott, has polymorphic nymphs like N. viridula (Kobayashi 1959, Kariya 1961). Light and dark color morphs have also been described for nymphs of the closely related Acrosternu m hjjar_e (Say) in the United States (Drake 1920).

PAGE 34

24 Similar Polymor p h i c_ S y s t ems Nymphal polymorphism has been described in another pentatomid, Perillus bioculatus Fabricius, which is also polymorphic as an adult (Knight 1924). Adult Perillus have complex striped patterns of melanization that vary in extent on the pronotum, scutellum, and corium. Unmelanized areas vary in hue from white to yellow to red; red accompanies increasing melanization. The darkest nymphal morphs are red with black thoraxes and wingpads while the lightest are white with some unmelanized areas on the thorax. Under normal conditions, nymph color forms coincide closely with adult color form. Knight concluded from his rearing experiments that temperature and humidity were the major factors accounting for the polymorphism. A diversity of dorsal black cuticular pigmentation patterns appears among nymphs of the lime aphid Eucal lipteru s t^il^ae (L.) (Kidd 1979). First generation nymphs are all yellow and unmelanized, but later nymphs develop black markings on the head, thorax, and abdomen in 2nd instar. Unlike Nezara patterns, these remain constant through remaining nymphal stages. Adult coloration is not variable. Kidd (1979) associated the appearance and greater frequency of melanics with crowding and leaf maturity Another larval polymorphism very similar to the Nezara system is found in the saturniid moth Saturni a ( Eudi a ) payo_nia L. First instars are black, while the last instars are usually solid pale green. In the 4th and 5th instar, larvae develop black patterns on each segment which can vary from thin black markings around the dorsal tubercles to a solid transverse band of black (Long 1953). Hintze-Podufal (1974) described the morphs extensively and, using larval exuviae, showed that the black

PAGE 35

25 markings were incorporated in the cuticle. The frequency of occurrence of melanic forms in the laboratory was influenced by crowding, light, humidity, temperature, and the freshness of the food (Long 1953, Hintze-Podufal 1977). Differences Between Morp hs Sex. There is no evidence that color morph expression in the 4th instar is related to the sex of the individual (Tables 2-1, 2-2). Since males are heterogametic (Kiritani et al. 1962), any genetic control over the polymorphism is probably autosomal. More field sampling (Table 2-3) is needed to determine whether there are differences in the sex ratio of morphs in the field. Since males develop slightly faster than females and are smaller as adults (Kiritani 1964), adaptive value of a color morph may differ between the sexes. Size and development rate A distinction should be made between growth rate (rate of increase in mass or size) and development rate (rate of maturation). Reaching a larger size faster may confer a number of advantages, including escape from predators (Hamilton 1973) and improved thermal stability (Casey 1981, May 1976). Larger male Nezara live longer, have better mating success and are more fertile (McLain 1981). Female fecundity and longevity increases with body weight (Kiritani and Kimura 1965). On the other hand, reaching reproductive maturity at an earlier age, regardless of size, can increase an individual's relative genetic contribution to succeeding generations (Hamilton 1973).

PAGE 36

26 Since there are no significant differences in dry weights of green and black 4th instar morphs reared in the laboratory (Table 2-4), color morph is probably not inherently linked with growth rates. However, green morphs appear to develop at a slightly faster rate. In most of the broods tested, there were proportionally more green morphs among the first nymphs to molt to the 4th instar than among the later-molting nymphs (Table 2-5). Black morphs average significantly longer 4th instar periods when equal numbers of each sex were reared on peanuts and water at 28 C (Table 2-6). At 23 and with the green beans and peanuts diet, black morphs averaged about half a day longer than green morphs of the same sex. Although the differences were not significant, the trends in these samples justify further experimentation. Inherent differences in black and green morph development rates may be magnified at lower temperatures. Since sex influences development rate, development of black and green morphs of both sexes should be monitored to the adult stage Reproductive success Differences in 4th instar coloration were not associated with differences in general reproductive success in the laboratory. Equal proportions of fully reproductive adult females were reared from representatives of all 4th instar morphs (Table 2-7). Sibling matings of like types (green-green and black-black) demonstrated equal fertility and fecundity of males and females derived from both morphs Conclusions Fourth instar color pattern is not clearly linked to other genetically or maternally inherited characters that could confer

PAGE 37

27 selective advantage. There is some evidence, however, that development rate may be slightly faster in green morphs than in black morphs. Since the differences overall appear minor, I must conclude that nymphal color polymorphism (1) is linked to some other adaptive feature that I have not measured, (2) is selectively neutral (ecologically irrelevant), or (3) has its own ecological function, relevant only in the natural habitat.

PAGE 38

CHAPTER THREE PARENTAL EFFECTS ON NYMPHAL COLOR MORPHS Introduction The ecological function of any color polymorphism must be examined in the context of the control of color pattern expression. For example, mimetic polymorphisms can only be maintained when there is strict genetic inheritance of a particular pattern (Ford 1945). Apostatic selection can maintain a cryptic polymorphism only if the rare morph is able to transmit its deviant alleles to its offspring (Ayala and Campbell 1974). However, if color patterns are determined exclusively by environmental factors, the specific color pattern of an individual will be independent of the color pattern of its parent. This situation is found when environment changes radically from one generation to the next and the adaptive value of a parental color pattern changes with it (Rowell 1971). For example, migratory locusts transform in one generation from cryptic, solitary forms to densely aggregating, conspicuously colored forms (Goodwin 1949). The late instar nymphs of the southern green stink bug, Nezara viridula are polymorphic (Jones 1918). The most diverse stage, the 4th instar, typically has forms ranging from black to light green. The darker color morphs have greater cuticular melanization in addition to darker subcuticular pigmentation. The blackest forms are identical in color pattern to the typically monomorphic 2nd and 3rd instars. The 28

PAGE 39

29 greenest forms are identical to the common green morph of the 5th instar and are the same hue as the green adult. (See Chapter 2 for complete descriptions ) If the control of color pattern in Nezara nymphs is purely genetic, then the ratio of morphs in the broods of any mating should remain the same regardless of environment. Black-black and green-green matings should yield offspring morph ratios at opposite extremes. If the variation in color pattern is a purely environmental effect, then all broods in the same environment should have the same variety and frequency of color patterns, regardless of parentage. Maternal effects would be indicated if the mother's environmental history or present health corresponds with changes in the color patterns of her offspring. In this paper, I present color morph ratios for mutiple broods from single-pair matings of Nezara viridula and the results of crossing like morphs. In addition, I examine possible maternal effects by comparing the color morph ratios of broods from parents reared on different diets and by comparing egg sizes where black color morph frequency changes between successive broods of the same parents. The relative importance of genetic, environmental, and maternal effects will be discussed. Materials and Methods General Approach Single-pair matings of virgin adult Nezara viridula provided series of egg masses which had common parentage. After rearing broods under identical conditions, I compared the frequencies of 4th instar color

PAGE 40

30 morphs in broods from the same parents and from different parents. In addition, I made comparisons between offspring of (1) pairs which had been different color morphs in the 4th instar and (2) pairs which had been reared and maintained on different diets. I also looked for correlations between several measurable characteristics of the egg mass and the frequency of black color morphs that resulted. Source of Specimens The eggs and nymphs used in this study came from parental stocks of Nezara viridula formed in June 1981, September-October 1981, and July-August 1983. Data sets originating from these stocks will be referred to as "summer 1981," "fall 1981," and "1983," respectively. I made allfield collections of Nezara in gardens and soybean fields in Gainesville, Florida. I formed the summer 1981 parental stock from the offspring of a pair of adults which had been collected the previous October and had overwintered in the laboratory. These siblings originated from the sameegg mass and were paired together according to their coloration during the 4th instar. The fall 1981 stock came from (1) 3rd, 4th, and 5th instar nymphs collected in the field in September and (2) nymphs reared from eggs laid by two gravid females captured in early September. Matings from the latter source were sibling crosses. Wherever possible, I mated adults which had had the same coloration in the 4th instar. All of the 1983 mating pairs were made up of adults reared from 2nd, 3rd, 4th, and 5th instar nymphs collected in mid-July 1983. All of the 4th instars were green morphs (I-II, Figure 2-1). With the

PAGE 41

31 exception of a special group of sibling matings, all matings were between bugs that had been collected on different days and on different plants. I formed a separate group of sibling mating pairs from a single dense aggregation of 2nd instars collected in the field. On the basis of my prior observations of nymph development and behavior, I concluded that these nymphs had hatched from the same egg mass. This group of bugs was subdivided and reared on different diets, as described below. Maintenance of Parental Stock Single mating pairs were kept in 100 mm plastic petri dishes lined with paper toweling. All mating pairs remained in a laboratory room (25-27 C) under 15L:9D incandescent and fluorescent lighting. Some indirect sunlight came through a window. Unless specified otherwise, all adults received a diet of fresh green beans ( Phaseolus vulgaris ) and shelled raw peanuts ( Arachis hypogaea ) purchased in local supermarkets. I used garden-grown green beans for a few weeks in June and September 1981. I provided fresh green beans every other day in 1981 and every day in 1983. I replaced peanuts when they became soft or moldy. I cleaned dishes and replaced paper toweling as needed. Each day I collected egg masses, which were usually laid directly on the paper toweling. If egg masses were laid elsewhere, I used white glue to affix them to a piece of toweling. Nymph Rearings In 1981, I kept all egg masses and nymphs with the parental stocks in the laboratory at 25-27 C, 15L:9D. In 1983, developing egg masses were moved to a 23 C constant temperature chamber on the 4th day of

PAGE 42

32 incubation. Nymphs hatched and developed in the chamber, which had 12L:12D fluorescent illumination. The nymphs from each egg mass were reared in two 100 mm plastic petri dishes. The two groups were formed by (1) allowing the entire egg mass to hatch and later dividing newly molted 2nd instars (1981 only) or (2) breaking the egg mass in two before hatch (1981 and 1983). The different techniques appeared to have no effect on survival or color morphs. In 1981, half of the nymphs of each egg mass were reared on green beans and peanuts while the other half received peanuts and water from a moistened piece of cheesecloth. In 1983, both halves of each brood of three mating pairs were reared on peanuts and water. For 12 other mating pairs, only one half of each of the first three broods received green beans and peanuts; the other half received treatments in conjunction with other experiments. For the remaining broods, both halves received green beans and peanuts. I provided fresh food and clean dishes every other day in 1981 and each day in 1983. Determination of Color Morph Frequencies When the nymphs in a petri dish reached the 4th instar, I assigned each nymph to one of eight color morph categories based upon the degree of melanization of the thorax (Chapter 2, Figure 2-1). For analysis, I combined the two lightest categories (green), the three intermediate categories (intermediate), and the three darkest categories (black). I calculated green, intermediate, and black color morph frequencies by dividing the number of nymphs in each category by the total number of nymphs reared in the dish. When both halves of a brood received identical diets, I combined the data from both dishes.

PAGE 43

33 Variation in Color Morph F r jegu e nc i e s_ Anipn_g_ Brood s In summer 1981 and in 1983, I reared continuous series of egg masses from different mating pairs. All series included the first egg masses produced by the mating pair. In 1983 I continued rearings of fertile egg masses until the female died. I made graphical comparisons of the frequencies of black, intermediate, and green 4th instar color morphs resulting from each egg mass or half egg mass for each mating pair Egg Mass Character istics and Resulting Color_ Morphs When dramatic changes in color morph frequencies occur between two broods from the same mating pair, a nongenetic factor must be responsible. Such changes occurred in my rearings of successive egg masses in 1983, despite my efforts to maintain constant rearing conditions. If the changes were associated with changes in the mother's reproductive state, we might expect to see changes in egg size, number of eggs per egg mass, and egg viability. I selected seven mating pairs for which the frequency of black morphs in their broods increased 33 to 78 percentage points from one brood to the next. For each egg mass, I counted the eggs and calculated the percent successfully hatched. I selected a random sample of 35 eggs and measured the diameter of the operculum using an ocular micrometer and a dissecting microscope. I ran a two-way ANOVA (Sokal and Rohlf 1973) of egg diameter on the two factors: (1) low vs. high black morph frequency and (2) the parents of the egg mass.

