Citation
Parasitization of Liriomyza trifolii (Burgess) by Diglyphus intermedius (Girault)

Material Information

Title:
Parasitization of Liriomyza trifolii (Burgess) by Diglyphus intermedius (Girault)
Creator:
Patel, Kirtikumar Jashbhai, 1951-
Publication Date:
Language:
English
Physical Description:
viii, 124 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Eggs ( jstor )
Female animals ( jstor )
Larvae ( jstor )
Leafminers ( jstor )
Mortality ( jstor )
Oviposition ( jstor )
Parasite hosts ( jstor )
Parasitism ( jstor )
Parasitoids ( jstor )
Tomatoes ( jstor )
Agromyzidae -- Biological control ( lcsh )
Dissertations, Academic -- Entomology and Nematology -- UF
Entomology and Nematology thesis Ph.D
Leafminers ( lcsh )
Tomatoes -- Diseases and pests ( lcsh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1987.
Bibliography:
Includes bibliographical references (leaves 110-123).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Kirtikumar Jashbhai Patel.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
030330403 ( ALEPH )
16694914 ( OCLC )

Downloads

This item has the following downloads:


Full Text









PARASITIZATION OF Liriomyza trifolii (Burgess) BY
Diglyphus intermedius (Girault)














By
KIRTIKUMAR JASHBHAI PATEL


















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


1987














ACKNOWLEDGEMENTS

My sincerest thanks to Dr. David J. Schuster, chairman of the

supervisory committee for his financial support, guidance and

advice.

I express my gratitude for the help given to me by Dr. Smerage

and Dr. Kerr while they served on the supervisory committee. I am

indebted to the late Dr. Sailer who had also served on the

supervisory committee and for motivating me in my career.

I would like to thank Dr. Portier and his graduate research

assistants for making helpful suggestions for the statistical

analysis of the experimental results.

The staff at the Gulf Coast Research Station helped me in

many ways while I was at the station doing the experiments. Their

help is gratefully acknowleged.

My colleagues and friends provided me with companionship and

moral support during my graduate years. Their friendship will

always be cherished.

My family provided me with financial and moral support. I am

grateful for the help I have received from them.

Finally, my wife Falguni has been patient with me while I was

busy with my studies and I wish to thank her for that.










ii












TABLE OF CONTENTS


Page

ACKNOWLEDGEMENTS ....................................... ii

LIST OF TABLES ...... ................................... v

LIST OF FIGURES.......................................... vi

ABSTRACT............................................. vii

CHAPTER

I INTRODUCTION.................................. 1

II LITERATURE REVIEW............................. 13

Tomato Production in Florida.................. 13
Importance ............................. 13
Production Methods Relevant to Pest
Management .... ............................. 14
Liriomyza trifolii (Burgess)................. 22
Description ................................. 22
Taxonomy.................................... 24
Distribution........ ........................ 24
Host Plant Range............................ 25
Life History ............................... 34
Chemical Control of Leafminers on Tomato.... 39
Biological Control of Leafminers on Tomato.. 41
Diglyphus intermedius (Girault)............... 47
Taxonomic Status, Hosts and Distribution.... 47
Adult External Morphology................... 47
Parasitoid Abundance in Florida............. 48
Biology and Life History.................. 48
Modeling Leafminer-Parasitoid System.......... 50
Introduction.............................. 50
The Smerage et al. Leafminer-Parasitoid
Model.......* .. *........................ ... 51

III PLANT PRODUCTION AND REARING TECHNIQUES FOR
Liriomyza trifolii (Burgess) AND Diglyphus
intermedius (Girault) ON TOMATO................ 57

Introduction............................... 57
Host PLant Production........................ 57
Rearing L. trifolii........................ 59
Rearing D. intermedius........................... 62




iii












Page
IV FECUNDITY AND LONGEVITY OF Diglyphus
intermedius (Girault) AND ITS CONTRIBUTION TO
HOST MORTALITY ............................... 64

Introduction ................................ 64
Materials and Methods....................... 65
Results and Discussion................ .... 66

V DIURNAL PATTERN IN OVIPOSITION AND HOST
MORTALITY.. .................... ............... 77

Introduction .................... .... ..... .. 77
Materials and Methods....................... 78
Results and Discussion....................... 79

VI HOST STAGE PREFERENCE.......................... 82

Introduction ............................. .. 82
Materials and Methods...................... 83
Results and Discussion....................... 84

VII IMPACT OF ALTERING HOST AND PARASITOID DENSITY
ON DAILY PARASITISM RATE...................... 87

Introduction............................ .... 87
Materials and Methods...................... 87
Results and Discussion...................... 91

VIII MULTIPLE OVIPOSITION AND ITS INFLUENCE ON
PROGENY SURVIVAL RATE........................ 98

Introduction........................ ...... 98
Materials and Methods.................... .... 98
Results and Discussion....................... 99

IX CONCLUSION..................................... 101

LITERATURE CITED...................... ............... 110

BIOGRAPHICAL SKETCH .. ............................... 124












iv














LIST OF TABLES


Page


Table 2-1. Current known distribution of L.
trifolii (Burgess) ........ ...... ........ ...... 26

Table 2-2. Host plant range of L. trifolii.................. 28

Table 2-3. Developmental times for L. trifolii
immature stages on some crop plants.............. 37

Table 4-1. Number of L. trifolii larvae killed
by, and the fecundity and longevity of
D. intermedius at different constant
temperatures (+ 1SD)..... ........... .... ..... ... 66

Table 5-1. Relationship between time of day and host
mortality and oviposition of 5-day old
female D. intermedius at 25-27 C (+ 1SD)......... 79

Table 6-1. Relationship between L. trifolii larva
age and mean no. of eggs laid and hosts
killed by D. intermedius (+ 1SD)................. 84

Table 7-1. Proportion of observations with 100%
parasitization of killed hosts at different
leafminer densities...... ....................... 95

Table 7-2. Frequency distribution of killed L.
trifolii larvae in classes of different
D. intermedius egg densities..................... 96

Table 8-1. Survival of D. intermedius eggs to adult
stage when placed at different densities on
L. trifolii larvae............................. 100

Table 9-1. Predicted number of parasitoid eggs
reaching the adult stage for every 100 hosts
killed ............... .......................... 106








v















LIST OF FIGURES


Page


Figure 1-1. Typical functional description of a single
host-single parasitoid parasitism................ 4

Figure 2-1. Conceptual model of the leafminer-parasitoid
system from Smerage et al. 1980.................. 52

Figure 4-1. Relationship between D. intermedius
fecundity, F, and temperature, T................ 68

Figure 4-2. Relationship between D. intermedius
longevity, L, and temperature, T................. 69

Figure 4-3. Relationship between # hosts killed, Hm,
and temperature, T.. ............................ 70

Figure 4-4. Mean daily D. intermedius-induced
host mortality rates and D. intermedius
oviposition at different temperatures........... 72

Figure 4-5. Influence of temperature on the number
of eggs laid daily by surviving D.
intermedius females.............................. 74

Figure 4-6. Influence of temperature on the number
of hosts killed by surviving D.
intermedius females............................ 75

Figure 7-1. Relationship between host density, nl,
and parasitism rates at different
parasitoid densities, n2 ........................ 93

Figure 7-2. Relationship between host density, n,
and parasitoid-induced host mortality rate
at different parasitoid densities, n2.' ......... 94











vi














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


PARASITIZATION OF Liriomyza trifolii (Burgess) BY
Diglyphus intermedius (Girault)

By

Kirtikumar Jashbhai Patel

May 1987

Chairman: D. J. Schuster
Major Department: Entomology and Nematology

Diglyphus intermedius (Girault) is a commonly encountered

parasitoid of Liriomyza species leafminers which are considered

important pests of tomato and several other commercially grown

crops in Florida. An understanding of the leafminer-parasitoid

interactions is important to any sound, integrated pest management

system on tomato. Those aspects of the biology of Diglyphus

intermedius that influence its parasitization of Liriomyza

trifolii (Burgess), reared on tomato, were determined from

laboratory experiments to obtain information on L. trifolii-D.

intermedius interactions.

The relationship between T, temperature in degrees Celsius, and

parasitoid fecundity, F, longevity, L, and parasitoid-induced host

mortality, Hm, were measured at 15.6, 19.4, 23.3, 27.2, and 31.1 C

and are described by the equations: F = -196.11 + 42.65T 1.1T2

(r = .65), L = 74.32 2.09T (r2 = .94), and Hm = 721.97 -

19.1T (r2 = .83)


vii












The daily parasitism rate was determined to be a function of

leafminer and parasitoid densities and is expressed by the equation:

Y = Kp.n2.(1 -exp(-n1/Kh) where Y is the number of hosts parasitized/

day, Kp is 7.3908 hosts parasitized/parasitoid-day and Kh is 0.0144

leafminer larvae/sq cm leaf area. Leafminer density, nl, ranged

from 0 to 0.06 larvae/sq cm leaf area. Parasitoid density, n2,

ranged from 1 to 5/cage.

The rate of daily host mortality is described by the equation

Z = Cp.n2.(1 -exp(-n2/Ch))where Z is the number of hosts killed/

day, Cp is 9.2064 hosts killed/parasitoid-day and Ch is 0.0165 leaf-

miner larvae/sq cm leaf area.

D. intermedius-induced host mortality and parasitization

were highest in the first 4 h of a 24 h period, with 12 h of light

followed by 12 h of dark. More 120-h-old larvae were parasitized

and killed than were 96-h-old larvae when both ages of larvae were

equally available to a female parasitoid. An average 0.8, 1.1, 1.35

and 0.98 parasitoid eggs became adults when 1, 2, 3, and 4 eggs were

placed on each host larva, respectively.

















viii













CHAPTER I
INTRODUCTION


The use of broad-spectrum insecticides on a large scale in the

1940s has led to the emergence of Liriomyza species leafminers

as important pests of tomato, chrysanthemum, celery and many other

commercial crops grown in Florida (Schuster 1981). The worldwide

spread of Liriomyza trifolii (Burgess) in the past decade, in

addition, made it an international pest (Poe and Montz 1981a). The

balance between leafminers and their natural enemies has presumably

been upset by the use of broad-spectrum insecticides under

Florida conditions. New or previously unused insecticides have

been introduced whenever contemporary ones became ineffective

(Leibee 1981b), but these have provided only short term relief.

The period of effective use of a newly introduced, broad-spectrum

insecticide has usually been followed by loss of control. The

exact mechanisms of the breakdown in control are not known.