PAGE 44

34 Effect of Parenta ge on Color Morph Frequencies Variation in color morph frequency can occur both among broods from the same parents and between broods of different parents. If parentage has an effect on offspring color morphs one would expect that the variation among sibling broods would be less than the variation among all broods in the population. I tested this hypothesis with a Kruskal-Wallis one-way analysis of variance by ranks (Siegel 1956) on green morph frequencies in broods from the summer 1981 and fall 1981 data sets. I restricted analysis to mating pairs for which I had data from three or more broods. If 4th instar coloration is heritable, one would expect a strong correlation between the parents' 4th instar color morphs and the frequency of those morphs among their offspring. For example, parents which were completely black in the 4th instar ought to have a higher proportion of black morph offspring than parents which were green in the 4th instar. I tested this idea using the summer 1981 and fall 1981 mating pairs which had been either black or green in the 4th instar. Since I had reared only one, two, or three egg masses of many of the mating pairs, I used only the first three egg masses of the other mating pairs in my analysis. For each mating pair, I found the average black and green color morph frequencies. I then tested the values obtained for green morph parents against those for black morph parents using a Mann-Whitney U-Test (Siegel 1956). Effect of Parental Diet on Color Morph Frequencies Because embryos are packaged in an egg shell with a supply of nutrients and hormones provided by the mother (de Wilde and de Loof

PAGE 45

35 1973a), the physiological state of the mother can play a role in the success of the offspring. Since nymph color morphs are affected by their own diet (Chapter 4), parental diet may have an effect as well. I tested this hypothesis in 1983 by rearing three sets of sibling 2nd instars on three diets: (1) green beans and peanuts, (2) peanuts and water, and (3) green beans alone. I matched pairs of virgin males and females which had been reared on the same diet and continued the diets until the death of the female. I divided each of the first two or three viable egg masses from these matings and reared half of the nymphs on green beans and peanuts and half on peanuts and water at 23 C. I compared color niorph frequencies, hatch success, and egg mass size. Results Variation in Color Morph Frequencies Among Broods Summer 1981 data se t. Frequencies of green, intermediate, and black morphs in the 4th instar varied between successive broods from the same parents (Figure 3-1). For most mating pairs (B, C, E, F), the first three half-broods reared on the green beans and peanuts diet were more similar to each other than to the last two or three broods. In those cases, the proportion of green nymphs declined in the later egg masses. The ratios of color morphs of half-broods reared on peanuts and water also varied. Again, in all cases the later broods had higher proportions of black morphs and lower proportions of green morphs than earlier broods. In addition to the variation between broods from the same parents, egg masses laid on the same day by different mating pairs often yielded

PAGE 46

Figure 3-1. Fourth instar color morph frequencies of successive broods (Summer 1981 data set). Half of each brood was reared on green beans (G) and peanuts (P), and half on peanuts and water (W) at 25-27 C. The egg masses of 6 different mating pairs are represented (A-F) All parents are siblings from one egg mass. Number of 4th instars indicated above bars.

PAGE 47

37 DATE EGGS LAID

PAGE 48

38 100806040206/16 6/20 6/24 7/1 7/7 6/16 6/20 6/24 7/1 7/7 G, P P, W Figure 31--Ccnt inued DATE EGGS LAID

PAGE 49

39 very different ratios of 4th instar morphs. For example, the June 16 brood of mating pair B had about 25% black morphs while broods of mating pairs C, E, and F had about 5% or fewer blacks on the diet of green beans and peanuts. 1983 data set. The color morphs of broods reared on green beans and peanuts in 1983 (Figure 3-2) varied less than the 1981 broods. For many mating pairs, the color morph frequencies remained consistent (many greens, few intermediates and blacks) for six or more successive broods (Figure 3-2 A-F, H, I). Sometimes the first brood (Figure 3-2 A), the second brood (Figure 3-2 D, H) or both (Figure 3-2 G) had distinctly higher proportions of darker morphs than the other broods preceding or following. But in all cases except mating pair A, the last broods produced by the mating pairs had higher proportions of black morphs and lower proportions of green morphs than earlier broods. The results of rearing broods on peanuts and water (Figure 3-3) were very different. Two mating pairs (Figure 3-3 A, B) had five successive broods in which color morph frequencies were identical, with black morphs dominating. But in several subsequent broods, green and intermediate morphs dominated. The broods of the third mating pair (Figure 3-3 C) had lower proportions of black morphs than the broods of the other mating pairs. The second and third broods showed a shift to dominating green morphs. Egg Mass Ch aracteristics and Resulting Color Morphs There was no consistent trend in either egg mass size or hatch success when consecutive broods with low and high black frequencies occurred (Table 3-1). Egg mass size and hatch success decreased in

PAGE 50

Figure 3-2. Fourth instar color morph frequencies of successive broods reared on green beans and peanuts at 23C (1983 data set). Each set of bar graphs incudes rearings of all fertile egg masses laid during the lifetime of one female mated to one male. The broods of 12 different mating pairs (A-L) are represented. Number of 4th instars indicated above bars. = only one half of egg mass represented in this rearing.

PAGE 51

41 1008060402045* I green ] intermediate black 84 A 51* s§ 100> 80z !j 60o LL 7/28 8/3 8/8 8/13 8/18 8/23 8/28 9/2 9/8 49" 54* I CC O 5 20101 108 93 B 8/2 8/6 8/11 8/16 8/25 9/3 9/11 1008060402061* 57* '9 8/14 8/19 8/24 8/29 9/4 9/9 9/14 9/19 9/30 DATE EGGS LAID

PAGE 52

42 8/13 8/18 8/23 8/29 9/4 9/10 9/16 9/22 9/28 10/4 8/13 8/17 8/22 8/28 9/3 9/8 9/13 10CH 8060402054 108 3 88 96 Urn 3
PAGE 53

43 100 /5 8/L4 8/22 a/30 9/6 9/13 9/20 9/28 10/8 100 8/9 8/13 8/18 8/23 8/28 9/3 9/10 9/15 9/20 100n 80 60 402053' 8/15 8/20 8/25 8/31 9/6 9/13 9/18 9/24 10/1 las DATE EGGS LAID Figure 3 -2-Cont inued

PAGE 54

44 lOCh 8,6 8/14 8/22 8/30 9/6 9/13 9/20 9/26 10/4 10/11 10/18 /14 8/19 8/24 9/1 9/6 9/12 9/19 9/25 10/2 10/9 10/14 100n /14 8/20 8/24 8/30 9/5 9/11 9/16 9/22 9/28 DATE EGGS LAID Figure 3-2--Continued

PAGE 55

Figure 3-3. Fourth instar color morph frequencies of successive broods reared on peanuts and water at 23C (1983 data set). Each set of bar graphs includes rearings of all fertile egg masses laid during the lifetime of one female mated to one male. The broods of 3 different mating pairs (A-C) are represented. Number of 4th instars indicated above bars. Two broods reared at 21C indicated by*.

PAGE 56

46 7/25* 7/31 8/5 8/U 8/18 8/24 8/31 9/6 9/12 9/18 9/24 8/1 8/9 8/17 8/23 9/2 9/11 9/20 9/28 DATE EGGS LAID

PAGE 57

47 co afi u i 00 01 i oo a —i c-i CN CN O O O O o o + 1 + 1 CN ON in m o o o o + 1 + 1 o o + 1 + 1 r m on oo in >o oo oo LA ITl in l/^l o o o +1+1 +1+1 ci r^ in cn On co
PAGE 58

48 only four out of seven comparisons. However, there was significantly lower survivorship in the broods with the higher proportion of black morphs (Wilcoxon T = 1, N = 7, p < 0.025, 1-tail). Egg size varied significantly with respect to parentage, but there was no effect attributable to the differences in color morphs (Table 3-2). Effect of Par en tage on Color Morph Frequenci e s Parentage had an influence on offspring 4th instar color morph frequencies (Table 3-3, 3-4). That is, broods from the same parents were more similar in coloration to each other than to broods from the general population. Differences between broods of sibling parents of the summer 1981 data set (Table 3-3) were not as strong (0.05 < p < 0.10) as those found in -the fall 1981 data set (Table 3-4, p < 0.01), where most parents were unrelated. In addition, parental influence was significantly related to the 4th instar coloration history of the parents (Table 3-5). Parents which had been green during the 4th instar had broods with higher green morph frequencies and lower black morph frequencies than parents which had been black in the 4th instar. However, one pair of black morph parents had two egg masses which yielded more than 90% green morphs and no black morphs. One egg mass of one green morph parent resulted in 82% black morphs and only 6% green. Effect of Pa rental Diej:_ on_ Color Morph F requenci e s Five out of nine mating pairs reared on peanuts and water failed to produce viable eggs, and only one pair produced norma 1 -looking egg masses (Table 3-6). In contrast, all mating pairs reared on the other

PAGE 59

49 Table 3-2. ANOVA of egg diameter (mm) on two factors: the resulting frequency of black 4th instar morphs (low or high) and the parents of the eeg mass. Source of Variation Sum of Squares df Mean Square F Frequency of Black Morphs 0.000817 1 0.000817 3.69 >0.05 Parentage 0.106533 6 0.017756 80.25 <0.01 Interaction 0.003345 6 0.000558 2.52 <0.05 Error 0.105308 476 0.000221

PAGE 60

50 Table 3-3. Variation in frequency (%) of green morphs in the 4th instar, with respect to parentage. Summer 1981 data set. Frequencies are listed in the order egg masses were laid, but some broods are missing from the sequences. Nymphs reared on green beans and peanuts. The effect of parentage is strong but not significant (Kruskal-Wal lis H = 12.8, df = 7, 0.5 < p < 0.10). Fourth instar color morph of parents (male, female) are indicated by G (= green), I (= intermediate), and B (= black). Parental Parents Morphs % Green Morphs in Each Brood Mean Tl-1 (F)* G,G 88 92 100 45 66 T2-1 (C)* G,G 93 86 95 25 50 T5-1 B,B 26 64 94 T7-1 (A)* B,B 79 52 38 51 58 T5-3 B,B 50 40 74 T4-2 (E)* 1,1 81 52 56 27 39 T7-2 (B)* B,B 21 71 57 71 31 23 T7-3 (D)* B,B 12 54 23 11 9 78

PAGE 61

51 Table 3-4. Variation in frequency (%) of green morphs in the 4th instar, with respect to parentage. Fall 1981 data set. Frequencies are listed in the order egg masses were laid, but some broods are missing from the sequences. Nymphs reared on green beans and peanuts. The effect of parentage is significant (Kruskal-Wallis H = 33.9, df = 13, p < 0.01). Fourth instar color morphs of parents (female, male) indicated by G (= green), I (= intermediate), and B (= black). Parental Parents Morphs % Green Morphs in Each Brood Mean X7

PAGE 62

52 Table 3-5. Influence of parental color morph in the 4th instar on the color morph frequencies of their offspring. Nymphs were reared on green beans and peanuts. Values given are averages of the first egg masses produced by each mating pair. The number of egg masses used to determine each value is given in parentheses. Nymphs reared on green beans and peanuts. U = Mann-Whitney statistic. Average Morph Frequency Among Offspring Green Morph Black Morph Green Parents Black Parents Green Parents Black Parents SUMMER 1981 97.6 (2)

PAGE 63

53 Table 3-6. Comparison of egg production of parents which had been reared and maintained on different diets. Parents were reared from a single aggregation of 2nd instar nymphs collected in the field. Parents' Diet Green Beans Peanuts Green Beans Peanuts Water Only No. mating pairs set up 6 9 3 No. pairs laying normal, 6 1* 3 viable eggs No. viable egg masses 6-10 8 3 produced in lifetime (range) Interval between egg masses 5-8 5-6 8 (days) No. eggs per egg mas §** n Average Range Hatch success (%) per egg mass""* n • 12 3 7 Average 97.5 92.4 76.8 Range 93.9 100.0 87.3 98.2 52.3 100.0 12

PAGE 64

54 diets reproduced normally. The number, size, and hatch success of egg masses laid by the one peanuts -and-water pair fell within the range of parents reared on green beans and peanuts. On the other hand, the three pairs reared on green beans alone produced fewer egg masses at longer intervals. These egg masses were slightly smaller and less viable on the average. Despite the differences in laying interval and viability, the 4th instar color morphs of broods from parents reared on green beans alone did not differ from broods of parents reared on green beans and peanuts (Figure 3-4). The only significant difference was for the broods of the parents reared on peanuts and water: the three half-broods reared on green beans and peanuts had much higher proportions of green morphs. Discuss ion Variation between Succ essive Broods Environment al v ariation The striking differences in the ratios of color morphs that occur among broods of a single parental pair (Figures 3-1; 3-2, 3-3) indicate that nongenetic factors are important in the determination of 4th instar color. The greater variation of summer 1981 broods (Figure 3-1) may be because rearing conditions, particularly temperature and photoperiod, were less controlled than in 1983 (Figure 3-2). Cooler rearing temperatures bring about higher frequencies of black morphs in laboratory populations (Kariya 1961). Chapter 5 of this dissertation deals further with this effect. Short photoperiod stimulates increased melanism in a number of insects (e.g., Watt 1969,

PAGE 65

55 co si '_ 4-1 4J 5-1 •H CO CD (%) AouanbaJj qd.ioy\| c (V i_ Qo c > u • ii T3 w en 5 to 01 to c -h 4-1 k. 6 CO ,— I en co u u C 5) Mil n u mj: n c