Damage due to insecticide-induced outbreaks of leafminers can

be reduced greatly if the insecticide is selected carefully along

with population monitoring and the use of economic thresholds

(Waddill 1981). Only recently have host specific insecticides

become available for leafminers (Robb and Parrella 1984). Some of

these chemicals are still awaiting clearance for commercial use on

tomato. (Fluker personal communication 1986). As a result of the

failure of long term control of the pest with insecticides,




1









2
growers have embraced the current philosophy of integrated pest

management (IPM) approach (Poe 1985).

Action thresholds have been established for leafminers on

tomato in Florida (Pohronezny and Waddill 1978, Schuster et al.

1980). The role that the natural enemies play in determining host

population densities has subsequently become crucial to any IPM

strategy for leafminer regulation. For example, only live

leafminer larvae are counted in determining if leafminers have

reached the threshold level that requires insecticide application.

The role of natural enemies is recognized and taken into

consideration by excluding larvae killed by parasitoids. As a

further refinement of IPM tactics, many current research programs

are directed toward determining the role of natural enemies in

regulating leafminer populations to make IPM programs more robust

(Parrella et al. 1985, Schuster 1985, Zehnder and Trumble 1985, and

Lindquist and Casey 1985).

Leafminer-parasitoid interaction was a focus of the leafminer

population dynamics model for celery proposed by Smerage et al.

(1980) and provides an excellent framework for the elucidation of

the role of natural enemies in regulating leafminer populations.

Their model pertained to within-field populations of eggs, larvae,

pupae and adults of leafminers and its parasitoids, broadly

expressed as processes that contribute to the overall dynamics of

the model. Other processes, such as parasitism, were also

incorporated into the model.









3
One hypothesis about the rate of parasitism is that it is a

function of host and parasitoid densities. Holling (1959), for

example stated that parasitism rate is directly dependent on host

and parasitoid densities. An increase in either density led to

higher parasitism rate. This hypothesis motivated Smerage et al.

(1980) to utilize the family of curves depicted in Figure 1-1 to

illustrate the general relationship between the host and parasitoid

densities and the number of hosts parasitized/day. The scale of

the family of curves was assumed by Smerage et al. (1980) to

fluctuate with temperature.

Musgrave et al. (1980) made some computer simulations of the

Smerage et al. (1980) model. Process and parameter descriptions

were derived from hypothetical values and preliminary estimates.

The results of the simulations were encouraging enough to warrant

further research to obtain more complete data to improve the model.

Several workers in Florida are striving for more complete

quantitative descriptions of the biologies of leafminers and its

parasitoids. This includes research in better estimates of

parameters. The ultimate aim of this research is to incorporate a

leafminer model into IPM strategies. As results become available,

they may be incorporated into the Smerage et al. (1980) model. An

alternative model may become necessary if the new information can

not be readily incorporated into the existing model without major

structural changes.








4














4Kp -
n2=4




S3Kp -
4"- n2 = 3



S2K -
P
S"2



K p i n 21





0 KH 2KH 3KH 4KH 5KH



Host Density

Figure 1-1. Typical functional description of a single host-
single parasitoid parasitism (figure obtained from
Smerage 1980).









5
The specific needs being addressed currently are 1) complete

quantitative information about the biologies of leafminers and their

most common parasitoids and 2) sampling procedures to estimate

populations of the leafminers and its major parasitoids. A major

thrust of research efforts has been the measurement of fecundity,

longevity and development of leafminers and its parasitoids on

different crop and weed hosts and at different temperatures.

Behavioral phenomena, such as peak periods of ovipositional and

feeding activities, have been reported to follow circadian

patterns. These are also being studied for the leafminer and its

parasitoids.

Another focal point of leafminer research is leafminer-

parasitoid interactions. In particular, the extent to which

leafminer and parasitoid densities determine the rate of leafminer

parasitism is being investigated. Parasitization of a host refers

to the act of utilizing a host for oviposition. Parasitoids may

also kill hosts without ovipositing on them. Therefore, the effect

of host and parasitoid density on the rate of parasitoid-induced

host mortality is also being investigated. The term parasitoid-

induced host mortality is from hereon used to refer to the sum of

hosts killed by parasitoids for ovipositional and non-ovipositional

purposes. In other words, it is the number of successful attacks

by parasitoids that lead to death of hosts.

More than one egg of a parasitoid may be deposited when a host

is parasitized. Multiple oviposition may lead to reduced

survivorship of the progeny of parasitoids. In addition, not all












host larvae may be equally acceptable to parasitoids for

ovipositional or non-ovipositional purposes. Therefore, there are

current investigations of multiple oviposition and parasitoid

preference for a particular size of leafminer larvae.

The objectives of the research presented here were


1) To determine the rates of parasitism and

parasitoid-induced host mortality as functions of leafminer and

parasitoid densities, with temperature being kept constant.


2) To determine the effect of temperature on parasitoid

fecundity, female parasitoid longevity and parasitoid-induced host

mortality.


3) To determine the role of time of day on parasitoid

oviposition and parasitoid-induced host mortality rates during

different times within a day.


4) To determine the influence of host age on parasitoid

oviposition and parasitoid-induced host mortality.


5) To determine the influence of multiple oviposition by the

parasitoid on the survivorship of the parasitoid progeny.


In the Smerage et al. model (1980), all parasitoid species

were aggregated into a single equivalent parasitoid population to

simplify the model. There are many parasitoids that attack

Liriomyza trifolii, however. In addition, the biologies of









7

these different species vary. For the research presented here it

was possible to investigate only one parasitoid species,

Diglyphus intermedius (Girault), with the resources and time

available.

D. intermedius was chosen as a representative natural

enemy of the leafminers for two reasons. D. intermedius is one

of the three most common parasitoids reared from leafminers on

tomato in Florida (the other two are Chrysonotomyia sp.

(Eulophidae) and Opius sp. (Braconidae)) (Schuster 1985). D.

intermedius is an ectoparasitoid, laying its eggs on or near the

host larva (Hendrickson and Barth 1978). This makes D.

intermedius easier to observe than either Chrysonotomyia sp.

or Opius sp., which are endoparasitoids (Lema 1976).

Before initiating any experimental work, the literature on

leafminer and D. intermedius was reviewed (see Chapter II).

To study the biology of the parasitoid and its interaction with

leafminers, experiments should ideally be done in the field. It is

not possible to vary a single factor and keep all others constant

in the field, however. All experiments were, therefore, done in

the laboratory. L. trifolii and D. intermedius were initially

collected from tomato fields and subsequently reared on tomato in

controlled environmental conditions to ensure an adequate supply for

use in the experiments. The insect rearing techniques are

described in Chapter III.

The first series of replicated experiments determined the

influence of temperature on D. intermedius fecundity, longevity









8


and D. intermedius-induced host mortality (Chapter IV).

Hendrickson and Barth (1978) had shown that D. intermedius

usually oviposited on 3rd-instar hosts although it killed all three

instars. Leafminer larvae in their 3rd-instar were kept in petri

dishes and placed in 5 incubators each set at a different constant

temperature in the 15.6 to 31.1 C range. A pair of parasitoids was

introduced into each petri dish. Killed hosts and parasitized

hosts were counted daily. Leafminer larvae from the previous day

were replaced with more 3rd-instar larvae. The petri plates were

maintained until the female parasitoids died.

In the temperature range studied, the temperature dependence

of F, the average fecundity per adult female lifespan, was

described by the equation:


F = -196.11 + 42.65T 1.1T2 (r2 = .65) (1)


When the number of hosts was a non-limiting resource, the

temperature dependence of Hm, the average number of hosts

killed/adult female lifespan, was described by the equation:


Hm = 721.97 19.1T (r2 = .83) (2)


Finally, the temperature dependence of L, the average female

parasitoid longevity in days, was described by the equation:


L = 74.32 2.09T (r2 = .94) (3)









9

In these and all subsequent equations, T denotes temperature in

degrees Celsius. As a female parasitoid aged, daily parasitism

and parasitoid-induced host mortality rates declined. Temperature

determined the adult female parasitoid age at which oviposition and

parasitoid-induced host mortality rates peak. With an increase in

temperature, these peaks are reached earlier.

Experiments to determine the effect of time of day on

parasitoid-induced host mortality and parasitization were done in

a controlled environment room (Chapter V). The room was maintained

at 25-27 C. Fluorescent lighting was turned on at 0800 h and off

at 2000 h. A 5-day-old female parasitoid female was introduced

into a petri dish containing 20 3rd-instar host larvae. Every 4

hours, for the first 12 h of a 24 h day, host larvae were replaced

with 20 more 3rd-instar larvae. The numbers of hosts killed and

parasitized were counted for each time period.

D. intermedius-induced host mortality and parasitization

were highest between 0800 and 1200 h. There was an almost complete

absence of parasitism and parasitoid-induced host mortality in the

absence of light. The data obtained in this experiment could be

utilized in determining, for example, when to spray during the day

to produce the least effect on parasitoid densities.

Whether D. intermedius preferred 3rd-instar larvae or 2nd-

instar larvae for oviposition was also determined (Chapter VI). An

equal number of 2nd- and 3rd-instar host larvae were made available

to females singly confined to petri dishes. Significantly more 3rd-

instar larvae were parasitized and killed than were 2nd-instar larvae.









10



Experiments done to determine rates of parasitism and

parasitoid-induced host mortality, as functions of host and

parasitoid densities, are reported in Chapter VII. Female D.

intermedius, along with tomato plants containing 3rd-instar

leafminer larvae, were confined to cages in a controlled

environment room. The temperature in the room was maintained

within the 25-27 C range. Each cage contained 1, 2, 3, 4, or 5

4-day-old female parasitoids. The density of leafminer larvae was

altered by exposing clean tomato plants to different densities of

leafminer adults for different time intervals. At the end of 12 h

of exposure to parasitoids, plants were removed, leaf area was

measured, and the numbers of hosts parasitized and killed were

recorded.

The data obtained from the experiments were utilized to obtain

estimates of parameters permitting description of the relationship

between the rate of parasitism and the densities of L. trifolii

and D. intermedius by the equation:


Y = Kp.n2(1 -exp(-n1/Kh) (r2 = .79) (T = 25-27C) (4)


where Y, the parasitism rate, is the number of hosts parasitized

per day. In (4), Kp, with a value of 7.3908, is a constant with

the units number of hosts parasitized/parasitoid-day. Densities

n2 and n, are, respectively, parasitoids/cage and leafminers/









11

leaf area (cm2) and Kh, another constant, is 0.0144 leafminer

hosts/leaf area.

Results of the experiments also were utilized to formulate

the similar but parametrically different equation (5) below that

expresses parasitoid-induced host mortality as a function of host

and parasitoid densities at 25-27C:


Z = Cp.n2(l -exp(-n1/Ch) (r2 = .78) (T = 25-27) (5)


where Z, the parasitoid-induced host mortality rate, is the number

of hosts killed per day. With a value of 9.2064, Cp is a constant

with the units number of hosts killed/parasitoid-day and Ch is

0.0165 leafminer hosts/leaf area.

The effect of multiple oviposition on survival of progeny is

important in estimating the parasitoid density in the following

generation. To study this effect, parasitoid eggs were kept at 4

different densities on host larvae in the experiment reported in

Chapter VIII. An average 0.8, 1.1, 1.35 and 0.98 eggs became

adults when 1, 2, 3, and 4 eggs, respectively, were kept on each

host larva.

The results of all experiments reported in this dissertation

are summarized in Chapter IX. In this final chapter, I have also

hypothetized that, Ke, the number of parasitoid eggs reaching the

adult stage/100 host larvae killed may be described by the

equation:


Ke = 93 -800nI (6).









12

In (6), n1 is the number of leafminers/leaf area (sq cm). The

relationship would be true at 25-27 C temperature range only since

survivorsip of eggs to adult stage was not reported at other

temperatures. This equation was obtained from the observed

distribution of parasitoid eggs densities on host larvae at

different host densities as reported in Chapter VII, and the

observed survival rates of different densities of parasitoid eggs

on host larvae as reported in Chapter VIII. Although it is not a

substitute for a comprehensive leafminer-parasitoid model which has

provisions for stage specific mortality, Ke provides a crude

estimate of the number of parasitoids expected in the next

generation. This could be a useful tool in the field if it is

necessary to obtain a quick estimate of the density of parasitoids

to be expected in the next generation.

In addition to summarizing the results and suggesting the

utility of Ke, I have also suggested that the classical biological

approach should not be neglected. Exploration of other regions for

leafminer parasitoids not found in Florida, and subsequent

importation and establishment of these in Florida should also be

seriously considered in the tomato IPM program.














CHAPTER II
LITERATURE REVIEW

Tomato Production in Florida

Importance

In monetary terms, the tomato is the most important vegetable

crop grown in Florida (Cantliffe 1985). Total value of the crop

increased from $122.3 million in the 1973/74 season to $390.6

million in the 1982/83 season (Van Sickle and Belibasis 1985).

Tomato acreage has increased from 31,500 acres in the 1974/75

season to 47,600 acres in the 1983/84 season. Manatee and

Hillsborough counties and adjacent areas accounted for 17,500

harvested acres in the 1983/84 season while Dade county harvested

12,800 acres in the same season and Collier and Hendry counties

harvested 9,375 acres. These three regions accounted for 40,075

acres of the 47,600 total acres harvested in Florida (Van Sickle

and Belibasis 1985). The yield of tomatoes has increased from an

average 796 25-pound boxes per acre in the 1973/74 season to an

average 1,250 boxes in the 1981/82 season (Van Sickle and Belibasis

1985).

An important aspect of tomato production in Southern Florida

is the perceived competition from West Mexico producers for the

U.S. domestic market. Both regions supply a large part of the

Eastern U.S. winter demand for fresh tomatoes. Van Sickle and

Belibasis (1985) have shown that Florida's share of the tomato

market has increased over the past 10 years, however. The


13








14

widespread use of hybrid varieties such as 'FTE-12', 'Duke' and

'Sunny' has contributed to Florida's competetive edge over West

Mexico, according to those authors. These hybrids have higher

yields, firmer fruits and concentrated production compared to the

traditional varieties. Fields are now picked only 2 or 3 times,

whereas the previous practice was to pick as many as 5 times. In

addition, the cost of marketing and shipping Florida tomatoes is

less than that for Mexican tomatoes (Van Sickle and Belibasis

1985).

The competitive edge has also been sustained by keeping down

the costs of pest control. Twice weekly sprays just to control

leafminers was the recommendation to growers by the University of

Florida's Institute of Food and Agricultural Sciences (IFAS) in the

1960's (Brogdon et al. 1970). Spray applications by a typical

grower for control of all pests were reduced to 14 per season by

1980 (Prevatt, personal communication). The costs (in 1978/1979

prices) of pest control ranged from $228.79 to $438.20 per acre in

1978/1979 season (Brooke 1980). In 1984/85, these costs (in

1984/1985 prices) ranged from $453.43 to $501.96 per acre (Van

Sickle and Belibasis 1985).

Production Methods Relevant to Pest Management

There are two tomato growing seasons a year in the Manatee-

Hillsborough area of Florida. The 110-day fall season starts in

late August and ends in December. The 120-day winter season lasts

from January to April. In the Dade, Collier and Hendry counties,









15

tomatoes are grown once a year in a growing season extending from

approximately November to February. In northern regions of the

state, tomatoes are grown during the spring months. Pest management

decisions are made throughout the growing season. These decisions

are discussed below.

Land selection. Most growers use the same field only in

alternate or every fourth season for growing tomato. Some growers

avoid using fields with a previous history of consistently bad

yields attributable to soil pathogens or nematodes (personal

communications).

Land preparation. Rotovating, forming raised beds, preparing

irrigation and drainage ditches, fertilizing, fumigating and

covering with strip- or full-bed plastic mulch are the cultural

practices commonly employed in preparing the land for planting. The

soil pH should ideally be maintained in the 6.0 to 6.5 range

(Hochmuth 1985). Crop losses due to Fusarium wilt (Fusarium oxysporum

f. sp. lycopersici (Saccardo) Snyder and Hensen races 1, 2, and 3),

Fusarium crown rot (F. oxysporum f. sp. radicis-lycopersici Jarvis

& Shoemaker), southern blight (Sclerotinium rolfsii Saccardo), and

and many nematode-incited diseases can be reduced by adjusting the

pH to within the 6.5 to 7.5 range (Jones and Overman 1985).

Other soil pathogens are not easily managed by simply

regulating the soil pH. These organisms include Verticillium wilt

(Verticillium albo-atrum Reinke and Berthold races 1 and 2), the

damping-off disease caused by Rhizoctonia solani Kuhn and Pythium

spp., Pyrenochaeta brown-rot (Pyrenochaeta lycopersici Schneider and









16

Geralach), and bacterial wilt (Pseudomonas solanacearum E. F. Smith).

These organisms are controlled with soil fumigants (Jones and

Overman 1985). Hochmuth (1985) recommends that at least 50% of the

nitrogen should be in the nitrate form. Blossom-end rot and soil

conditions favorable to fusarium wilt can occur as a result of

reduced calcium uptake if much of the nitrogen is in the ammoniacal

form. He also points out that some pesticides contain elements

that are essential micro-nutrients for the tomato. These micro-

nutrients should not be given separately if previous soil analyses

indicated that they were lacking and if pesticides containing the

micronutrients are to be used. Toxicity could occur as a result of

build up of some elements. Excess copper could induce iron

deficiency, for example.

Tomato yields can be increased greatly by reducing the

diseases caused by soil-borne pathogens by using other methods in

addition to proper liming and fertilizing regimens. This involves

the use of grower-imposed quarantines, exclusion of infected

materials, removal and incineration of crop residues, the use of

resistant cultivars, and soil fumigation.

Resistant cultivars. 'Sunny', 'Duke', 'FTE 12', 'Hayslip',

and 'Floradade' tomato cultivars accounted for 96% of the total

acreage in Florida in the 1984/85 season (Hawkins 1985). Most

cultivars in current use in Florida are resistant to race 1

verticillium wilt but not to race 2 (Jones and Overman 1985). They

are also resistant to F. oxysporum f. sp. lycopersici races 1 and 2








17

and to gray leaf spot (Stemphylium solani Weber) (Maynard 1986).

Scott (1985) has shown true resistance to Fusarium wilt race 3 in

LA716 (Lycopersicon penellii (Correll)D'Arcy) accession. Advanced

testing of lines resistant to bacterial spot (Xanthomonas campestris

pv. vesicatoria (Doidge) Dye), bacterial wilt (Pseudomonas

solanacearum E. F. Smith) or fusarium crown-rot should occur in the

next two years (Scott 1985).

Soil fumigation. It is essential to use broad-spectrum

fumigants even when all the other tactics mentioned above are used

to control the soil-borne pathogens. The fumigants include

methylbromide + chloropicrin, chloropicrin + nematicides, and

methylisothiocyanate + nematicides (Jones and Overman 1985).

Weed control. Nightshade (Solanum americanum Mill.) is the

most important weed in southwest Florida (Gilreath 1985). It belongs

to the family Solanaceae as does the tomato. Because of the close

genetic relationship between the two plant species, tomato would

likely be injured if herbicides were used to control nightshade.

Herbicides can be used to control nightshade if the tomato is

grown on mulched, seepage irrigated beds and the spray is directed

to the row middle so that it does not contact the tomato plant

(Gilreath 1985). Gilreath (1985) found that paraquat (Paraquat R),

metribuzin (SencorR /LexoneR), and other labeled herbicides

will control nightshade if applied to nightshade in the 2 to 4 true

leaf stage of development. Older plants and those hardened by cold

or other factors would not be adequately controlled by labeled

hebicides. Gilreath (1985) has evaluated several new currently








18


unregistered herbicides for their efficacy against nightshade and

other weeds in tomato. He found that a combination of oxyfluorfen

(GoalR) and flauzifop (Fusilade 2000R) provided the best

control of many weeds, including nightshade and grasses, and that

there was a rapid kill of emerged weeds. It is his hope that

GoalR will be registered for use in mulched tomato middle in the

next two years.

Other weeds of tomato include the nutsedge (Cyperus spp.),

ragweed (Ambrosia artemesifolia L.), Medicago spp., common beggar-

tick (Bidens pilosa L.), and downy ground cherry Physalis pubescens

L.). Almost all weeds other than nightshade are easily controlled

by currently registered herbicides. Nutsedge is controlled by

methylbromide, for example (Dunn 1985).

Foliar disease management. Regular maneb/mancozeb and

copper applications are made to control bacterial infections.

They have to be used carefully. Certain copper and mancozeb

combinations may lead to higher incidence of target spot

(Corynespora cassiicola (Berk. & Curt)), than the use of

chlorothalonil (Jones and Jones 1984). Conover and Gerhold (1981)

showed that maneb or mancozeb alone were more effective than a

combination of basic copper sulfate with either maneb or mancozeb

in controlling late blight (Phytophthora infestans (Montagne) de

Bary) and gray leaf spot. Metalaxyl, which is now registered for

use on tomato, gives excellent control of the tomato late blight,

and related fungi (Pohronezny 1985). Samoucha and Cohen (1984)









19

have shown a high degree of cross-resistance to other compounds by

fungus biotypes that have resistance to metalaxyl, however.