PAGE 66

56 Ishii and Hidaka 1979), and nymphs in summer 1981 may have perceived changing day length because of a laboratory window. Nutritional variation. The summer 1981 rearings (Figure 3-1) clearly show that rearing nymphs on the peanuts and water diet results in increased incidence of melanism in the 4th instar. The effects of diet are examined further in Chapter 4. Purchased green beans differed in maturity and probably nutrition from week to week, and Nezara does show a preference for young fruit over mature in the field (Drake 1920). I tried to minimize this variation by purchasing enough beans at once to feed all cultures for several days. Peanuts varied less because a single bag could supply nymphs for weeks. However, in 1981 rapid growth of fungus on old peanuts sometimes occurred between cleaning periods. I often found nymphs feeding on those infested peanuts. In 1983, I successfully prevented fungus outbreaks by removing old peanuts sooner. The water source provided for nymphs on the peanuts and water diet might have retained salivary enzymes of drinking nymphs. Such a variable chemical cue might account for some variation between broods reared on peanuts and water. Maternal effec ts. Despite the possible influences of rearing conditions on brood color morph frequencies, there are indications that maternal factors affect the variation between successive broods. Broods from egg masses laid on about the same day by different parents should experience very similar rearing conditions. If conditions change to favor black morphs, then all broods should exhibit an increase in black morphs relative to the preceding broods. This was not always the case (Figures 3-1, 3-2, 3-3). In fall 1983 (Figure 3-2), for example, pair E's last brood on September 13 had a marked shift toward darker morphs,

PAGE 67

57 while pair F's September 12 brood had a high green morph frequency, consistent with preceding and succeeding broods. With the exception of pair A, all of the brood series reared on green beans and peanuts in 1983 (Figure 3-2) showed distinct shifts toward darker morphs in the last one or more viable egg masses laid before the female's death, independent of the date. A similar trend was observed in 1981 (Figure 3-1). Offspring of older Nezara viridu la females tend to have lower larval survivorship, lower adult longevity, and lower fertility than offspring of younger females (Kiritani and Kimura 1967). My data (Figure 3-1, 3-2) provide evidence that female senility also affects coloration of offspring. Wellington (1965) showed that the quantity of yolk allocated to tent caterpillar eggs was closely related to the activity and vigor of the hatchlings. The last eggs laid by a female had less yolk and produced less vigorous hatchlings. However, Nezara egg mass size, egg size, and hatch success did not vary significantly when sudden increases in black morphs occurred in a brood series (Table 3-1, 3-2). On the other hand, nymph survivorship to the 4th instar decreased in six out of seven cases, and increased by only 0.9% in the one other case (Table 3-5). Although increases in black morphs were not preceded by measurable changes in egg size, a maternal effect may still have been transmitted in the yolk. In addition to providing lipids, proteins, and carbohydrates, maternal cells of Oncopeltus (Lygaeidae) transfer RNA to developing oocytes (de Wilde and de Loof 1973a). Yolk deposition in Rhodnius (Reduviidae) and other insects is activated by secretions from the corpus allatum that seem identical to juvenile hormone (de Wilde and de Loof 1973b). Corpus allatum activity is reduced in diapausing

PAGE 68

58 insects (de Wilde and de Loof 1973b) and may change with aging. Female cecropia moths (Saturniidae) supply significant quantities of juvenile hormone to their eggs, but the significance to embryonic and nymphal development is unknown (Gilbert 1964, Gilbert and King 1973). Effect of Parentag e on Average_Color_Morph_ Fr equen c i e s In spite of the brood-to-brood variations, there was an overall significant effect of parentage on the coloration of broods (Table 3-3, 3-4). The effect was not as strong among the sibling parents of summer 1981 (Table 3-3), where broods from half of the parents averaged 51-61% green morphs In the fall 1981 data set, the range of green morph frequencies was wider and more evenly distributed (Table 3-4). Here there were two sets of sibling pairs (Zl, Z2 and Zl 1 Z13, Z14, Z15), and the other pairs were unrelated matings. The parental effect on nymph color morph was definitely associated with the color morph of the parents (Table 3-5). On the average, the broods laid by parents which had been black nymphs themselves in the 4th instar had relatively more black morphs than the broods of parents which had been green morphs. However, the ratios of morphs obtained are not typical of other genetically controlled polymorphisms. For example, melanic patterns of Coelophora inaequalis (F.) (Coleoptera: Coccinellidae) are controlled by eight alleles of a single gene (Houston and Hales 1980). Numerous intermediate forms are heterozygotes but the blackest morph is a simple homozygote. Nezara 4th instar black color cannot be a simple homozygotic trait because black morph crosses did not result in 100% black morph offspring (Table 3-5). Likewise, if inheritance of 4th instar color pattern were polygenic (Strickberger

PAGE 69

59 1968), then black-black crosses would at the least produce a majority of black morphs. Instead, most crosses produced less than 50% black morphs and one cross produced no black morphs at all (Table 3-5). Therefore, the parental effects observed are either the result of maternal inheritance or a complex system of modifier genes. Maternal transmission of environmentally influenced polymorphisms has been observed in other insects. Melanism induced by low temperature in the adult wasp Habrobracon and the lygaeid bug Oncopeltus can be passed on to the next generation ( Wigglesworth 1965). On the other hand, there are examples of melanic polymorphisms which vary with environment but are also influenced by genetic inheritance. These include a noctuid moth larva (Long 1953), a tortricid moth larva (Baltensweiler 1977), a geometrid moth (Majerus 1981), Colias butterflies (Roland 1982), and a syrphid fly (Heal 1979). Effect of P arental Diet Parental diet had a definite effect on fertility (Table 3-6), but its relationship to offspring coloration is still unclear. Most adult pairs reared on peanuts and water were infertile. Yet, egg mass size and hatch success of masses from the onp fertile pair were similar to egg masses laid by parents reared on green beans and peanuts (Table 3-6). Egg masses laid by parents reared solely on green beans were smaller and had lower hatch success than eggs of other parents. Nevertheless, out of all of the half-broods reared on either diet, only the broods laid by the peanut-fed parents were different (Figure 3-4). Their half-broods reared on green beans and peanuts had twice the frequency of green morphs in comparison to nymphs from the other

PAGE 70

60 parents. Unfortunately, it is impossible to separate genetic and maternal causes in this case. Further repetitions would be difficult to obtain because there is such high infertility among parents reared on peanuts and water. A better approach might be to rear nymphs on the standard diet and switch adults to different diets. Conclusions Nezara viridula adults contribute significantly to the distribution of color morphs among their 4th instar offspring. In particular, the parents' coloration during their own 4th instar is associated with higher proportions of that color type among their offspring. However, this research did not determine the exact mode of inheritance, whether genetic or maternal. A maternal effect is suggested by a consistent pattern of higher black morph frequencies in the broods of senile females. In addition, higher black morph frequencies in successive broods are associated with lower nymph survivorship. Genetic and nongenetic parental effects are difficult to assess because of the flexible response of nymph coloration to environment. Nezara nymph diet (Figure 3-1; Chapter 4) and rearing temperature (Kariya 1961; Chapter 5) can cause as much variation between two halves of a single brood as can be found between broods of any two parental pairs. Knight (1924) looked for patterns of genetic inheritance of the color polymorphism of nymphs and adults of another pentatomid, Perillus bioculatus Fabricius. After three years of breeding experiments, he finally concluded that color patterns of individual bugs were more strongly influenced by external conditions, such as temperature, than by inheritance from parents. The same appears to be true for Nezara viridula nymphs

PAGE 71

CHAPTER FOUR EFFECT OF NYMPH DIET ON COLOR MORPH RATIOS Introduction Nymphs of the southern green stink bug, Nezara viridula, undergo dramatic changes in coloration during postembryonic development. Although all instars retain the same basic patterns of white and red abdominal spots, dorsal ground color changes from reddish brown in the 1st instar to black in the 2nd instar, and ultimately to green in the adult. The green ground color may be attained as early as the 3rd instar and is accompanied by reduced melanization of the head capsule, thorax, legs, and abdomen (Chapter 2). The consequence of the variation in the timing of this color change is that 3rd, 4th, and 5th instar populations can be polymorphic. Third instar nymphs are usually black and 5th instar nymphs are usually green, but 4th instar nymphs often range from black to green with intermediate stages. Jones (1918) provides descriptions of Nezara nymphs, including dark and light forms of the 4th and 5th instars. A comparison of morphs is detailed in Chapter 2 This investigation is based on my initial observation that the frequency of melanization among laboratory-reared 4th instar nymphs differed from that of nymphs reared in an outdoor green bean garden, even though both groups were from the same parental stock. I then 61

PAGE 72

62 discovered differences that seemed to be associated with the type of laboratory diet the nymphs had received. Food plant species, plant quality, and amino acid content of diet have been associated with differences in the degree of and frequency of occurrence of melanism in a variety of insects. Under crowded conditions, larvae of the moth Plusia gamma L. (Lepidoptera: Plusiidae) range from light green to almost black, depending on the amount of diffuse melanin in the cuticle (Long 1953). Long found that larvae reared on broccoli were distinctly less melanized than larvae reared on five other plant species. In solitary cultures, 20% of Er innyis ello L. larvae (Lepidoptera: Sphingiidae) reared on Poinset tia were brown and 80% were green. However, up to 90% of larvae reared on Euphorbia were brown (Schneider 1973). In contrast to these examples, the polymorphic melanic pattern of larvae of Papilio dem odocus (Lepidoptera: Papilionidae) in South Africa is not determined by diet, even though the distribution of pattern type is closely associated with food plant species in the field (Clarke et al. 1963). In that case, the authors suggest that visual predators have selected for different genetically-determined cryptic patterns on the different plant backgrounds. In general, poor food quality increases melanism in insects. E. ello larvae were more commonly brown when fed defoliated Poinsettia shoots instead of fresh shoots (Schneider 1973). Nymphs of the grasshopper Paulinia acum inata (de Geer) which feeds exclusively on a single aquatic fern species ( Sa lvinia ) changed from green to brown or black in successive instars when switched from fresh green fern to old brown fern (Meyer 1979). A higher proportion of lime aphid nymphs, Eucal 1 ipte rus t iliae (L.), developed black cuticular bands when they were reared on

PAGE 73

63 mature leaves instead of young leaves (Kidd 1979). On the other hand, Hintze-Podufal (1977) concluded indirectly from her experiments that Saturnia (Eudia) pav onia L. larvae (Lepidoptera : Saturniidae) feeding on drier food plants become less melanized than those feeding on fresh, moist food. Goldberg and De Meillon's (1948) study on nutritional needs of mosquito larvae ( Aedes egypt i L. ) produced direct evidence that nutritional components of diet affect melanizat ion. Larvae reared on artificial diets free of the amino acids tyrosine and phenylalanine did not develop normal melanic larval pigmentation. In this study, I compare the effects of a diet of fresh green beans, shelled raw peanuts, or a combination of the two on the frequency of occurrence of color morphs in the 4th instar. In addition, I test diet changes during the 2nd and 3rd instars to determine whether a sensitive period for 4th instar color determination exists. The effect of diet on development time and mortality also gives insight to the mechanism by which 4th instar color pattern is determined. Materials and He t hod s General Ap p ro ach All experiments were based on matched comparisons of two groups of siblings from a single egg mass. This protocol controlled for genetic or maternal factors that might influence color pattern development of nymphs. Although the number of nymphs varied in each repetition of an experiment, matched experimental pairs had close to the same number of nymphs

PAGE 74

64 Parental Stocks Nymphs and adults of Nezara viridula from Gainesville, Florida, populations were reared in the laboratory on a diet of green beans ( Phaseolus vulgaris ) and shelled raw peanuts (Arachis hypogaea) purchased in local supermarkets. In 1981, adults were kept as pairs in 100 mm plastic petri dishes or in groups of five pairs in quart mason jars. Every other day, I provided fresh food and clean paper toweling on the bottoms of dishes and jars, cleaning containers whenever necessary. Direct illumination was provided by incandescent and fluorescent lights; some indirect sunlight came in through a laboratory window. The light regime was 15L:9D. Room temperature was 25-27C. The parental stock for 1983 experiments was composed exclusively of adults collected in the nymphal stages. Virgin adults were placed together as permanent mating pairs in separate petri dishes. They were maintained as in 1981, except that each pair received fresh green beans daily, instead of every other day. Adults usually laid eggs directly on the paper toweling on the bottom of the petri dish or on strips of paper hanging in the mason jar. Eggs laid elsewhere were easily pried off with a razor blade and glued to paper toweling with white glue. This procedure had no effect on hatching success. I collected eggs daily, cutting out the small square of toweling around each egg mass. I kept these egg masses in petri dishes which had a 2.5 x 2 x 0.3 mm piece of cellulose sponge or cotton cheesecloth taped to the lid and moistened to provide humidity. In 1981, egg masses were assigned arbitrarily to different experiments, without respect to parentage. In 1983, however, all repetitions of each