A weather-based system, BLITECAST, predicts outbreaks of late

blight and is used in many parts of the country. It has not been

very useful under Florida conditions, however. First outbreaks of

late blight occur much later than predicted by BLITECAST

(Pohronezny 1985).

Insect pest management. The tomato is damaged by direct and

indirect insect pests. Direct pests attack the marketed product,

the fruit. The armyworms (Spodoptera eridania (Cramer), S. exigua

Hubner), and S. dolichos (F.)), the tomato pinworm Keiferia

lycopersicella (Walsingham)), the tomato fruitworm (Heliothis zea

(Boddie)), the tobacco budworm (H. virescens (F.)), and the southern

green stink bug (Nezara viridula (L.)) are some of the direct

insect pests of tomato.

Indirect insect pests of tomato include leafminers as well as

most of the pests listed above when they damage plant parts other

than the fruit. Defoliation could result in reduced photosynthetic

activity and reduced fruit size. The sun may scorch fruit on the

vine if leaf shade is removed by defoliation (Musgrave et al. 1975).

Pathogens may exploit damaged parts for invasion of the plant.

Alternaria alternata (Fries) Keisler, which is weakly parasitic,

and Xanthomonas campestris pv. vesicatoria enter leafmines to

invade the tomato (Keularts 1980).

Chemical control. Several effective insecticides are available

and in current use by growers (Johnson 1985). Permethrin (AmbushR),









20


methomyl (LannateR), endosulfan (ThiodanR), and methamidophos

(MonitorR) are some of the popular insecticides for control of

tomato insect pests. Avermectin (AvidR) and cyromazine (TrigardR)

are new insecticides that are effective against leafminers (Schuster

and Everett 1983), but they are currently not registered for use

against leafminers on tomato in Florida. The effect of insecticides

on leafminers and their natural enemies is described in detail in

the section on leafminer control in this chapter.

Integrated pest management. There are essentially two types

of strategies for insect management available to growers on tomato.

Some growers treat their crops when needed. Other growers spray the

crop on a regular basis whether they know pests are present in the

field or not. Both strategists employ resistant cultivars, intensive

pre-plant treatment of soil with nutrients and multi-purpose

fumigation, hebicides and other practices such as staking and

mulching.

Growers who treat their crops for insects only when needed,

make their decisions on recommendations from professional scouting

consultants or use their own judgement. The employment of

professional scouts is now widespread (Pohronezny 1985).

Integrated Pest Management (IPM) programs were initiated in the

Homestead region in 1976 (Pohronezny and Waddill 1978) and, with

modifications, in the Manatee-Ruskin area in 1978 (Schuster et al.

1980). Funding for the programs was from state and federal sources

but scouts were encouraged to form private firms and solicit









21

contracts from the growers. There are now at least two tomato IPM

consultant companies in the Manatee-Ruskin area, and one in the

Homestead region.

Sampling methods remain very similar to those initially

developed by Phoronezny and Waddill (1978). Scouts in both areas

spend most of their time in the field sampling leafminers and

pinworms. All other scouting activity is secondary in that

observations of other pests are made at the same sampling stations

and while walking from one station to the next. Schuster et al.

(1980) described the procedure, technique and leafminer thresholds

for the Manatee-Ruskin area. Each field is scouted twice weekly

throughout the growing season. More stations are allocated to

field boundaries than to the middle of a field, as pinworm and

leafminer populations seem to be distributed in clumps, with higher

densities of these insects being found near boundaries. There is

one sampling station for every 2.5 acres of a field. Plants are

sampled for leafminers, pinworms, and other pests.

Live leafminer larvae are now counted instead of the earlier

method of counting total mines in deciding whether or not

leafminers have reached a population level that required chemical

treatment. In considering only live larvae and excluding larvae

killed by parasitoids, the role of natural enemies in regulating

leafminer populations is recognized (Pohronezny et al. 1984).

The threshold for treating lepidopteran pests has been altered

from 1 larva/field to 1 egg/field after fruiting (Schuster et al.

1980). Pena (1983) revised the tomato pinworm threshold to 0.67









22

larva/plant or 0.83 foliar injury/plant. To save time on the

laborious task of counting mines, Schuster and Beck (1983)

established a visual rating system for assessing total leafmines.

Liriomyza trifolii (Burgess)

Description

Adult L. trifolii are 1.3 mm long. Their wing length is

1.5 mm (Burgess 1880). A full description of a neotypic male

specimen reared from alfalfa was given by Spencer (1965). Spencer

(1973) also illustrated male genitalia, a character which he uses

to differentiate closely related Agromyzidae. L. sativae

Blanchard is often confused with L. trifolii. These species

can be readily identified by differences in four features. The

curvature of the aedeagus is more pronounced in L. trifolii than

in L. sativae. The L. trifolii mesonotum is distinctly greyish-

and matte while that of L. sativae is shining black. The area of

the L. trifolii upper orbits and much of the hind margin of the

eye is yellow and both vertical bristles are on a yellow background.

The upper orbits and much of the hind margin in L. sativae is dark

and the vertical bristles are on a dark background (Spencer 1981a).

Knodel-Montz and Poe (1982) used scanning electron microscopy

to study female adults. They showed that L. sativae and L.

trifolii females can be distinguished by the denticles and the

egg guides on the genitalia. The denticles are angular and the egg

guide is V-shaped in L. trifolii. In L. sativae, the

denticles are elongate and the egg guide is acutely angled. Zehnder









23
and Trumble (1983) utilized electrophoretic and scanning electron

microscopy techniques to separate the two species.

L. trifolii eggs are "oval, creamy or translucent and

approximately 0.2 x 0.1 mm (E. A. Mortimer personal communication).

The larva, which is initially colorless, darkens to yellow as it

matures and the pupa is orange-yellow" (Bartlett and Powell 1981,

p. 186).

A distinguishing characteristic of agromyzid larvae is that

the anterior pair of spiracles is located dorsally, adjacent to the

dorsomeson. The anterior spiracles are laterally located in other

cyclorrhaphan larvae (Peterson 1979). The number of bulbs on the

posterior is sometimes useful in separating agromyzid species.

There are 3 spiracular bulbs in L. sativae while 6 bulbs occur

in L. huidobrensis Blanchard, another closely related species

(Spencer 1981a).

The larva has 12 segments (the head, 3 thoracic segments, and 8

abdominal segments). Most, and sometimes all, segments have

distinctive minute tubercles in bands that are mainly confined to

the segmental boundaries. Muscle scar patterns on the body are also

found. The body tends to be cylindrically uniform but the anterior

and posterior taper (Allen 1956). Agromyzid larvae have 2 mandibles

The mandibles and the rest of the much reduced mouth parts form the

cephalopharyngeal skeleton.








24

Taxonomy

"Article 35 of the International Commission on Zoological

Nomenclature states that a name, once placed in homonymy, must be

rejected. As a result, the economically important and well-known

Liriomyza trifolii (Burgess) requires a new name" (Zoebisch

1984, p.5). Zoebisch (1984) has shown from his review of literature

on L. trifolii taxonomy that the species ought to be renamed

since it is a homonym of Agromyza trifolii Kaltenbach. L.

trifolii (Burgess) was first described and named by Burgess (1880)

as Oscinis trifolii. Coquillet (1898) placed 0. trifolii in the

genus Agromyza after having synonymized the species with Agromyza

diminuta. 0. trifolii was placed in the genus Liriomyza by

de Meijere (1925). He was also the first to note the homonymy of

A. trifolii (Burgess) and A. trifolii Kaltenbach. Spencer (1981a)

synonymized L. alliovora of Frick (1955) and also 1. archboldi of

Frost (1962) with L. trifolii.

Distribution

L. trifolii is of nearctic origin. Several authors put

its origin in Florida (Spencer 1981b; Parrella and Keil 1984) but

the first description of the species made by Burgess (1880) was of

a specimen collected from white clover in the District of Columbia.

This is important to note as many of the authors who place the

origin of the species in Florida also claim that the worldwide

spread of the species resulted from the shipping of infested plant

material from Florida during the last decade. The species was

known to exist in Oregon, California, Indiana as well as Florida








25


and was suspected to occur throughout the U.S. as early as the

1950's (Frick 1959). This is much earlier than the reported spread

into the rest of the U.S. from Florida as claimed by Parrella and

Keil (1984).

Irrespective of the contention that Florida is the origin of

L. trifolii, the species is now distributed worldwide. It has

been recorded in N. America, S. America, Africa, Europe, and Asia

(Table 2-1). A distribution map of L. trifolii is available

from the Commonwealth Institute of Entomology (CIE) in London,

England (CIE 1984).

Host Plant Range

L. trifolii is a polyphagous species (Spencer 1964). The

reported host plant range of L. trifolii is presented in Table

2-2. Only literature sources published since 1981 have been

utilized in the table because there was much confusion about the

leafminer species involved prior to Spencer's (1981a) clarification

on the differences between L. trifolii and L. sativae.

Hence, the citation of a host plant does not necessarily mean a

first record. It is interesting to observe that no author notes

more than 100 species as hosts while the total worldwide host plant

range is 148 species in 31 families. For example, Stegmaier (1981)

does not record watermelon as a host of L. trifolii in Florida,

while Poe and Montz (1981a, 1982) record that the insect was reared

from watermelon in Texas. Fagoonee and Toory (1984) also recovered

L. trifolii from watermelon in Mauritius.









26




Table 2-1. Current Known Distribution of L. trifolii (Burgess).



Region Area Literature source



N. America Califonia Poe and Montz 1982
Florida "
Georgia "
Hawaii "
Indiana Frick 1959
Iowa Stegmaier 1968
Massachussetts Vittum 1982
Mexico Poe and Montz 1982
Minnesota "
North Carolina "
Nova Scotia, Can. "
Ohio
Ontario, Can.
Oregon Frick, 1959
Pennsylvania Poe and Montz, 1982
South Carolina "
Texas "
Washington Frick 1959
Wisconsin Price 1981

S. America Colombia Poe and Montz 1982

Carribean US Virgin Is. Poe and Montz 1982
Bahamas Spencer and Stegmaier 1973
Barbados Spencer 1963

Europe Denmark Bartlett and Powell 1981
Finland Tuovinen and Aapro 1981
France d'Aguillar and Martinez 1979
Holland Poe and Montz 1982
Italy Arzone 1979
Malta Bartlett and Powell 1981
Spain "
Sweden Nedstam 1981
U.K. Anon 1977
W. Germany Bartlett and Powell 1981


continued...









27




Table 2-1 continued.



Region Area Literature source



Africa Canary Is. Poe and Montz 1982
Kenya de Lima 1979
Mauritius Fagoonee and Toory 1983
Reunion Vercambre 1980
Senegal Spencer 1985
Tanzania Spencer 1985

Asia Israel Spencer 1985
Japan Minkenberg and Lenteren 1986








28


Table 2-2. Host plant range of L. trifolii



Botanical Literature
name source


Acanthaceae
Thunbergia sp. Poe and Montz 1982
Amaryllidaceae
Alstromeria sp. Bartlett and Powell 1981
Amaranthaceae
Amaranthus retroflexus L. Fagoonee & Toory 1984
A. viridis L. Zoebisch & Schuster 1984
Celosia sp. Poe and Montz 1982
Apiaceae
Apium graveolens Stegmaier 1981
var. dulce (Mill.) Pers. Stegmaier 1981
Coriandrum sativum L. Fagoonee & Toory 1984
Daucus carota L. Broadbent 1982
D. carota
var. sativae L. Stegmaier 1981
Hydrocotyle bonariensis
Comm. ex Lam Fagoonee & Toory 1984
H. umbellata L. Stegmaier 1981
Pastinaca sativa L. Powell 1981
Asclepiadaceae
Asclepsias syriaca Broadbent 1982
Asteraceae
Ageratum conyzoides L. Fagoonee & Toory 1984
Ambrosia artemisiifolia L. Broadbent 1984
Arctium sp. Broadbent 1982
Aster cordifolius L. Smith and Broadman 1986
Aster sp. Stegmaier 1981
Baccharis halimifolia L. Stegmaier 1981
Bellis perennis L. Fagoonee & Toory 1984
Bidens alba L. Zoebisch & Schuster 1984
B. pilosa L. Stegmaier 1981
Calendula officinalis L. Powell 1981
Callistephus chinensis (L.) Nees Stegmaier 1981
Centaurea cyanus L. Powell 1981
Chrysanthemum leucanthemum Smith and Broadman 1986
Chrysanthemum X morifolium Ramat Poe and Montz 1981a
Dahlia sp. Stegmaier 1981
Eclipta prostrata L. Fagoonee & Toory 1984

continued...








29




Table 2-2 continued.



Botanical Literature
name source


Erechititis heiracifolia (L.) Raf. ex DC Stegmaier 1981
Eupatorium coelostinium L. Stegmaier 1981
E. serotinum Mich X. Stegmaier 1981
Gaillardia aristata Pursh Stegmaier 1981
Galinsoga ciliata (Raf.)Blake Stegmaier 1981
G. quadriradiata Ruiz.& Pay. Smith and Broadman 1986
Gazania sp. Bartlett and Powell 1981
Gerbera sp. Fagoonee & Toory 1984
Gerbera jamesoni Bolus ex Hook Stegmaier 1981
Gamochaeta pennsylvanica (Willd.)Cabrera
(Gnaphalium spathalium Lam.) Stegmaier 1981
Helianthus annus L. Stegmaier 1981
H. bipinnatus Cav. Smith and Broadman 1986
Hymenopappus scaboseasus L'Her Stegmaier 1981
Lactuca canadensis L. Stegmaier 1981
L. sativa L. Stegmaier 1981
L. serrioala L. Broadbent 1984
Launaea cornuta (Oliv. & Hearn) Jeffrey Spencer 1985
Melanthera aspera Mich X Stegmaier 1981
Mikania scandens (L.) Willd. Genung 1981
Senecio glabellus Poir. Stegmaier 1981
S. jacobaea L. Powell 1981
S. vulgaris L. Broadbent 1982
Solidago sp. Broadbent 1982
Sonchus asper (L.) Hill Stegmaier 1981
S. oleraceous L. Genung 1981
Synedrella nodiflora (L.) Gaertn. Stegmaier 1981
Tagetes erecta L. Stegmaier 1981
T. indica L. Fagoonee & Toory 1984
Tagetes sp. Fagoonee & Toory 1984
Taraxacum officinale Weber Broadbent 1984
Tithonia diversifolia (Hemsl.) Gray Spencer 1985
Tridax procumbens L. Stegmaier 1981
Xanthium sp. Stegmaier 1981
Zinnia sp. Stegmaier 1981
Balsaminaceae
Impatiens sp. Poe and Montz 1982

continued...









30



Table 2-2 continued.



Botanical Literature
name source


Boranginaceae
Cordia myxa L. Fagoonee & Toory 1984
Brassicaceae
Brassica campestris
var. rapa L. Fagoonee & Toory 1984
B. chinensisL. Fagoonee & Toory 1984
B. juncea (L.) Czern. Fagoonee & Toory 1984
B. oleraceaea var.
botrytis L. Fagoonee & Toory 1984
B. oleraceaea. var.
capitata L. Fagoonee & Toory 1984
B. oleraceaea. var.
gemmiferae L. Fagoonee & Toory 1984
Capsella bursa-captoris (L.) Medic. Powell 1981
Raphanus raphanistrum L. Broadbent 1982
R. sativus L. Fagoonee & Toory 1984
Thlaspi arvense L. Broadbent 1984
Caryophyllaceae
Dianthus sp. Poe and Montz 1982
Gypsophila sp. Poe and Montz 1981a
Stellaria media (L.) Vill Broadbent 1982
Chenopodiaceae
Beta vulgaris L. Stegmaier 1981
Chenopodium album L. Stegmaier 1981
Spinacea oleracea L. Stegmaier 1981
Convolvulaceae
Ipomoea batatas (L.) Lam. Fagoonee & Toory 1984
Cucurbitaceae
Citrullus lanatus
(Thunb.) Matsumura & Nakai Poe and Montz 1981a
Cucumis melo L. Stegmaier 1981
C. sativus L. Stegmaier 1981
Cucurbita mixima Duchesne Fagoonee & Toory 1984
C. pepo L. Stegmaier 1981
Lagenaria siceraria
(Molina) Standl. Fagoonee & Toory 1984
Melothria pendula L. Genung 1981



continued...









31




Table 2-2 continued.



Botanical Literature
name source


Euphorbiaceae
Ricinus communis L. P. Parkman, personal communication
1986
Fabaceae
Arachis hypogea L. Fagoonee & Toory 1984
Cajanus cajan (L.) Druce Fagoonee & Toory 1984
Cicer arietinum L. Fagoonee & Toory 1984
Crotolaria incana L. Stegmaier 1981
Glycine max (L.) Merr. Fagoonee & Toory 1984
Lathyrus odoratus L. Powell 1981
Medicago sativa L. Stegmaier 1981
M. lupulina L. Smith and Broadman 1986
Phaseolus aureus Roxb. Stegmaier 1981
P. coccineus L. Powell 1981
P. lunatus L.
(limensis McFad.?) Fagoonee & Toory 1984
P. vulgaris L. Fagoonee & Toory 1984
Pisum sativum L. Stegmaier 1981
Trifolium repens L. Stegmaier 1981
Trigonella foenum-graecum L. Fagoonee & Toory 1984
Vicia faba L. Fagoonee & Toory 1984
V. angustifolia Reichard Smith and Broadman 1986
Vigna luteola (Jacq.) Benth. Genung 1981
V. unguiculata (L.) Walp
(repens?) Fagoonee & Toory 1984
Lamiaceae
Ajuga remota Benth. Spencer 1985
Lamium amplexicaule L. Broadbent 1984
Salvia splendens Ker-Gawl. Poe and Montz 1982
Lillaceae
Allium cepa L. onion Stegmaier 1981
A. sativum L. Fagoonee & Toory 1984
A. porrum L. Fagoonee & Toory 1984
A. schoenoparsum L. Poe and Montz 1982
Malvaceae
Abelmoschus esculentus J. C. Wendl. Stegmaier 1981
Malva moschata L. Smith and Broadman 1986
M. neglecta Wallr. Broadbent 1982

continued...









32




Table 2-2 continued.



Botanical Literature
name source


Oxalidaceae
Oxalis corniculata L. Fagoonee & Toory 1984
0. corymbosa DC. Fagoonee & Toory 1984
0. latifolia HBK. Fagoonee & Toory 1984
Plantaginaceae
Plantago lanceolota L. Broadbent 1984
P. major L. Broadbent 1984
Polemoniaceae
Phloxdrum mondii L. Fagoonee & Toory 1984
Polygonaceae
Polygonum convolvulus Smith and Broadman 1986
P. persicaria L. Broadbent 1984
Primulaceae
Primula sp. Bartlett and Powell 1981
Ranunculaceae
Ranunculus repens L. Powell 1981
R. sp. Powell 1981
Rosaceae
Crataegus monogyna Jacq. Powell 1981
Scrophulariaceae
Antirrhinum majus L. Fagoonee & Toory 1984
Linaria canadensis (L.) DC. Fagoonee & Toory 1984
Solanaceae
Capsicum sp. Stegmaier 1981
Lycopersicon esculentum Mill. Stegmaier 1981
Petunia sp. Stegmaier 1981
Physalis sp. Stegmaier 1981
P. pubescens L. Zoebisch & Schuster 1984
Solanum mauritianum Scop. Fagoonee & Toory 1984
S. dulcamara L. Powell 1981
S. indicum L. Fagoonee & Toory 1984
S. melongena L. Stegmaier 1981
S. americanum L. Zoebisch & Schuster 1984
S. nigrum L. Fagoonee & Toory 1984
S. tuberosum L. Spencer 1985
Tropeolaceae
Tropaeolum majus L. Powell 1981
T. peregrinum L. Powell 1981

continued...









33




Table 2-2 continued.



Botanical Literature
name source


Turneraceae
Piriqueta caroliniana (Walt.) Urban 'Stegmaier 1981
Verbenaceae
Lantana camara L. Fagoonee & Toory 1984
Verbena officinalis L. Fagoonee & Toory 1984
Zygophyllaceae
Kallstroemia mixima (L.) Hook.& Arn. Stegmaier 1981
Tribulus terrestris L. Poe and Montz 1982









34

Life History

Eggs are laid singly in the leaf meosophyll by a female

inserting her ovipositor through the upper epidermis and creating a

depression or 'stipple' which varies in size and shape according to

the age of the female (Personal communication Dong 1980). The

larva feeds on the mesophyll, creating a serpentine mine. A trail

of fecal material, alternating from one side to another, is

deposited in the mine. The mature third-instar larva makes a semi-

circular hole in the epidermis and drops to the soil. As with

other Agromyzidae, L. trifolii pupates within the cuticle of

the last larval instar. Adults emerge from the puparia and can

mate on the first day (personal observation). After mating, female

begin ovipositing in leaves. Females also feed from puctures made

in the leaves. These punctures are larger than oviposition

punctures (Zoebisch 1984).