PAGE 75

65 experiment used egg masses from different mating pairs. In addition, the egg masses in each experiment were all the same rank order. For example, the first egg masses laid by 22 females were assigned to the first experiment; the second egg masses were assigned to another. Experimental Proce dures Rearing experiments in 1981 were started in two ways: (1) One to two days before hatch, egg masses were broken into two parts having about the same number of viable eggs, and the toweling was cut between the halves. Each half was arbitrarily assigned to one of two treatments (or the same treatment if a control) and placed in a 100 mm petri dish with the appropriate diet (Figure 4-1). (2) The entire egg mass was allowed to hatch in a petri dish having only moistened sponge or cheesecloth. First instar nymphs appeared to behave normally and usually aggregated on or next to the water source by the second day. Within 24 hours of the first molt, the 2nd instars were divided arbitrarily into separate dishes with the appropriate diet treatment (Figure 4-1). The data from rearings begun in either of the two ways were pooled, since there appeared to be no consistent effects attributable to the starting technique. The egg-splitting technique was faster and easier, and was used for all 1983 experiments. All 1983 experimental treatments were started on the 4th day after eggs were laid. Figure 4-2 outlines the experimental design used to determine whether a diet difference during the 2nd or 3rd instar affects color morph ratios. In 1981, experimental and control dishes were kept in the same lab area as the adults and developing eggs. Every day, I supplied distilled

PAGE 76

66 DIET TREATMENT EXPERIMENTAL GROUPS (1981) Half A: green beans (2 segments) / peanuts (2) egg mass < water \ Half B: peanuts (3-4) water CONTROL GROUPS (1981) Half A: green beans (2 segments) / peanuts (2) egg mass < water \ Half B: green beans (2 segments) peanuts (2) water Half A: peanuts (3-4) / water egg mass < \ Half B: peanuts (3-4) water EXPERIMENTAL GROUPS (1983) Half A: green beans (2 segments) / peanuts (2) egg mass < \ Half B: green beans (3-4 segments) CONTROL GROUPS (1983) Half A: green beans (3-4 segments) egg mass < Half B: green beans (3-4 segments) Half A: green beans (3-4 segments) egg mass < Half B: further divided at 2nd instar into groups of 15 or fewer nymphs per dish, each with 3-4 green bean segments. Figure 4-1. Experimental design for testing the effects of diet on color morph ratio. Green bean segments were 4-7 mm long; raw peanuts weighed about 1 g each. Water was provided by a moistened piece of sponge or cheesecloth attached to the lid of the rearing dish.

PAGE 77

67 DIET TREATMENTS DURING EACH INSTAR First I nstar Second Inst ar Third Instjrr --(A) green beans / peanuts egg mass --> green beans > green beans --< peanuts peanuts \ --(B) peanuts water egg mass --> peanuts water peanuts water --(A) green beans / peanuts (B) peanuts water -(A) green beans > green beans / peanuts peanuts egg mass --> water only --< \ -(B) green beans > peanuts peanuts water -(A) peanuts > green beans / water peanuts egg mass --> water only --< \ -(B) peanuts > peanuts water water -(A) water > green beans > green beans / only peanuts peanuts egg mass -< \ -(E) water > green beans > peanuts only peanuts water Figure 4-2. Experimental design for determining the effect of a diet switch in the 2nd or 3rd instar on the color morph ratios in the 4th instar. See Figure 4-1 caption for details of food quantities. Last treatment executed in 1983; all others run in 1981.

PAGE 78

68 water to the sponge or cheesecloth as needed and replaced moldy peanuts and green beans. Every other day, I transferred nymphs to a clean petri bottom with new paper toweling, fresh green beans, and new peanuts. During the 1983 experiments, nymphs were reared in an environmental chamber at 23 C, with a 12L:12D light regime. Regular care of nymphs was the same as in 1981, except that I provided fresh green beans each day instead of on alternate days. I also recorded nymph deaths and the date of appearance of the first 4th instars in the dish. Because the molt to the 4th instar is asynchronous, I chose to terminate rearings when 85-100% of the nymphs in a dish had molted. Usually only three or fewer 3rd instars remained. I assigned 4th instars to eight color morph categories based on the degree of melanization of the dorsal cuticle of the thorax (Figure 2-1). Pattern types ranged from unmelanized (only a few minute black dots) to completely black (except at the lateral margins, which remained golden yellow). The unmelanized nymphs had a light green ground color on the thorax and abdomen. Nymphs with black thoraxes had very dark black-brown abdomens, although the only melanized areas were black margins around the medial scent gland openings and semicircular markings around the connexivum (margin of the abdomen). Nymphs in the middle color morph categories usually had a more golden-yellow color in the unmelanized part of the thorax, and their abdomens were a very dark, blackish green. Data Analysis I combined the eight original color morph categories into 3 major categories for analysis:

PAGE 79

69 (1) green : the two lightest colored categories (I-II, Figure 2-1); the maximum extent of melanization is a central U-shaped pattern of black lines, a few spots, and finger-like patterns projecting inward from the lateral margins; thorax ground color green; abdomen ground color light green to dark green. (2) intermediate : the three middle categories (III-V, Figure 2-1); black markings on thorax extensive, but covering less than 3/4 of area; thorax ground color green to golden yellow; abdomen ground color dark green to black-brown. (3) black : three darkest categories (VI-VII, Figure 2-1); more than 3/4 of thorax black (except for yellow lateral margins); unmelanized spots are golden-yellow; abdomen ground color black-brown. I calculated color morph frequency for each half of an egg mass by dividing the total number of 4th instar nymphs in a category by the total number of 4th instar nymphs counted. Frequencies are expressed here as percentages. Differences between treatments were tested with the Wilcoxon Matched Pairs Signed Ranks Test (Siegel 1956). Percent mortality during each instar was calculated by dividing the number of deaths at the end of the instar by the number of nymphs present at the beginning of the instar. Differences were tested with the Wilcoxon Test. Development time for an egg mass half was defined as the number of days between egg-hatch and the first appearance of 4th instar nymphs in the group. Differences between two halves were tested by the Sign Test (Siegel 1956).

PAGE 80

70 Results Single-Item Diets The frequency of melanism is greater among 4th instar nymphs reared on a single food item than on the combined green bean plus peanut diet (Table 4-1). Compared to the combination diet, the peanut diet resulted in fewer green morphs and more black morphs, but about the same frequency of intermediate morphs. The green bean diet also led to a shift toward darker morphs in comparison to their siblings reared on the combination diet. There were no significant differences between the morph frequencies of sibling groups reared on identical diets. Nymphs reared on green beans alone produced liquid waste at a much faster rate than their siblings on green beans and peanuts. The 30-70 2nd or 3rd instar nymphs from one half of an egg mass usually produced enough waste in 24 hours to completely saturate three layers of paper toweling. Their siblings which were feeding on both green beans and peanuts only slightly dampened the paper linings. To check the effect of increased humidity and waste product, I ran an additional eight control masses on the green bean diet. For these controls, I divided the 2nd instars from one of the half-mass groups into separate dishes containing no more than 15 individuals. These "uncrowded" nymphs did not differ significantly from their "crowded" siblings with respect to morph frequency (Table 4-2). In fact, the crowded group had more green morphs and fewer black morphs in five out of the seven cases with differences

PAGE 81

71 Table 4-1. The effect of diet on morph ratios. Samples are matched by assigning different treatments to each half of an egg mass. Numbers 1 3 are 1981 experiments; 4 and 5 are from 1983. Values in parentheses are standard errors of the mean. GB = green beans, PN = peanuts, W = water, N = number of cases with differences, T = statistic for Wilcoxon Matched Pairs test. SS = p < 0.01, S = p < 0.05, NS = p > 0.05

PAGE 82

72 Table 4-2. Effect of rearing nymphs in a single dish ("crowded") or in groups of less than 15 ("uncrowded" ) with a diet of solely green beans. Matched by dividing single egg mass into 2 treatment groups. N = number of cases with differences, T = Wilcoxon statistic, NS = p > 0.05 Treatment of Each Half Crowded Uncrowded No. of egg masses tested 8 Mean no. of 4th instars 40.6 + 4.0 39.5 + 2.5 Mean % green morphs 9.6+5.4 5.4+2.7 N = 7, T = 7, NS Mean % intermediate morphs 17.5 + 4.3 11.1 + 4.1 N = 7, T = 5, NS Mean % black morphs 72.9 + 9.0 83.5 + 6.1 N = 7, T = 6, NS

PAGE 83

73 Switching Diet s A diet switch during either the 2nd or 3rd instar affected the outcome of morphs in the 4th instar (Figure 4-3). Nymph groups switched from peanuts and water to green beans and peanuts at the beginning of the 3rd instar had more green morphs and fewer black morphs than their sibling groups which remained on peanuts and water (Figure 4-3, Treatments 2 and 4). However, the reverse diet change (from green beans and peanuts to peanuts and water) only had an effect when 1st instars were unfed (Figure 4-3, Treatments 1 and 3). In comparison to siblings which received green beans and peanuts during both the 2nd and 3rd instars, nymphs switched to peanuts and water during the 2nd instar showed a significant shift towards black morphs (Figure 4-3, Treatment 5). Mortality and Gro wt h Ra te Mortality in the 2nd and 3rd instar was higher among nymphs reared on peanuts and water compared to their siblings reared on green beans and peanuts (Table 4-3). Mortality among 2nd instar nymphs reared on green beans alone was significantly higher than their siblings reared on green beans and peanuts. Overall mortality in all experimental groups was extremely low. Nymphs reared on single-item diets developed more slowly than those reared on the combination of green beans and peanuts (Table 4-4).

PAGE 84

rE z z 0. Q-5 z z a. a. 00 00 5 93

PAGE 85

75 >-, 4J 4-1 tO QJ

PAGE 86

76 Table 4-4. Development time (number of days) from egg-hatch to 4th instar for nymphs reared on different diets. Observations were made once per day in late afternoon. Recorded time is appearance of the first 4th instar nymph(s) in the group. GB = green beans, PN = peanuts, W = water. GB, PN W, PN diff. GB PN GB diff. Mean

PAGE 87

77 Discuss ion Diet-Induced Mela nism These results differ from previous studies by showing that in comparison to a single item diet, a combination of foods can reduce the occurrence of melanization (Table 4-1). In addition, Nezara nymphs are melanic in the earlier instars and make a transition to an unmelanized cuticle, whereas previous studies have been concerned with non-melanic immatures which develop melanization in later stages (Long 1953, Hintze-Podufal 1974, 1977, Kidd 1979). Under the conditions tested, once a Nezara nymph had molted to an unmelanized cuticle in the 3rd, 4th, or 5th instar, successive molts were also unmelanized. I could not induce melanic adults. Meyer (1979) was able to induce dark grasshopper nymphs to molt into green nymphs and back again by changing the diet. He also obtained dark adults from dark nymphs. Hintze-Podufal (1977) reversed the direction of melanization between the 4th and 5th instar larvae of E. pavonia under some experimental conditions. However, melanic aphid nymphs (Kidd 1979) and 4th instar moth larvae (Schneider 1973) retained their developed melanism throughout the remaining immature stages. The melanization effect does not seem to be caused by the increased humidity from the artificial water source or the green beans. The green bean and peanut diet resulted in fewer melanized 4th instar nymphs even when the water source was included as part of the treatment (Table 4-1, Treatment 1). Nymphs fed only green beans showed the same degree of melanization whether in a large group in one dish where moisture quickly accumulated or in small groups in dry containers (Table 4-2).

PAGE 88

Timing of Color De terminat ion The color pattern expressed during the 4th instar seems to be determined by the length of time nymphs are allowed to feed on both green beans and peanuts (Figure 4-1). In the cases where significant differences in color morph frequency occurred, the group which had fed on both green beans and peanuts for the longer time had the higher proportion of non-melanics (green). Changes in food type during either the 2nd or 3rd instar affected the ratio of color morphs in the 4th instar. Thus, there is not a single "critical time" that determines the pattern. Hintze-Podufal (1977) found that changes in experimental conditions in the 2nd or 3rd instar affected the 4th instar color pattern of a saturnid larva, and conditions in the 3rd and 4th instar affected the 5th instar color pattern. Nutritional Defi c its a nd_ E xc esses The higher mortality among the 2nd and 3rd instars reared on peanuts and the 2nd instars reared on green beans (Table 4-3) suggests that these single-item diets are nutritionally poorer than the combination diet. The longer development time of nymphs reared on the single-item diet (Table 4-4) also points to a nutritional deficit. Green beans are about 85% water compared to 5% water in peanuts and by weight contain only about 10% of the protein content of peanuts (Souci 1981). They also differ in amino acid and vitamin content by weight and relative proportion. Researchers who rear large laboratory colonies of Nezara v iridu la have best results with a regular diet of green beans and raw peanuts, with other items added from time to time (Harris and Todd

PAGE 89

79 1981). Nezara reared on artificial diets take longer to mature and suffer higher mortality (Jensen and Gibbens 1973). Nutrition has been linked to other kinds of polymorphisms in insects, such as castes in social insects (Wigglesworth 1954). In Drosophila the penetrance of the mutant character tetraltera increased when flies were reared on food which prolonged development (Wigglesworth 1954). Kidd (1979) suggested that the decline in soluble nitrogen in mature leaves was responsible for increased melanization among lime aphid nymphs. The important role of the amino acid tyrosine in the production of normal dark pigmentation in Aedes mosquito larvae (Goldberg and De Meillon 1948) can be explained by the fact that tyrosine is a melanin precursor (Neville 1975, Kiguchi and Kimura 1981). Because quinones and phenols are used in the polymerization process that forms melanin (Bursell 1970), melanization may be a means of disposing excess quantities of these potentially toxic by products of metabolism (Wigglesworth 1965). Hormonal Rol e The hormones responsible for insect growth and development also play a role in melanin production. Tyrosine is present in high concentrations in moth larval tissues immediately before ecdysis, and its release from the fat body seems to be caused by ecdysone (Kiguchi and Kimura 1981). Juvenile hormone does not affect this tyrosine release (Kiguchi and Kimura 1981) but does prevent melanization in at least two moth species (Kiguchi 1972, Hintze-Poduf al 1976). The presence of the hormone bursicon is essential for the sc lerot izat ion of the cuticle, and its concentration may be correlated with the degree of melanization (Neville 1975).

PAGE 90

Thus, diet may be affecting Nezara nymph color patterns directly, by providing chemical precursors for melanin production, or indirectly, by affecting the nymph's production of ecdysone, bursicon, or juvenile hormone. When nymphs are restricted to only one food item, they may have to take in excess amounts of certain amino acids or other melanin precursors in order to obtain minimum amounts of essential nutrients. Therefore, one would expect more melanism in nymphs which have been restricted to a single item diet for a longer time (Figure 4-1). Secretion of the hormone ecdysone is likely affected by the poorer diets, since development time is increased (Table 4-4). I have no data that links bursicon or juvenile hormone to the melanizing effects of diet. However, unmelanized cuticle is an adult characteristic of Nezara and the completely melanized 3rd and 4th instar nymphs are identical in color and pattern to 2nd instar nymphs. Thus in this case, juvenile hormone may act to retain melanism, rather than prevent it. Ecological Functio n of Die t-Indu ced Mela nism Does the melanization of the later instars provide any advantage to the nymph? The melanic cuticle may serve as a safe dumping site for toxic metabolites (Wigglesworth 1965). Melanin is known to add mechanical strength and to shield UV light (Neville 1975). It also strongly absorbs radiant energy in the visible spectrum (Watt 1968), and its possible role in thermoregulation is further discussed in Chapter 5. The melanized patterns may aid in avoidance and escape from visually-oriented predators. The most common escape response by a Nezara nymph is to simply let go and drop. A black or intermediate nymph would be cryptic if it dropped on to bare soil or leaf litter. A

PAGE 91

green morph would more easily evade detection if it dropped on to green vegetation. This potential crypticity may be important, since some vertebrate predators do eat nymphs (Stam 1978). In the typical seasonal pattern of Nezara adults move to different host plants as they become available, and nymphs are free to feed on many plant parts (leaves, stems, flowers, and fruits; Todd and Herzog 1980). However, in late fall, suitable young host plants are no longer available. Since Nezara nymphs disperse very little (Panizzi et al. 1980), an egg mass deposited in a patch of senescent food plants would leave the nymphs with a limited diet of stems and mature pods. Since the nymphs are poiki lothermic additional solar energy absorbed by the melanized cuticle could help to counteract slower growth rates caused by either poor diet or cooler ambient temperatures. It is critically important for nymphs to reach maturity before winter begins because only adults are able to overwinter successfully (Drake 1920, Todd and Herzog 1980). Cone lusions Diet has a significant effect on the expression of melanism in Nezara nymphs. This flexible color polymorphism provides a range of opportunities for further investigation into hormonal control of genetic expression and the ecological advantages of such flexibility. Rearing experiments on living plants are needed to test whether differences in availability of flowers, developing fruit, and/or mature fruit can produce differences in color morph ratios. Comparisons of field collections in young and senescent food patches at the same time of year should be made to confirm the diet effect in the field.

PAGE 92

CHAPTER FIVE EFFECT OF TEMPERATURE ON COLOR MORPH RATIOS Introduction Pigmentation darkening in insects has often been associated with cooler temperatures (Wigglesworth 1965). Most often, the dark coloration is due to an irreversible increase in melanin pigment. The best studied examples are in the Lepidoptera. Early spring and late fall broods of Colias and Na thai is (Pieridae) have more melanic scales on the underside of the wings than mid-summer broods (Watt 1969, Douglas and Grula 1978, Hoffman 1978). Dark autumnal forms have also been described for a skipper (Hesperiidae ) in Japan (Ishii and Hidaka 1979) and a noctuid moth in England (Myers 1977). The sycamore aphid has melanic pigmentation only in the spring and fall seasons (Dixon 1972). Melanic forms of some polymorphic caterpillars become more frequent at lower temperatures (Fye 1979, Baltenswei ler 1977). In addition to seasonal dark forms, increased melanism has also been found among insects at higher latitudes and altitudes. For example, the frequency of melanics in alpine Colias (Roland 1982) and Scandanavian bumblebees (Pekkarinen 1979) is greater than in their congeners in lower regions. In the cases where temperature has been associated with within-spec ies variation, short photoperiod and cold rearing temperature are the important factors that stimulate an increase in melan izat ion. In one

PAGE 93

83 saturnid moth caterpillar, however, melanics increased at higher rearing temperatures (Hintze-Poduf al 1977). Polymorphism in 4th and 5th instar nymphs of Nezara viridula is a result of variation in the amount of melanin in the cuticle and the shade and hue of subcuticular pigments in non-melanized areas. Coloration ranges from unmelanized forms, which have light green subcuticular pigmentation, to black forms, which have completely melanized thoraxes and black-brown subcuticular pigmentation. Jones (1918) provides complete descriptions of nymphs; diagrams of thoracic melanization patterns are provided in Chapter 2 of this thesis. The strong influence of diet on this polymorphism is discussed in Chapter 4. In a brief experiment in October 1981, I compared laboratory rearings on green beans and peanuts with caged outdoor rearings on green bean plants. All of the surviving 4th instar nymphs in the outdoor cages were black morphs, while many of their indoor siblings were green or intermediate in coloration. Earlier that summer, uncaged nymphs reared outdoors on green bean plants were usually less melanized than nymphs reared in the laboratory. Because night-time temperatures in October dropped below laboratory temperature, I hypothesized that temperature was a factor controlling melanization of Nezara nymphs. This is a report of laboratory rearings of Nezara viridula nymphs at different temperatures and consequent shifts in color morph frequencies. In addition, I include a laboratory test of the influence of melanization on heat absorption during basking at low ambient temperature and a discussion of its possible adaptive significance.

PAGE 94

84 Materials a nd Methods Rearing Nymphs at Diffe rent Tempe ratures A parental stock of Nezara viridula was formed from 4th and 5th instar nymphs from a single Gainesville, Florida, population collected in September 1982. I kept sexes in isolation until I united them as permanent mating pairs in separate 100 mm plastic petri dishes. Adults were maintained in the laboratory as described in Chapter 4. I collected egg masses daily and assigned them by random number to one of four experimental treatment groups (Figure 5-1). After incubation in the laboratory for three days, I divided each egg mass in half along the longer axis, and randomly assigned each half to one of the two rearing temperatures. On the fourth day of development, I placed the eggs in 100 mm petri dishes with the appropriate diet and moved them to environmental chambers with 12L:12D lighting. Nymphs hatched out after 1-2, 2-4, or 5-8 more days at 28, 23, and 18 C, respectively. I monitored their progress daily, supplying fresh green beans at least every other day, and replacing peanuts when they became soft or moldy. I added fresh distilled water daily to the cheesecloth roll in the peanuts and water treatments. Every other day, I transferred nymphs to a clean petri dish bottom with fresh paper toweling. I recorded and removed dead nymphs daily and noted the date of first appearance of 4th instars in each dish. I counted and preserved the nymphs in each dish as soon as (1) all nymphs reached the 4th instar, (2) any nymphs molted to the 5th instar, or (3) two days elapsed without further molting. I assigned the 4th instar nymphs to one of eight color morph categories according to the

PAGE 95

85 Nymph Diet = Gre en Be a ns, Peanut s Treatment Group 1 Half A 28C egg mass < Half B 23C Treatment Group 2 Half A 23C egg mass < Half B 18C Nymph Diet = Peanuts, Water Treatment Group 3 Half A 28C egg mass < Half B 23C Treatment Group 4 egg mass < Half A 23 C Half B 18C Figure 5-1. Experimental design for testing effect of rearing temperature on color morph ratio.

PAGE 96

86 degree of melanization of the thorax. (See Chapter 2 for complete descriptions.) For data analysis, I combined the eight original color type categories into the three major categories describe in Chapter 4: green, intermediate, and black. Color-type frequencies, percent mortality in each instar, and development time were calculated and statistically tested as in Chapter 4. Cold Shock Ex per imen t From a second parental stock collected in July 1983, I obtained 20 egg masses. Each was the third egg mass laid by 20 different mating pairs. Three days after they were laid, I divided egg masses in half and randomly assigned each half to one of these treatments: (1) constant temperature-constant 23 C; green beans and peanuts diet; 12L:12D light regime. (2) cold shock-start out at 23 C, but on first day of 2nd instar, nymphs moved to 18 C; after 72 hours, nymphs moved back to 23 C; same diet and light regime as (1). For an additional control, I took the fourth egg masses produced by the same set of mating pairs and randomly assigned each half to one of two groups (A or B). Both groups were reared at a constant 23 C and with the green beans and peanuts diet. Heating Curve s Using an indoor laboratory set-up, I compared heating rates of live 4th instar black and green morphs under artificial lighting. A horizontal platform was made of 0.5 mm thick styrofoam and covered completely with white paper. A 26 ga (0.46 mm) hypodermic needle

PAGE 97

87 microprobe (Type MT-26/2, Bailey Instruments) was inserted through the styrofoam from below. Thus nymphs could be positioned horizontally on the platform surface and impaled by the probe from the underside. The light source was a 200 W incandescent lamp 22 cm directly above the platform surface. Nymph body temperature was measured with the microprobe and a digital readout instrument (Model BAT12, Bailey Instruments), which has a 0.1 s time constant and a 0.1 C resolution. I measured shaded ambient temperature with a mercury thermometer shaded by aluminum foil and positioned 5 cm above the platform. The entire apparatus was kept in a constant temperature room, which maintained a constant 15 C when the lamp was off. Test animals were selected arbitrarily, with an effort to match sizes, from among black and green 4th instar nymphs reared from three unrelated egg masses. Each nymph was cooled on ice for at least 30 min, then quickly transferred to the platform. I inserted the probe through the underside at the midline, just at the posterior margin of the thorax and penetrating as far as possible (2-4 mm). I then secured the bug with a thread across the body, made taut by a weighted plastic ring. These steps took less than a minute, and the impaled bug was always several degrees below ambient temperature at this point. I allowed the nymph to warm up slowly to 14.7 C; then I switched on the lamp to begin the run. Every 30 s, I recorded body temperature from the digital readout and shaded ambient temperature from the mercury thermometer. Most runs continued for 10 min, but some were stopped as early as 8 min if body temperature leveled off. All nymphs survived the runs and behaved normally after they were released.

PAGE 98

88 At least 10 min elapsed between runs to allow the apparatus to return to 15.1 C. A dish of ice placed on the platform facilitated complete cooling. Black and green morphs were run alternately. I measured live weights on a Mettler balance after each run, rather than before, because the nymphs lost hemolymph when the probe was inserted. Since all factors were held constant in each run, except nymph color and size, I could make direct comparisons of nymph body temperature at any given time. The influence of size on the heating rate was determined by plotting body temperature at 8 min against live weight and was tested by the Spearman rank correlation coefficient (Siegel 1956). I used the Mann-Whitney U-test (Seigel 1956) to test for significant differences in body temperatures after 4 min and 8 min of exposure to light. I constructed heating curves by plotting body temperature against time. Resu It s Lower rearing temperatures increased the degree of melanization among 4th instar nymphs (Figure 5-2). With a green bean and peanuts diet, 98% of nymphs were green at 28 C while only 2% were green at 18 C. Less than 1% were black at 28 C while almost 70% were black at 18 C. The average frequency of intermediates did not increase beyond about 30%, but greater frequencies did occur in individual cases. The same cooling effect occurred with the peanuts diet, except that at any given temperature, the proportion of melanics was greater. Also, the green, unmelanized morphs usually had a more yellowish green background color than nymphs reared on green beans and peanuts.