Longevity. Adult lifespan varies with environmental

conditions. Temperature and food sources are especially important

regulators of longevity. Charlton and Allen (1981) found that at

23.8 C females lived 22.7 days and males lived 13.9 days when the

the flies were given honey in addition to the host plant, black-

eyed pea. Interestingly, when only honey was provided, males

lived an additional 1.9 days; the female lifespan, on the other

hand, was shortened by 6.4 days. Supposedly, males also feed from

the punctures made by females (Musgrave et al. 1975); however, if

this were the case, males would live longer when the host plant was

made available in addition to honey.









35


Leibee (1981a) found that temperature influenced leafminer

longevity on celery. The females lived 27.7, 28.3, 16.8, and 14.6

days at 15, 20, 25, and 30 C respectively when a 10% honey solution

was provided with celery. Parrella et al. (1983b) showed that

female lifespans were 14, 12, and 10 days on chrysanthemum, celery,

and tomato, respectively, when honey was also provided (the authors

do not report at what temperature the flies were maintained).

Fecundity. Like longevity, the egg laying capacity of

leafminers also varies. Once again, temperature, host plant and

food supplements influence the fecundity of flies. L. trifolii

laid 239.67, 405.67, 288.25, 182.33, and 24.33 eggs at 35, 30, 25,

20, and 15 C, respectively, on celery when the flies were provided

with a 10% honey solution (Leibee 1981a). Flies maintained at 23.8

C on black-eyed pea laid 177 eggs (Charlton and Allen 1981). An

average 439 eggs was deposited when the flies were given honey

also. The mean numbers of viable eggs were 298, 212, and 39 on

chrysanthemum, celery and tomato when honey was also given to the

flies (Parrella et al. 1983b). The daily oviposition rates peaked

on day 1, 2, and 4 at 35, 30, and 25 C, respectively, on celery.

Also, oviposition was much reduced at lower temperatures and almost

stopped at 15 C (Leibee 1981a). Even though the flies lived much

longer at the lower temperatures, they cannot be very damaging to

the host plants since fewer eggs are laid.

Provision of additional carbohydrate sources results in

increased fecundity. Naturally occurring carbohydrate sources









36

include floral and extra-floral nectars, and aphid and other

homopteran honeydew secretions. If such sources were exploited by

the flies, their fecundity would be increased. Zoebisch and

Schuster (in press) have demonstrated that potato aphid,

Macrosiphum euphorbiae (Thomas), honeydew on tomato leaflets

increased the fecundity and longevity of L. trifolii under

laboratory conditions.

Developmental times for L. trifolii on some crop plants.

Temperature regulates the rate at which L. trifolii develop.

Host plants also affect developmental times (Table 2-3). It is

interesting to note that Leibee (1981a) reports much longer

developmental times for the species on celery than other authors

report on chrysanthemum, pink bean and tomato. For example, at

15.0 C, larvae on celery mature after 25.8 days while larvae pupate

in 12.6 days at 14.8 C on pink bean and in 10.1 days at 15.6 C on

tomato. Why development takes much longer on celery is not clear.

Celery is not a "bad" host plant for L. trifolii since the fly

lays comparable numbers of eggs on celery and favored host plants,

such as the chrysanthemum (Parrella et al. 1983b).

Mortality. Temperature also influences leafminer mortality.

Only 9.4% of the pupae survived at 35 C while almost 10 times as

many pupae survived to the adult stage at temperatures ranging from

15 to 30 C (Leibee 1981a). Parrella et al. (1983b) found that at

37.8 C 100% pupae died on White Hurricane cv. chrysanthemum. The

host plant also influences mortality of the immature stages.

Charlton and Allen (1981) reported that total larval and pupal









37



Table 2-3. Developmental times for L. trifolii immature
stages on some crop plants.



Temp #days as Host Literature
C egg larva pupa plant source


14.8 10.7 12.6 28.0 pink bean Charlton and Allen 1981
15.0 10.0 25.8 celery Leibee 1981a
15.6 10.1 tomato Schuster and Patel 1985
20.0 3.0 8.0 10.6 mum Charlton and Allen 1981
4.2 6.7 9.4 pink bean Charlton and Allen 1981
4.4 12.0 celery Leibee 1981a
21.1 7.1 tomato Schuster and Patel 1985
25.0 2.2 4.7 8.2 mum Charlton and Allen 1981
2.9 4.5 8.5 pink bean Charlton and Allen 1981
2.3 8.0 celery Leibee 1981a
26.7 4.4 tomato Schuster and Patel 1985
30.0 3.8 5.1 6.7 mum Charlton and Allen 1981
2.2 3.7 6.8 pink bean Charlton and Allen 1981
6.8 celery Leibee 1981a
32.2 3.5 tomato Schuster and Patel 1985
32.5 5.2 4.5 7.6 mum Charlton and Allen 1981
2.0 3.4 6.9 pink bean Charlton and Allen 1981
2.0 5.4 celery Leibee 1981a









38

mortality was 26.9, 26.8, 52.6, and 99.0% on pink bean, black-eyed

pea, 'show-off' mum and 'yellow knight' mum, respectively. As

humidity is increased, the percentage of pupae becoming adults also

increases. More than 60% of the pupae held at any humidity higher

than 50% RH survived to become adults.

Circadian behavior. L. trifolii exhibits diurnal patterns

in many of its primary behavioral phenomena including feeding,

oviposition, larval emergence, and adult emergence (Charlton and

Allen 1981). Most of these activities occurred primarily in the

late morning hours. Most L. trifolii larvae emerged from

black-eyed pea leaves between 0830 and 1330 h (Charlton and Allen

1981). No larvae exited mines between 1830 and 0630 h. Adult

emergence was concentrated between 0930 and 1230 h and no adults

emerged from 1530 to 0730 h. Oviposition and feeding continued

throughout the daylight hours, but the period of greatest oviposition

activity was from 1130 to 1430 h. In this same period, the ratio

of punctures: eggs was also at its lowest. Feeding occurred

throughout the day, although the number of feeding punctures

was the lowest during 0630 to 0830 h and after 2030 h.

Zehnder and Trumble (1984b) used yellow sticky trap cards to

monitor adult leafminer activity in tomato. They found that, in

1981, more L. trifolii were trapped between 1500 and 2000 h.

But in 1982, equal numbers were trapped between 0700-1100 h and

1100-1500 h with fewer adults trapped during 1500-2000 h. It

is interesting to note that the period of greatest oviposition









39
activity recorded by Charlton and Allen (1981) coincides with the

lower daytime sticky trap counts monitored by Zehnder and Trumble

(1982) in 1982. Zehnder and Trumble (1982) contend that the high

trap count periods are associated with greater female oviposition

activity, arguing that the more active females find the best

oviposition sites and increase the time available for oviposition.

Diapause. Attempts by Charlton and Allen (1981) to induce

diapause in Californian L. trifolii failed. Larew et al. (1986)

have shown that L. trifolii can overwinter outdoors in Maryland.

They suspect that the insect probably diapauses in the pupal stage.

Chemical Control of Leafminers On Tomato

Leibee (1981b) discussed past and contemporary methods of

control, the importance of leafminer parasites, and the development

of insecticide resistance in Liriomyza spp. leafminers on

vegetables in Florida. He discussed the effectiveness or non-

effectiveness of insecticides used in Florida until 1981 but did

not specifically name the leafminer species involved because of the

difficulty in verification since voucher specimens were not

retained. Several authors have reported on the efficacy of various

chemicals against L. trifolii on tomato since 1981.

In a detailed study, Schuster and Everett (1983) examined the

effects of avermectin, cyromazine, permethrin, fenvalerate (Pydrin R

and methamidophos on oviposition and survival of L. trifolii.

Avermectin and cyromazine were effective in controlling the insect

on tomato in the field. In the laboratory, avermectin killed L.

trifolii larvae and pupae and also inhibited oviposition.









40

Cyromazine had no effect on oviposition. Very few adults emerged

from foliage treated with avermectin and none emerged from

cyromazine treated foliage because no larvae from this treatment

pupated successfully. Methamidophos resulted in 80% of the adults

dying within 24 hours after treatment in the laboratory. Permethrin,

cyromazine, or avermectin did not induce as high a mortality as did

methamidophos. Avermectin is a macrocyclic lactone natural product

isolated from the soil microorganism Streptomyces avermitilis n.

sp. (Brown and Dybas 1982). It is effective against insects from

several orders and against mites and nematodes. Its mode of action

involves inhibition of gamma amino-butyric acid (GABA) mediated

neuromuscular junction impulse transmission. Although the chemical

has no apparent direct effect on adult mortality, it apparently

inhibits the use of the ovipositor. Females are, thus, unable to

make oviposition or feeding punctures (Schuster and Everett 1983).

Cyromazine is an insect growth regulator (IGR) with a mode of

action that is suspected to be hormonal (Parrella et al. 1983a).

Leibee (1985) examined the dosage-response relationship of

cyromazine on the development of L. trifolii. Increasing

dosage led to higher larval mortality as well as increased numbers

of larviform and other abnormally developed pupae. Such pupae

failed to become adults. Lindquist and Casey (1985) collected more

pupae from methomyl treated greenhouse tomato plants than from

cyromazine treated plants. Their study showed that methomyl had an

adverse effect on the parasite population density while neither









41

cyromazine nor parasitoid introduction altered the parasitoid

population density.

Neem seed extract has been tested for activity against L.

trifolii. The leaves and seeds of the neem tree, Azadirachta

indica Juss. have been used in India for hundreds of years as

sources of insect repellents and insecticides (Jacobson 1981). It

is believed that the principle insecticidal compound, azadirachtin,

has IGR properties (Rembold et al. 1982). The chemical is absorbed

by roots and translocated to the leaves (Larew et al. 1984, Webb

et al. 1984). Those authors also report the efficacy of neem seed

extract against the puparia when used as a soil drench, suggesting

that the active ingredient can be absorbed by the pupae through the

puparial cuticle. Webb et al. (1984) have also shown that neem

seed extract and the purified azadirachtin are highly active

against L. trifolii.

Intensive control methods, including the use of several

insecticides, are employed whenever outbreaks of L. trifolii

occur in the greenhouses in England (Powell 1981). The methods

include fumigation of empty greenhouses with methyl bromide, twice

weekly applications of heptenophos (applied as a high volume, HV,

foliar spray), oxamyl applied in granules, permethrin (HV foliar

spray), and resmethrin (ultra low volume spray) application.

Biological Control of Leafminers on Tomato.

Poe and Montz (1981b) summarized the parasitoids recovered

from L. trifolii, L. sativae, L. huidobrensis, and L brassicae.

All parasitoids recovered so far were Hymenoptera and included









42


species in the families Braconidae, Cynipidae, Eucoilidae,

Eulophidae, and Pteromalidae. Some species, including Opius

dimidiatus Ashmead (Braconidae), Ganaspidium sp. (Eucoilidae),

Halticoptera aenea (Walker) (Pteromalidae), and the eulophids,

Chrysocharis ainsliei Crawford, Closterocerus cinctipennis (Ashmead),

Diglyphus intermedius, and D. begini (Ashmead) have been recovered

from more than one species of leafminers. Schuster (1985) reared

the following parasitoids from leafminers infesting tomato in the

Bradenton area during 1980 to 1984: Diglyphus intermedius, Diglyphus

sp., Chrysonotomyia punctiventris (Crawford), Diaulinopsis

callichroma Crawford, Chrysocharis parksi Crawford, Opius sp.,

Halticoptera circulus (Walker), and Ganaspidium sp.

Zehnder and Trumble (1984a) indicated that L. sativae, L.

trifolii, and some of their natural enemies could discriminate

between host plant species of adjacently grown celery and tomato.

Diglyphus begini, D. intermedius, Chrysocharis parksi, C.

ainsliei, Halticoptera circulus, and Chrysyonotomyia

punctiventris were the most abundant parasitoids. C. parksi

was more numerous on tomato, which was a favored host of L.

sativae. D. intermedius was more active in celery where L.

trifolii was the predominant leafminer species. The host plant,

or the host leafminer species, or both may be involved in observed

differential densities of the parasitoid species.

The impact of insecticides on parasitoids has been studied by

several workers. Hills and Taylor (1951) and Shorey and Hall (1963)









43

demonstrated that DDT led to an increase in leafminer populations

due to a reduction in parasitoid density. The subsequent switch

to methoxychlor, dieldrin, endrin, and lindane (Wene 1955) and also

to parathion, ethion, and diazinon (Getzin 1960) showed the same

results.

Synthetic pyrethroid residues were found to be less toxic to

D. intermedius than was methomyl when pyrethroids were first

introduced to control leafminers (Waddill 1978). Fenvalerate was

the least toxic pyrethroid evaluated. A combination of leptophos

(PhosvelR) and endosulfan (ThiodanR) sprayed on tomato foliage

killed fewer parasitoids than the control (water spray) or any other

insecticide tested (Poe et al. 1978). Weekly applications of

oxamyl (Vydate R) on tomato foliage reduced the number of

parasitoids reared from excised tomato foliage compared to the

water control (Schuster et al. 1979). A single application of

oxamyl at rates ranging from 30g Al (active ingredient)/100 1 to as

high as 119.8g AI/100 1 or methamidophos at 89.9g AI/100 1 did not

reduce parasitoid emergence.

Getzin (1960) recommended integration of chemical and

biological control methods by selecting chemicals that would kill

leafminers but not the parasitoids. Such chemicals have not been

available until recently. Most insecticides used for leafminer

control have been broad-spectrum and have induced mortality in

parasitoid and leafminer populations.

Schuster and Price (1985) noted that Opius sp. and D.

intermedius were proportionately greater in nonsprayed tomatoes









44


than in sprayed tomatoes where Chrysonotomyia punctiventris was

more abundant. Organophosphates depressed Diglyphus spp. while

favoring Chrysonotomyia spp. in celery (Zehnder and Trumble

1985). The pyrethroid permethrin, on the other hand, seemed to

favor Diglyphus species. Schuster (1985) noted that the

proportions of different parasitoid species varied from season to

season. Methomyl and permethrin application reduced population

densities of all parasitoids including D. intermedius when

compared with a water check (Schuster and Price 1985).

Endosulfan in combination with Bacillus thuringiensis var

kurstaki Berliner did not significantly reduce parasitoid numbers

after 2 applications while endosulfan alone did reduce parasitoid

numbers. D. intermedius did not decrease significantly even

when endosulfan was applied alone. This suggests a differential

activity of endosulfan against the various parasitoids of the

leafminers. Such differential activity was also noted with

methomyl which was more injurious to D. interemdius and Opius

spp. compared to Chrysonotomyia punctlventris.

Methomyl also reduced the percentage of parasitism of

leafminers on celery in California as a result of a reduced rate of

adult parasitoid survival indicated by the increased number of dead

adult parasitoids collected from trays placed under the plants

(Trumble 1985). The ratio of leafminers to parasitoids reared was

lowest for cyromazine and avermectin treated plants (1.18:1 and

1.41:1 respectively) in 1982.









45

In 1983, avermectin and a water check had the lowest leafminer

to parasitoids ratios (3.80:1 and 4.16:1 respectively). These

results suggest that the effects of insecticides could not be

readily predicted from one year to the next. Also they do not

prove conclusively that there is differential susceptibility of

leafminers and parasitoids to cyromazine. Application of

avermectin as well as other chemicals altered the composition of

the parasitoid complex. Avermectin reduced the numbers of D.

intermedius reared, but at the same time increased the proportion

of D. begini in comparison to other treatments. In 1983, no

reduction in D. intermedius populations was observed in the

avermectin treatment. Although Trumble (1985) stated that changing

the pesticide shifted the composition of the parasitoid complex, he

failed to qualify his conclusions by not emphasizing the

differences observed in the composition of the parasitoid complex

during the two years in which in the experiments were done.

From a single season of experiments done in 1984, Zehnder and

Trumble (1985) reported that out of six insecticides (permethrin,

diazinon, naled (DibromR), methamidophos, mevinphos (Phosdrin )

and endosulfan) applied to celery, permethrin produced greater

parasitoid mortality than any other treatment. The authors also

reported that the composition of the parasitoid complex varied with

the treatments. Organophosphates favored C. punctiventris

emergence while permethrin treatment led to recovery of more

Diglyphus species.









46
Parasitoid population levels are monitored in a field by

recording numbers of live leafminer larvae. Spraying decisions are

based on live larvae counts rather than the earlier practice of

using total mine counts, which ignored the contribution of

parasitoids to regulating leafminer populations (Pohronezny et al.

1984). Most workers (e.g. Trumble and Nakakihara 1983) report

parasitoid abundance in fields in terms of % parasitism. This

method is faulty because no correlation between % parasitism and

actual parasitoid densities in the field is established. Schuster

and Beck (unpublished) have formulated a sampling plan for estimating

leafminer densities. This plan could be incorporated into a program

for correlating % parasitism and parasitoid-host densities in

conjunction with measurements made by Marlow et al. (1983) of plant

area at different growth stages of the plant.

Utilization of parasitoids is more intense in greenhouse

tomato cultivation. Success in regulating leafminer populations

with releases of parasitoids into greenhouses has been documented

in several parts of the world. Lindquist and Casey (1985) released

Diglyphus into a greenhouse after an initial population of L.

trifolii had been established. This treatment gave the same

tomato yields as alternative treatments where either cyromazine or

methomyl was used as the control agent. Petitt (1984) managed L.

sativae infestations on greenhouse tomato by vacuuming the

plastic mulch, increasing parasitoid populations, particularly of

Opius sp., and by not using any chemical insecticides in the

greenhouse. Lyon (1984) obtained greater success in regulating









47


L. trifolii populations on greenhouse tomato in France by

initially establishing populations of Diglyphus isaea Walker

and Dacnusa sibirica Telenga in the greenhouse and the surrounding

vegetation rather than waiting to release the parasitoids into

infested greenhouses.

Diglyphus intermedius (Girault)

Taxonomic Status, Hosts and Distribution

D. intermedius is one of eighteen species belonging to

this genus in the family Eulophidae. The species was described by

Girault from one female reared in 1916 from Phytomyza chrysanthemi

Kowarz (Gordh and Hendrickson 1979). Hosts of D. intermedius include

Agromyza frontella (Rondani), Liriomyza brassicae (Riley) L.

sativae, 1. trifoliearum Spencer, L. trifolii, Phytomyza atricornis

Meigen, P. chrysanthemi, and P. nigra Meigen (Kamm 1977; and Gordh

and Hendrickson 1979). Gordh and Hendrickson (1979) consider the

species to be neartic and neotropical in distribution.

Adult External Morphology

The general trend of sexual dimorphism in the subfamily

Eulophinae is expressed by the presence of dorsal rami on the

funicular segments of the male antennae. This is not evident in

Diglyphus, the male of which is distinguished from the female by

smaller size and longer marginal fringe on the forewings (Gordh and

Hendrickson 1979). The genus Diglyphus has two-segmented funicles

and parallel longitudinal grooves on the scutellum. D. intermedius

and D. isaea are the only two species in this genus that closely









48

resemble each other. In the female D. intermedius, only the basal

0.25-0.35 of the hind tibia is metalic colored with coloration

fading along the middle 0.30. In D. isaea, the hind tibia is

predominantly metallic colored and the metallic color does not

fade distally. D. begini has pale markings on its antennal scapes

the middle and hind tibiae are without apical duskiness and the

basal cell of the forewing is sparsely or moderately setose (Gordh

and Hendrickson 1979). D. isaea, an introduced palearctic

species, has not been recorded from Florida.

Parasitoid Abundance in Florida

D. intermedius was at least as numerous as any of the 5

other parasitoids reared during the spring of 1975 from excised

tomato foliage from Immokalee (Poe et al. 1978). D2. intermedius

accounted for 46.6% and Chrysonotomyia formosa Westwood accounted

for 49.7% of all parasitoids reared from excised tomato foliage

during the winter and spring of 1975 (Schuster et al. 1979).

Schuster and Price (In Press) observed that D. intermedius

continues to be one of the most common parasitoids found on tomato

in Florida.

Biology and Life History

Biological notes on a related species, D. begini, were

first made by Webster and Parks (1913). Hendrickson and Barth

(1978) studied the life history and biology of D. intermedius

in the laboratory. They reported that D. intermedius was a

solitary ectoparasitoid of leafminers, although occasionally 2 to 5

parasitoids developed on a single Liriomyza trifoliearum larva.