PAGE 99

00 ir 89 t£~ CM 03 O =3 C CL ^E € A I— O o o oo o -ro o CM u jc *o u o O -i 00 u ai o n 0) U li

PAGE 100

90 While rearing these nymphs at the different temperatures, I noticed dramatic differences in 3rd instar color patterns between matched sibling groups. In addition to the typical black patterns, unmelanized patterns appeared. These were similar to the green and intermediate patterns of 4th instar nymphs, except that the background colors were usually more yellowish. I classified 3rd instar groups by overall aspect: all black, mostly black (some intermediates), mixed (few blacks or greens, many intermediates), and mostly light (few or no blacks, many greens and intermediates). Only nymphs reared at 28 C fell into the mixed or mostly light categories, and their sibling counterparts at 23 C were always classified as all black or mostly black. More groups with unmelanized 3rd instars occurred with the diet of green beans and peanuts (Table 5-1). In all but one of the unmelanized 3rd instar groups, more than 95% of the nymphs molted into green 4th instar morphs, regardless of diet (Table 5-1). Development time increased with lower temperatures (Table 5-2). The mean number of days from egg-hatch to the emergence of the first 4th instar approximately doubled for each 5 C decrease. At any given temperature, nymphs reared on peanuts and water took longer to develop than nymphs reared on green beans and peanuts. These increases averaged only 0.85, 1.5, and 7.4 days at 28, 23, and 18C, respectively (Table 5-2). However, they are associated with major differences in color morph frequencies (Figure 5-2). Average mortality from the 1st through 3rd instar increased slightly between 28 and 23 C, but only the 23 to 18 C differences were significant (Table 5-3). Overall percent mortality at

PAGE 101

Table 5-1. Color differences in 3rd instar nymphs reared at 28C. Date are numbers of groups, where each group is composed of the nymphs from half of an egg mass, reared in a single dish. 3rd Instar Classification Mostly Light/Mixed Mostly Black/Black Green beans and peanuts diet Peanuts and water diet 4th instars 95-100% green 4th instars < 95% green 14

PAGE 102

Table 5-2. Development time (number of days from egg-hatch to 4th instar) of nymphs reared at different temperatures. Recorded time is the appearance of the first 4th instar nymph(s) in each dish containing one half of an egg mass. Observations made once per day in afternoon, n = number of half-egg mass groups used in calculation of means, N = number of cases with differences, x = number of fewer signs in the signs test, p = 1-tail probabilities. No. of Masses Treatment Tested Mean No, of Days S.E. Sign Test Green beans, peanuts diet 28 U C 23C 23"C 18C 16 10 16

PAGE 103

93 CO 73 £ o u

PAGE 104

94 18 C was four to five times greater than the values at 23 C. The greatest mortality occurred in the 2nd instar, often after several days of feeding and growth. Surprisingly, exposing 2nd instar nymphs to three days of 18 C temperature did not increase melanism in the 4th instar (Table 5-4). In fact, the groups of nymphs exposed to the cold shock had a slight, but significant, increase in green morph frequency and decrease in black morph frequency. The decrease in melanism occurred despite the induced 1.9 day increase in development time (Table 5-4). There were no significant differences in color-type frequencies when the two halves of egg masses from the same set of parents were both reared under constant conditions The body temperatures of 4th instar nymphs heated under incandescent lighting were independent of live weight for the range of weights tested (Figure 5-3). The differences in live weight between the green and black test groups were not significant (Table 5-5). Therefore, I constructed a single heating curve for each morph (Figure 5-4) by averaging the body temperatures of all runs. Green morphs heated up more slowly after the first minute and began leveling off at temperatures about 1 1.4 C lower than black morphs. The body temperature differences were significant (Table 5-5). Discussio n Nymphs Reared a_t_ Different Temp e rat u r e s The coloration of 4th instar nymphs of Nezara viridula is strongly dependent upon rearing temperature (Figure 5-2). Kariya (1961) made

PAGE 105

95 co w *J 3 c cq x en w hi u h
PAGE 106

% i-l 'M U I o \ I I I I I I I I I I OOoor s c

PAGE 107

97 Table 5-5. Comparison of live weights and body temperatures of green and black morphs used to construct heating curves. n = number of nymphs Green Morphs Black Morphs Statistical Test Live Weight (mg) n 13 12 Range 17.2 45.2 15.5 40.6 Mean 28.7 25.4 S.E. Mean 2.8 2.5 Body Temp. ( C) at 4 min. n 13 12 Range 30.8 33.9 32.7 35.0 Mean 32.56 33.73 S.E. Mean 0.24 0.24 Body Temp. (C) at 8 min. t-test p = 0.39 Mann-Whitney U p < 0.01 n 13 12 Range 34.7 37.9 36.3 39.7 Mean 36.38 37.78 S.E. Mean 0.31 0.25 MannWhitney U p < 0.01

PAGE 108

98 Figure 5-4. Heating curves of 4th irtstar nymphs. Values are mean body temperatures for black morphs (closed circles) and green morphs (triangles), and mean shaded ambient temperature (open circles). Vertical bars are + 1 S.E. Mean.

PAGE 109

99 first notice of this temperature dependency in an article published in Japanese. He presented color morph frequencies obtained for 4th and 5th instars of N. viridula and its congener, N. ante nnata reared at 20-30 C. His N. viridula rearings resulted in only about 43% green morphs at 30 whereas I obtained over 95% green morphs at 28 on the green beans and peanuts diet. At 25 and below, Kariya's cultures were more than 80% black morphs. His results for N. antennata were erratic; the highest frequencies of green morphs (30-40%) occurred at 27.5 and 22.5C. This tendency in Nezara for increased melanization with lower temperature is similar to the temperature responses found in the pentatomid Perillus bioculatus (Knight 1924), certain lepidoptera adults (Wigglesworth 1965, Myers 1977, Ishii and Hidaka 1979, Majerus 1981) and larvae (Ba ltensweiler 1977, Fye 1979), a polymorphic syrphid fly (Heal 1979), and a braconid wasp (Wylie 1980). However, Nezara has the unusual aspect of being darker in the earlier nymphal stages, such that the effect of cold temperature is to retain the same melanic pattern present in the earlier instars, rather than induce a new pattern. Green, unmelanized 3rd instar nymphs can be induced at a high temperature (Table 5-1), and almost all of the subsequent 4th instar nymphs are also green. This suggests that the control of cuticular melanization is related to quantitative factors which are subject to metabolic influence. That is, high temperature, by increasing metabolic rate, may increase the rate of production of, or destruction of, some key hormone or chemical precursor responsible for cuticular melanization. Since black 2nd instars always occur, the biochemical mechanism for melanizing cuticle is genetically present in all nymphs.

PAGE 110

100 The question is, then, what happens metabolical ly to cause such a wide variation in expression in later instars? Since all Nezara adults are green, the conversion from black nymphs must be fool-proof and ultimately invariant. One possible candidate for control is juvenile hormone (JH), since melanized cuticle is a feature of early juvenile stages of Nezara and JH is responsible for maintaining juvenile characteristics in insects (Bursell 1970). Williams (1961) proposed such a mechanism to account for larval color changes (from black to green) in cecropia silkworms. Nevertheless, the action of JH in other immatures ( lepidoptera ) is to prevent melanization (Kiguchi 1972, Safranek and Riddiford 1975, Hintze-Poduf al 1976). It is imperative, then, to test JH titres in black and green morphs of 4th instar nymphs. Also, by switching newly molted, unmelanized 3rd instar nymphs to cold temperatures, one could determine whether or not the lack of melanization is completely irreversible, as it appears when nymphs are reared at continuously nigh temperatures. The lack of green beans in the diet did not prevent the appearance of unmelanized cuticle in the 3rd and 4th instar nymphs at 28 C, even though light color types were less frequent (Figure 5-2, Table 5-1). The yellower tint of many of these nymphs is reminiscent of nymphs which have been preserved in alcohol. That is, the blue component (similar to anthocyanin, Wigglesworth 1965) seemed diminished, allowing the yellow component (probably xanthopter ine Wigglesworth 1965) to dominate. The blue component cannot be exclusively derived from green beans, because I have frequently obtained normally green 4th instars, as well as adults, reared only on the peanut diet at 23 C. Therefore, high temperature must interfere directly with the production of pigment from precursors which are available in peanuts.

PAGE 111

101 As expected, nymphs reared on the peanuts and water diet were darker than nymphs reared on green beans and peanuts at the same temperature. The effects of diet and temperature seem to be additive and not synergistic. That is, with either diet a 5 temperature drop has about the same degree of effect on the color type frequencies. The additive effects of temperature and diet are not related strictly to increased development time. For example, between 28 and 23 C on the green bean and peanut diet, the frequency of green morphs dropped almost 50%, and development time increased by eight days (Figure 5.2, Table 5.2). But, even though the frequency of green morphs reared on the peanuts-only diet at 28 C was 30% less than that for the green bean and peanut diet at the same temperature, development time increased by only 0.85 days. Similarly, nymphs reared on green beans and peanuts at 18 C took about 23 more days to develop than nymphs on the same diet at 23 C, while an equivalent color morph difference between the two diet treatments at 23 C was obtained with only a one-day increase in development time. Therefore, the diet's effect on the frequency of melanization can be said to have a greater effect than rearing temperature, and the mechanism by which it exerts this effect may be different Effects of Cold Shock The cold shock experiment gives further insight into the independence of development time from other factors controlling melanization (Table 5-4). Even though nymphs exposed to 18 C for three days took about two days longer to reach the 4th instar, they were slightly less, rather than more, melanized. Thus, the nymphs seemed to have

PAGE 112

102 responded directly to temperature change (from cold to warm), rather than to the actual temperature. This is in contrast to experiments in which diet was switched after the 2nd instar (Chapter 4). There, switching from a peanuts diet in the 2nd instar to a diet of green beans and peanuts in the 3rd instar still resulted in more melanics than nymphs which had been fed green beans and peanuts constantly. Again, hormonal control of melanization may explain these results. Cold temperature may switch off, or slow the rate of, glandular secretion. Switching back to a warmer temperature may cause over-compensation of glandular function. Another possible explanation is that temperature exerts its influence only at the time of the molt. I have never noticed distinct differences in color types when I have moved late 3rd instars to cold chambers, but this should be tested systematically. Additional cold shock experiments should cover the periods immediately before and after the 2nd and 3rd instar molt. Melanism and Basking I nse c ts The responsiveness of nymph coloration to temperature strongly suggests that there is an adaptive value to the retention of black coloration in later instars in cold environments. Of the possible functions of melanized cuticle, the most widely investigated has been its role in thermoregulation (Hamilton 1973, Casey 1981). The most important avenue for heat gain in an insect is through solar radiation (Digby 1955, Casey 1981). Even though long infra-red radiation is absorbed equally well by any color tissue (Casey 1981, Watt 1968), energy in the visible portion of the spectrum is absorbed more completely by a black surface than one of a single hue. Watt (1968)

PAGE 113

103 demonstrated this difference with spectrophotometry analyses of melanic and immaculate yellow wings of Colias The absorptive difference accounted for a 2-3 C higher thoracic temperature in basking melanic butterflies in comparison to immaculate forms. Similar differences in body temperature of dark and light basking insects have been found in desert tenebrionid beetles (Edney 1971), locusts (Hill and Taylor 1933), and the salt-marsh caterpillar (Fye 1979). In all of these cases, dark color has been associated with basking behavior. Roland (1982) demonstrated that the greatest advantage to melanic butterflies comes when ambient temperatures are cold and radiant input is low, conditions which exist in the spring and fall seasons, especially early in the morning. Under these conditions, the darker basking insect is able to attain the critical equilibrium body temperature necessary for activity while the light-colored insect remains inactive. Likewise, basking larvae can feed longer at faster rates than non-basking larvae (Casey 1977), and dark morphs may be able to mature more quickly than light morphs (Fye 1979). Nezara v iridul a adults and nymphs bask (Waite 1980). Basking seems to be most prevalent in the early morning and late evening hours, and it lasts longer on cloudy, cool days (Waite 1980). In my personal observations of the behavior in adults, 5th, 4th, and occasionally late 3rd instars, I have noticed that the bug is usually positioned on the upper surface of a leaf such that the entire dorsal surface of the bug is perpendicular to the sun's rays. In addition, I have seen them many times centered near the base of the leaf, perhaps to minimize convective heat loss and maximize re-radiation from the leaf surface.

PAGE 114

104 In view of these observations, the results obtained in the laboratory basking experiment (Figures 5-3, 5-4) support the hypothesis that melanization in 4th instars can function to raise equilibrium body temperature during basking. The insects I used are among the smallest by weight for which temperature during basking has been measured (May 1976). Although Casey (1981) asserts that heating rate decreases and equilibrium body temperature increases with increasing size in insects, May's (1976) empirical study showed that for many small insects, heating rate is constant or slightly positive. In the range I tested, there were no size-based differences (Figure 5-3). The effect of coloration, however, was significant: black, unmelanized forms reached equilibrium body temperatures 1-1.4 higher than green, unmelanized forms in the same size range (Figure 5-4, Table 5-5). This temperature difference is similar to that obtained for different morphs of the salt-marsh caterpillar, in which black morphs are induced by coldtemperature rearings (Fye 1979). When the two morphs of this caterpillar were reared together with intermittent radiative input, the black morphs developed at a faster rate. Since Nezara nymphs do not fly or actively pursue prey, the thermal advantage of melanism, if any, probably relates to increased feeding rates and metabolic rates. Nezara reproduction extends into September and October. Nymphs are often present in November, but only adults can successfully overwinter (Todd and Herzog 1980). Therefore, as autumn nymphs are exposed to increasingly cooler weather, any adaptation which can speed development time is advantageous. At moderate temperatures and without radiative input, black Nezara nymphs develop at the same or slightly slower rate than green morphs

PAGE 115

105 (Chapter 2). However, the situation may be reversed if basking is allowed. During some preliminary basking behavior observations made in a cold room at 15 C, I noticed that 4th instar nymphs moved from shade into the rays from an incandescent light soon after it was turned on. Only after basking for a while, unmoving, did they begin to feed. One should expect that black nymphs could begin to feed earlier. However, feeding behavior, and possibly basking behavior, can vary depending on whether the nymph is newly molted or just about to molt again. Therefore, behavioral comparison experiments would have to be carefully designed to match for age as well as weight. A more definitive experiment would be one in which black and green morphs would be reared at a cold air temperature with limited periods of basking allowed, similar to Fye's (1979) work on the salt-marsh caterpillar. However, Fye did not provide a control in which the two color morphs were reared at the same air temperature with no basking; this would be essential to demonstrate that the thermal advantages of the dark pattern is responsible for faster development rates, rather than some other metabolic feature of black morphs. The final point to address is why the color pattern is so flexible with respect to temperature: why not remain black throughout all of the nymphal instars? Two reasons are suggested from the literature. At high temperatures, insects exhibit heat avoidance behavior. This usually includes the cessation of all activity and deliberate shade seeking (Heath 1967, Watt 1968, Sherman and Watt 1973, Casey 1976, Roland 1982). Light-colored insects are able to remain active into the hotter parts of the day (Edney 1971, Watt 1969, Roland 1982). Since the heat load could increase as Nezara nymphs grow larger (Casey 1981), heat stress may be

PAGE 116

106 an important factor for 4th and 5th instars during the hot days of summer. One would expect that black 4th or 5th instars would overheat easily upon exposure to direct sunlight and could be restricted in their feeding activity, while green morphs would be less susceptible. The second factor at work is the conspicuousness of the black morph against green foliage. Heinrich (1979) suggested that predator pressure is reponsible for selecting against basking behavior in caterpillars. Thus, even if being black during the summer is not thermally disadvantageous to the larger nymphs, it may be more advantageous to be an inconspicuous green.

PAGE 117

CHAPTER SIX CONCLUSION Control of Nezara Nymph P olymor phism This research has shown that the variation in Nezara virid ula nymph coloration does not fit into any simple polymorphic model (Chapter 1). Coloration changes with successive nymphal stages, yet variation also occurs within an instar (Chapter 2). Although color pattern variation is continuous, it does not fall in a normal distribution in the population (Chapter 2). There is very little, if any, difference in the biological aspects of green and black morphs (Chapter 2). Variation in color does not appear to be genetically based, yet parental influences can be measured (Chapter 3). Color variation within an instar is strongly affected by diet (Chapter 4) and temperature (Chapter 5). In addition, Kiritani (1964) found that rearing in crowded conditions increased the frequency of melanism, although the effect may have actually been due to food quality differences. This polymorphic system is not unique to Nezara and related species: at least one moth species, Saturnia pavonia, experiences very similar patterns of variability (Long 1953, Hintze-Podufal 1974 and 1977). In the absence of direct genetic control of color pattern expression (Chapter 3), variation may be the result of differential production of hormones. The juvenile hormone ( JK ) is a likely candidate 107

PAGE 118

108 because it reduces melanization in some caterpillars (Kiguchi 1972, Safranek and Riddiford 1975, Hintze-Podufal 1976), its secretion from the corpus allatum can be activated by feeding (de Wilde and de Loof 1973), and its activity and secretion rate can be reduced by cold temperature (Gilbert 1964). However, in Nezara JH would have to increase melanization and its secretion would have to increase with colder temperature. As an alternative suggestion, there may exist another hormone that can act to prevent melanization but does not begin its secretion until the JH titer has declined. Melanization at cold temperature can then be explained by inactivation of this other hormone. Hormone levels may vary enough because of environmental and maternal effects to account for the variation of color pattern that occurs at the stage of transition from black to green. The Adaptive Sig nifica nce of Nezara Nymph Polymorphism Why has this loosely controlled system of color determination survived the rigors of natural selection, which should favor the stable expression of one color form over the others? One may suggest that the slush in the system is selectively neutral. In other words, the actual timing of color change makes no difference in the fitness of a nymph, as long as it changes sometime between the 2nd instar and the adult stage. Nevertheless, such selectively neutral gene groups rarely occur (Ford 1945). The alternative is that this color variation system is adaptive, and that individuals with flexible expression of coloration are favored. I have shown that melanic 4th instar nymphs can gain heat at a faster rate than non-melanized nymphs (Chapter 5). As Hamilton (1973)

PAGE 119

109 suggested, black coloration in the younger stages may serve to increase thermal gain and, thereby, increase metabolic and growth rates. As larger size is reached, an insect becomes more conspicuous to visual predators (Endler 1978). Thus, a switch to green color in larger nymphs (and adults) would be advantageous at the point where thermal considerations and predator pressure balance out. If the air temperature is close to or above the insect's preferred body temperature, exposure to direct sunlight will cause overheating, and light-colored forms will be favored (Watt 1969). Therefore, at higher temperatures, light morphs should appear at earlier stages, both to avoid heat and to achieve crypticity. At low air temperatures, the retention of melanism may be essential if development is to proceed at all. This may be especially important in the fall, since only adult forms can successfully overwinter. Short photoperiod (which induces melanism in some butterflies, e.g., Douglas and Grula 1978) and poor diet may serve as preliminary cues to the approach of cold weather. One could again argue that the variation which occurs within an instar, even when all are reared under identical conditions, is coincidental and selectively neutral. However, this variation may function as a balanced polymorphism: rarer color forms may be overlooked by search-imaging predators. Without selection against this nongenetic variation, it would be retained. Areas of Futur e Research Research on Nezara nymph coloration should continue along three courses: physiological, behavioral, and ecological. Physiological research should pursue the question of the induction and maintenance of

PAGE 120

me 110 lanism. In particular, studies should examine possible roles of the juvenile hormone. Nymphs should be reared on food laced with JH to see if black morphs will be induced. Black and green morphs of the 4th and 5th instars should be analyzed for JH titers. Extirpation of the corpus allatum should be attempted on green and black 4th instar nymphs to see if black 5th instar nymphs can be induced. The possible effects of photoperiod on melanism should be examined in relation to corpus allatum activity Behavioral studies should concentrate on nymph basking, aggregation behaviors, and the relation of color to predation. Is it more than coincidental that color change occurs at the same stage where aggregation behavior ceases? Do black morphs feed earlier and longer than green morphs at low temperature? Can green morphs feed longer in direct sunlight when temperature is high? Predation experiments should be designed to test predator perception and recognition of the different morphs. These tests should also question the importance of the scent gland in defense against such predators. Ecological studies should examine the success of the different color forms in the field. Different morphs could be induced in the laboratory and released in separate field plots at different seasons. Periodic recaptures would then give evidence for the relative growth rates and survival of the morphs. Extensive field collections should be made in early summer and late fall to determine whether natural conditions shift the entire population towards melanism in the 4th instar. Field tests of different plant parts, species, and ages should be made to further elucidate the adaptiveness of the color response to diet. Finally, the nymph color patterns of other hemipterans in similar and different habitats should be studied and compared with Nezara.

PAGE 121

LITERATURE CITED Ayala, F. J., and C. A. Campbell. 1974. Frequency-dependent selection. Ann. Rev. Ecol. Syst. 5: 115-138. Baltensweiler, W. 1977. Color polymorphism and dynamics of larch bud moth populations Zeiraphera dinj.ana ( Lepidoptera : Tortricidae ) Mitt. Schweiz. Entomol. Ges. 50: 15-23. Bursell, E. 1970. An introduction to insect physiology. Academic Press, London and New York. Buschman, L. L. and W. H. Whitcomb. 1980. Parasites of Nezara viridula (Hemiptera: Pentatomidae ) and other Hemiptera in Florida. Florida Entomol. 63: 154-162. Cain, A. J., and P. M. Sheppard. 1950. Selection in the polymorphic land snail Cepaea nemoralis Heredity 4: 275-294. Casey, T. M. 1976. Activity patterns, body temperature, and thermal ecology in two desert caterpillars (Lepidoptera: Sphingidae). Ecology 57: 485-497. Casey, T. M. 1977. Physiological responses to temperature of caterpillars of a desert population of Manduca sext a (Lepidoptera: Sphingidae). Comp Biochem. Physiol. 57A: 53-58~! Casey, T. M. 1981. Behavioral mechanisms of thermoregulation. Pages 79-114 in B. Heinrich, editor. Insect thermoregulation. John Wiley and Sons, New York. Clarke, C. A., C. G. C. Dickson, and P. M. Sheppard. 1963. Larval color patterns in Papilio demodocus Evolution 17: 130-137. Corpuz, L. R. 1969. The biology, host range, and natural enemies of Nez ara viridula L. (Pentatomidae, Hemiptera). Philipp. Ent. 1: 225-239. de Ruiter, L. 1952. Some experiments on the camouflage of stick caterpillars. Behaviour 4: 222-232 de Wilde, J., and A. de Loof. 1973a. Reproduction. Pages 12-96 in M. Rockstein, editor. The physiology of insecta, 2nd ed. Academic Press, New York. Ill

PAGE 122

112 de Wilde, J., and A. de Loof. 1973. Reproduction-endocrine control. Pages 97-158 rn M. Rockstein, editor. The physiology of insecta, 2nd ed Academic Press, New York. DeWitt, N. B., and G. L. Godfrey. 1972. The literature of arthropods associated with soybeans. II. A bibliography of the southern green stink bug, Nezara viridula (Linnaeus) (Hemiptera: Pentatomidae) Illinois Natur. Hist. Surv. Biol. Notes 78: 1-23. Digby, P. S. B. 1955. Factors affecting the temperature excess of insects in sunshine. J. Exp. Biol. 32: 279-298. Dixon, A. F. G. 1972. Control and significance of seasonal development of colour forms in the sycamore aphid, Drepanosyphum platanoides (Schr.). J. Anim. Ecol. 41: 689-697. Douglas, M. K., and J. W. Grula. 1978. Thermoregulatory adaptations allowing ecological range expansion by the pierid butterfly Nathal is iole Boisduval. Evolution 32: 776-783. Drake, C. J. 1920. The southern green stink-bug in Florida. Quart. Bull. State Plant Board of Florida 4: 41-94. Edney E. B. 1971. The body temperature of Tenebrionid beetles in the Namib desert of Southern Africa. J. Exp. Biol. 55: 253-272. Endler, J. A. 1978. A predator's view of animal color patterns. Pages 319-364 in M. K. Hecht, W. C. Steere, and B. Wallace, editors. Evolutionary biology, vol. 11. Plenum Publishing Corp., New York. Ford, E. B. 1945. Polymorphism. Biol. Rev. 20: 73-88. Fye R. E. 1979. Salt marsh caterpillars: temperature control of melanism in larvae. Sci. Educ Adm. Agric. Res. Results ARR-W (4): 1-11. Genung, W. G. and V. E. Green, Jr. 1974. Food habits of the meadowlark in the Everglades in relation to agriculture. Environ. Entomol. 3: 39-42. Gilbert, L. I. 1964. Physiology of growth and development: endocrine aspects. Pages 150-226 in M. Rockstein, editor. The physiology of insecta. Academic Press, New York. Gilbert, L. I., and D. S. King. 1973. Physiology of growth and development: endocrine aspects. Pages 250-379 rn M. Rockstein, editor. The physiology of insecta, 2nd ed. Academic Press, New York. Gillett, S. D. 1978. Environmental determinants of phase polymorphism of the desert locust Schistocerc a gjegarij reared crowded. Acrida 7: 267-288.

PAGE 123

n: Golberg, L. and B. De Meillon. 1948. The nutrition of the larva of Aedes aegypti Linnaeus. 4. Protein and amino acid requirements. Biochem. J. 43: 379-387. Goodwin, T. W. 1949. Carotenoid distribution in solitary and gregarious phases of the African migratory locust and the desert locust. Biochem. J. 45: 472-479. Hailman, J. P. 1977. Optical signals: animal communication and light. Indiana Univ. Press, Bloomington, Indiana. Hamilton, W. J. III. 1973. Life's Color Code. McGraw-Hill, New York. Harris, V. E., and J. W. Todd. 1981. Rearing the southern green stink bug Nezara virid ula with relevant aspects of its biology. J. Georgia Entomol. Soc 16: 203-211. Heal, J. 1979. Color patterns of Syrphidae. 2. Eristalis intricarius Heredity 43: 229-238. Heath, J. E. 1967. Temperature response of the periodical "17-year" cicada, Magicicada cas sini Amer. Midland Natur. 77: 64-76. Heinrich, B. 1979. Foraging strategies of caterpillars: Leaf damage and possible predator avoidance strategies. Oecologia 42: 325-337. Hill, L. and H. J. Taylor. 1933. Locusts in sunlight. Nature 132: 276. Hintze-Podufal, C. 1974. Untersuchungen zur Farbmusterbildung bei den Larven von Eudia pavonia L. (Lepidoptera : Saturni idae ) Biol. Zbl. 93: 545-559. Hintze-Podufal, C. 1976. Juvenile hormone and analogs in the food plant of Notodontidae and their effect on post embryonic development. Z. angew. Entomol. 82: 177-186. Hintze-Podufal, C. 1977. The larval melanin pattern in the moth Eudia pavonia and its initiating factors. J. Insect Phvsiol. 23: 731-738. Hoffman, R. J. 1978. Environmental uncertainty and evolution of physiological adaptation in Colias butterflies. Am. Nat. 112: 999-1015. Houston, K. J., and D. F. Hales. 1980. Allelic frequencies and inheritance of color pattern in Coelophora inaequalis (F.) (Coleoptera: Coccinel lidae) Austr. J. Zoo IT" 281^669-678 Ishii, M., and T. Hidaka. 1979. Seasonal polymorphism of the adult rice plant skipper Parnara gutt ata guttata (Lepidoptera: Hesperiidae) and its control. Appl. Ent. Zool. 14: 173-184.

PAGE 124

114 Jensen, R. L. and J. Gibbens. 1973. Rearing the southern green stink bug on an artificial diet. J. Econ. Entomol. 66: 269-271. Jones, J. S., B. H. Leith, and P. Rawlings. 1977. Polymorphism in Cepaea : a problem with too many solutions? Ann. Rev. Ecol. Syst. 8: 109-143. Jones, T. H. 1918. The southern green plant bug. USDA Bulletin 689: 1-27. Kariya, H. 1961. Effect of temperature on the development and the mortality of the southern green stink bug, Nezara viridula and the oriental green stink bug, N. antennata (in Japanese, English summary). Jap. J. Appl. Entomol. Zool. 5: 191-196. Kettlewell, H. B. D. 1973. The evolution of melanism, the study of a recurring necessity. Clarendon Press, Oxford. Kidd, K. A. C. 1979. The control of seasonal changes in the pigmentation of lime aphid nymphs, Eucal lipteru s til iae Entomol. Exp. Appl. 25: 31-38. ~~ ""' Kiguchi, K. 1972. Hormonal control of the coloration of larval body and the pigmentation of larval markings in Bombyx mori. Part 1: Endocrine organs affecting the coloration of larval body and the pigmentation of markings. J. Seric. Sci. Jpn. 41: 407-412. (English abstract.) Kiguchi, K., and S. Kimura. 1981. Hormonal control of larval coloration in the silkworm Bombyx mori: Changes in tyrosine contents during larval molting. J. Seric. Sci. Jpn. 50: 435-443. (English abstract.) Kiritani, K. 1964. The effect of colony size upon the survival of larvae of the southern green stink bug, Nezara viridula. Jap. J. Appl. Entomol. Zool. 8: 45-54. ~ Kiritani, K. N. Hokyo, and K. Kimura. 1962. Differential winter mortality relative to sex in the population of the southern green stink bug, Nezara viri dula (Pentatomidae Hemiptera). Jap. J. Appl. Entomol. Zool. 6: 242-246. Kiritani, K., N. Hokyo, and K. Kimura. 1963. Survival rate and reproduct ivity of the adult southern green stink bug, Nezara viridula in the field cage. Jap. J. Appl. Entomol. Zool. 7: 113-124. Kiritani, K. and K. Kimura. 1965. The effect of population density during nymphal and adult stages on the fecundity and other reproductive performances. Jap. J. Ecol. 15: 233-236.

PAGE 125

115 Kiritani, K. and K. Kimura. 1967. Effects of parental age on the life cycle of the southern green stink bug, Nezara viridula L. (Heteroptera: Pentatomidae ) Appl. Entomol. Zool. 2: 69-78. Knight, H. H. 1924. On the nature of the color patterns in Heteroptera with data on the effects produced by temperature and humidity. Ann. Ent. Soc. Amer. 17: 258-272. Kobayashi, T. 1959. The developmental stages of some species of the Japanese Pentatomidae (Hemiptera). VII. Developmental stages of Nezara and its allied genera (Pentatomidae s. str.) (in Japanese, English summary). Jap. J. Appl. Entomol. Zool. 3: 221-231. Long, D. B. 1953. Effects of population density on larvae of Lepidoptera. Trans. R. Ent. Soc. Lond 104: 544-585. Majerus, M. E. N. 1981. The inheritance and maintenance of the melanic form nigrescens of Pachycnemia hrpppcas_tanaria f. nigrescens (Lepidoptera: Ennominae). Ecol. Entomol. 6: 417-422. May, M. L. 1976. Warming rates as a function of body size in periodic endotherms. J. Comp. Physiol. Ill: 55-70. McLain, D. K. 1981. Female choice and the adaptive significance of prolonged copulation in Nezara viridula (Hemiptera: Pentatomidae). Psyche 87: 325-336. Meyer, A. 1979. Color polymorphism in the grasshopper Paulinia acuminata Entomol. Exp. Appl. 25: 21-30. Mitchell, W. C. and R. F. L. Mau. 1969. Sexual activity and longevity of the southern green stink bug, Nez ara vi ridula Ann. Ent. Soc. Amer. 62: 1246-1247. Myers, A. A. 1977. The effect of temperature during pupal actiphase on imaginal coloration in Mythimna loreyi (Lepidoptera: Noctuidae) Entomol. Gaz. 28: 75-79. Neville, A. C. 1975. Biology of arthropod cuticle. Springer, Berlin. Panizzi, A. R. M. H. M. Galileo, H. A. 0. Gastal, J. F. F. Toledo, and C. H. Wild. 1980. Dispersal of Nezara viridula and Peizodorus guildinii nymphs in soybeans. Env Ent. 9: 293-297. Owen, D. F. and R. G. Wiegert. 1962. Balanced polymorphism in the meadow spittlebug, Philaenus spumarius Am. Natur. 96: 353-359. Pekkarinen, A. 1979. Morphometric color and enzyme variation in bumble bees ( Hymenoptera, Apidae, Bombus ) in Fenncscandia and Denmark. Acta Zool. Fenn. 158: 1-60.

PAGE 126

116 Rex, M. A. 1972. The relationship of island area and isolation to color polymorphism in Liguus f asci atus (Pulmonata: Bulimulidae) Breviora (Mus. Comp. Zool., Cambridge, Mass) 391: 1-15. Roland, J. 1982. Melanism and diel activity of alpine Colias (Lepidoptera : Pieridae). Oecologia 53: 214-221. Rowell, C. H. F. 1971. The variable coloration of the Acridoid grasshoppers. Adv. Insect Physiol. 8: 145-198. Safranek, L. and L. M. Riddiford. 1975. The biology of the black larval mutant of the tobacco hornworm, Manduca sext a J. Insect Physiol. 21: 1931-1938. Sasakawa, M. 1967. Phase polymorphism of the larger pellucid hawk moth, Cephomodes hy las L. Sphingidae. Effect of population density on the larval coloration and development. Sci. Rep. Kyoto Pref. Univ. Agric. 19: 29-36. Schneider, C. 1973. Uber den Einfluss verscheidener Umweltf aktoren auf den Farbungspolyphanismus der Raupen des tropisch-amerikanischen Schwarmers Er innyis el lo L. (Lepidopt., Sphingidae). Oecologia (Berl.) 11: 351-370. Sheppard, P. M. 1962. Some aspects of the geography, genetics, and taxonomy of a butterfly. Pages 135-152 _i_n D. Nichols, editor. Taxonomy and geography. Systematic Association, London. Sherman, P. W. and W. B. Watt. 1973. The thermal ecology of some Colias butterfly larvae. J. Comp. Physiol. 83: 25-40. Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill, New York. Singh, Z. 1973. Southern green stink bug and its relationship to soybeans: bionomics of the southern green stink bug Nezara viridula Linn. (Hemiptera: Pentatomidae ) in Central India. Metropolitan Book Co., Delhi, India. Smith, A. G. 1978. Environmental factors influencing pupal color determination in Lepidoptera Part 1. Experiments with Papil io polytes Pap il io demoleus, and Papilio polyxenes Proc R. Soc Lond. B. Biol. Sci. 200~: 295-330. Sokal, R. R. and F. J. Rohlf. 1973. Introduction to bios tat is tics W. H. Freeman, San Francisco. Souci, S. W. 1981. Food composition and nutrition tables 1981/82. Wissenschaf tliche Ver lagsgesel lschaf t Stuttgart. Stam, P. A. 1978. Relation of predators to population dynamics of Nezara viridula (L.) in a soybean ecosystem. Dissertation. Louisiana State University, Baton Rouge, Louisiana.

PAGE 127

117 Strickberger, M. W. 1968. Genetics. Macmillan Company, New York. Todd, J. W., and D. C. Herzog. 1980. Sampling phytophagous Pentatomidae on soybean. Pages 438-478 ^n M. Kogan and D. C. Herzog, editors. Sampling methods in soybean entomology. Springer-Verlag, New York. Waite, G. K. 1980. The basking behavior of Nezara viridu la (L.) (Pentatomidae: Hemiptera) on soybeans and its implication in control. J. Austr. Entomol. Soc. 19: 157-159. Watt, W. B. 1968. Adaptive significance of pigment polymorphisms in Colias butterflies. I. Variation of melanin pigment in relation to thermoregulation. Evolution 22: 437-458. Watt, W. B. 1969. Adaptive significance of pigment polymorphisms in Colias butterflies. II. Thermoregulation and photoperiodical ly controlled melanin variation of C olia s eurytheme Proc Nat. Acad. Sci. (U.S.) 63: 767-74. Wellington, W. G. 1965. Some maternal influences on progeny quality in the western tent caterpillar, Malacosoma pl uvia le (Dyar). Can. Ent. 97: 1-14. " Wigglesworth, V. B. 1954. The physiology ofinsect metamorphosis. Cambridge Univ. Press, Cambridge. Wigglesworth, V. B. 1959. Metamorphosis and differentiation. Sci. Amer. 200: 100-102, 104, 106, 108, 110. Wigglesworth, V. B. 1965. The principles of insect physiology. Methuen and Co., London. Williams, C. M. 1961. The juvenile hormone. II. Its role in the endocrine control of molting, pupation, and adult development in the cecropia silkworm. Biol. Bull. 121: 572-585. Wylie, H. G. 1980. Color variability among females of Micronotus vitta tae (Hymenoptera : Braconidae). Can. Entomol. 112: 771-774. Yukawa, J., and K. Kiritani. 1965. Polymorphism in the southern green stink bug. Pac. Insects 7:639-642.

PAGE 128

BIOGRAPHICAL SKETCH Carol Brown Johnson was born Carol Ann Brown on July 7, 1953, in Waco, Texas. During her childhood years in the central Texas area, she spent many hours in creeks and fields. Her parents, Bryce and Lillian Brown, are biologists with a love for natural history and geology. Therefore, family vacations to Mexico, Central America, and Colorado became educational adventures for Carol. After graduating valedictorian from University High School in Waco, she attended Baylor University. Majoring in biology, she received her Bachelor of Science degree summa cum laude, honors program with distinction, in August 1975. She then began the graduate program in zoology at the University of Florida, where she gained experience teaching laboratory classes in physiology, embryology, invertebrate zoology, and general biology. She completed a Master of Science degree in August 1978 with a thesis on the Florida Tree Snails of the Everglades National Park. While a Ph.D. candidate, she taught lecture courses in general biology and wrote laboratory procedures for introductory biology courses. On May 8, 1982, she married Douglas A. Johnson. J 18

PAGE 129

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Thomas C. Emmel, Chairman Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Carmine A. Lanciani Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Pauline 0. Lawrence Associate Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Reece I. Sailer Graduate Research Professor of Entomology and Nematology This dissertation was submitted to the Graduate Faculty of the Department of Zoology in the College of Liberal Arts and Sciences and tc the Graduate School, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1984 Dean for Graduate Studies and Research

PAGE 130

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ..-.e.eji Thdrtias C. Emmel, Chairman Professor of Zoology l/Vfrv*-?e<7 I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. L&miw* U g\&w„. vk^&aai Carmine A. Lanciani Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Cjw^AlL U 'jj pUd^jUT^^ Pauline 0. Lawrence Associate Professor of Zoology I certify that I have read this study and that in my opinion— it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ^r t\wa>tM Reece I. Sailer Graduate Research Professor of Entomology and Nematology This dissertation was submitted to the Graduate Faculty of the Department of Zoology in the College of Liberal Arts and Sciences and to the Graduate School, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1984 Dean for Graduate Studies and Research

PAGE 131

UNIVERSITY OF FLORIDA 3 1262 08285 325 9