49


It paralyzed and killed all 3 larval instars, but it usually

deposited eggs on 3rd-instars only. Incomplete larval development

ocurred when eggs were deposited on 2nd-instar hosts, possibly

because of an inadequate food supply. Developmental times were

recorded for parasitoids reared at a constant temperature of 25.5 C

and 60% RH and a photoperiod of 16L:8D. Hosts were reared on bush

snap bean, Phaseolus vulgaris cv Bountiful and lima bean P.

limensis cv Thaxter. As it was difficult to observe parasitoids

inside leaf tissue, artificial mines were constructed with small

cardboard rings sandwiched between a glass microscope slide and

coverslip. The egg, larval, and pupal stages lasted 1, 4, and 6

days, respectively, under the conditions provided.

Patel and Schuster (1983), using a method similar to that of

Hendrickson and Barth (1978), measured developmental rates of D.

intermedius at 15.6, 21.1, 26.7, and 32.3 C constant temperatures.

The quadratic regression equation Y = -0.2028 + 0.0214T 0.0004T2

described the relation found between temperature (T) and the

developmental rate/day (Y). At 32.2 C, parasitoid larval mortality

averaged 88.3% which was twice the mortality at other temperatures.

The manner of pupation in D. intermedius is unusual. The

larva changes from light yelow to lime bluish green and constructs

6-8 pillars made from meconium voided at the time of pupation. The

pillars extend vertically between the two epidermal layers and are

usually laid in pairs along the length of the pupa. It has been

suggested that the pillars protect the pupa from being crushed if








50

the leaf dries out and also prevents the pupa from rolling about in

its mine. Hendrickson and Barth (1978) noted that a female D.

intermedius (N = 6) produced an average 40.2 progeny and lived

from 3 to 4 weeks.

Modeling the Leafminer-Parasitoids System

Introduction

Modeling is fast gaining acceptance as a major tool in life

system studies, including insect population dynamics research

(Ruesink 1975, Huffaker et al. 1976, Barfield and Jones 1979,

Shoemaker 1980, Getz and Gutierrez 1982, Stimac 1982). A system

can be defined as a composition of interacting objects (Smerage

1980). A model is simply a description of a system and may take

many forms. Two factors have contributed to the popularity of the

modeling approach: 1) Computers have made it possible, at a rapid

pace, to realistically simulate and predict the behavior of real

life systems using models. 2) It is becoming increasingly cheaper

to simulate rather than establish experiments to develop and test

system behavior hypotheses.

The leafminer model of Smerage et al. (1980), for example,

can be utilized in predicting leafminer and parasitoid population

dynamics much more cheaply and quickly than a field experiment ever

could. Models such as that by Smerage et al. also provide a frame-

work in which research towards better understanding of a system can

be rationally organized, expressed and analyzed. The model of

Smerage et al. has been a motivating factor for much of the research








51


done on leafminers in Florida. It is therefore appropriate to

review briefly here this model.

The Smerage et al. Leafminer-Parasitoid Model

This model represents insect population systems in general,

although it specifically described the leafminer-parasitoid

population system. Figure 2-1 illustrates a. conceptual model of

the leafminer-parasitoid system. The model pertained to the

coupling of within-field egg, larval, pupal and adult populations

of leafminers, labeled C1, C2, C3, and C4, respectively,

with the egg, larval, pupal, and adult populations of parasitoids,

C5, C6, C7, and C8, respectively, by the parasitism process Hp.

The densities of each class of populations is labeled n., where the

subscript i, is denoted by the same number as the population. For

example leafminer egg population, C1, has density nI.

Transformations H1,...,H8 represent development and

physiological mortality. The flow of an individual, from one store

to the next, resulted from the individual developing into the next

stage. For example, upon completion of the egg stage, a leafminer

individual flowed from the leafminer egg store to the leafminer

larval store through H1. Flow into the egg store resulted from

oviposition, HO

Mortality of an individual in a particular store, for example

in the leafminer egg store, resulted in the flow of that individual

into a 'sink' for dead individuals. Sinks are denoted by triangular





















Hpo
T











ST TVFR



HH n 26 n 3 H n84 G,
2 3



2 C3 4 T nS8













12 14
n ) 5 n6H I n 8 G 8
7~~ n^ S8 ^ \ ^^






16 1







Figure 2-1. Conceptual model of the leafminer-parasitoid system from Smerage
et al. (1980). (see text for explanation)


en
1\









53

symbols in Figure 2-1. Temperature dependence of a process is

indicated by the terminal labeled T.

Insecticide mortalities affecting larvae and adults were

incorporated as extrinsic factors. In Figure 2-1, flow sources,

YI2' YI4A YI6, and YI8 remove individuals from their
respective age classes at prescribed rates that reflect insecticide

induced mortality.

Population density sources ns2, ns4, and n8 represent

external sources of leafminer larvae, leafminer adults, and

parasitoid adults, respectively. Movement of adults between the

field is in response to a gradient in an environmental variable.

G4 and G8 denote the processes that regulate movement of leafminer

and parasitoid adults, respectively. Process G4 may be a function

of the quality of vegetation VF, in the external reservoir, and

VR in the field to which leafminers respond.

Parasitism, Hp, was incorporated as the link between leafminer

and parasitoid populations, a process by which attacked hosts were

killed and a parasitoid egg was simultaneously oviposited on each

host larva. In this model, the rate of parasitism was a function

of host and parasitoid densities and environmental variables. The

exponential and the rectangular hyperbolic functions represented by

equations (7) and (8) are alternative closed-form descriptions of

the family of curves for the rate of parasitism when plotted as in

Figure 1-1.


Y = Cp.Kp.n2 (1 exp(nl/Kh)) (7)








54

Y = C .Kp.n2 (nl/(n1 + Kl)) (8)


In (7) and (8), Y is the number of hosts parasitized/day and Kp is

the specific parasitism rate (hosts/adult-day). In (7) and (8) Kh

and Kl represent the critical host density above which Y is nearly

independent of host density, n1. Parasitoid density is denoted

by n1. The capacitance, or the ground area or volume occupied by

the parasitoids is represented by C

The mathematical model of the system is not reviewed here. It

is sufficient to report here that a mathematical model results in

the system being described as a single entity by a set of complex

equations systematically obtained from the descriptions of the

processes and structure (Smerage 1980). The reader is referred to

the chapter on mathematical model in Smerage et al. (1980) if

interested in more details or mathematical description of the model.

Several simplifying assumptions were made in formulating the

model: 1) the celery crop was a non-limiting host resource for

leafminer reproduction and development, 2) all parasitoid species

were aggregated into an equivalent single species population, 3)

development rates of each species were linear functions of

temperature, 4) oviposition rates were constant and independent of

temperature, 5) both populations were uniformly distributed over

the field, 6) parasitism was an exponential function of host and

parasitoid densities, and 7) movement of both populations was by

diffusion or drift.









55

A combination of hypothetical and preliminary estimates of

parameter values obtained from data collected from greenhouse,

laboratory, and field studies were utilized to run computer

simulations of the Smerage et al. model (Musgrave et al. 1980).

These parameters included durations of 40, 40, 90, and 35 degree-

days for the leafminer egg, larval, pupal, and adult stages.

Corresponding parasitoid durations were 25, 40, 140, and 70 degree-

days.

Leafminer sex-ratio was observed to be 1:1 with each female

laying 4 eggs/day. The parasitoid sex ratio was not measured but

a parasitoid was assumed to lay 1 egg per host, killing 2 hosts/day.

Natural mortalities, other than mortality due to parasitization,

for the egg, larval, pupal, and adult stages for each egg laid,

respectively, were assumed to be 0.4, 0.2, 0.2, and 1.0 for the

leafminer, and 0.2, 0.2, 0.1, and 1.0, respectively, for the

parasitoid. The critical host density (number of hosts/area), Kh,

and the specific parasitism rate (number of hosts parasitized/

parasitoid adult-day), Kp, as defined in equation (4) in Chapter I,

were assumed to be 5 larvae/sq m and 2 larvae/adult-day, respectively.

Simulations using several initial conditions provided

considerable insight into the influence of various factors on the

leafminer-parasitoid population dynamics. Important findings

included the observation that sudden outbreaks of leafminers could

only be suppressed by the parasitoids 20 to 30 days after the

outbreak started. If external reservoirs of leafminers were large,

chemical control of leafminers inside the field was ineffective and









56


required frequent and routine applications, and that under these

conditions could even be considered wasteful (Musgrave et al. 1980).

More importantly, the work of Smerage et al. (1980) in

formulating the leafminer model, and that of Musgrave et al. (1980)

in simulating the model has been a significant factor in

determining the direction of research on leafminers in Florida.

This is especially the case with the work presented in Chapters IV

through IX of this dissertation. As with other research currently

ongoing in Florida, these chapters describe the work done to obtain

better estimates of parameters and descriptions of the processes

involved in the leafminer-parasitoid system.














CHAPTER III
PLANT PRODUCTION AND REARING TECHNIQUES FOR
Liriomyza trifolii (Burgess) AND
Diglyphus intermedius (Girault)

Introduction

Colonies of L. trifolii and D. intermedius were

maintained to insure that healthy, homogeneous populations of

insects were always available. The techniques used for rearing the

insects as well as the host plants are described in this chapter.

As many as 1000-2000 adults per day of each insect species were

produced by using this method.

Several plant species can be used for rearing L. trifolii

(see section on host plant range in chapter II). Perhaps the

fastest and easiest plants to grow are the bean plants, which have

been used by several researchers working on L. trifolii (e.g.

Charlton and Allen 1981). Tomato cultivars 'Walter' and 'Hayslip'

were used because the results of the experiments were to be used to

understand the leafminer-parasitoid relationship on the tomato.

Host Plant Production

Planting. Tomato seeds were planted in SpeedlingR

cellular styrofoam trays. In each cell, 4 to 8 seeds were placed

1/2 cm deep in SpeedlingR soil mix. The soil pH was maintained

between 5.5 and 6.5. Acidic soil mixes were corrected to ideal pH

by adding lime or sodium hydroxide. Two trays were planted every

Monday, Wednesday, and Friday. As many as 12 trays were planted




57









58

per week when more plants were needed for the experiments. Seed

germination took 3 to 10 days, depending on the temperature.

Trays were watered gently every day, although twice daily

watering was required on very hot days. Plants were fertilized

twice weekly, 379 g of 20-20-20 (N, P, K) Miller's brand nutrileaf

solution was alternated with a solution of 80 g Mg(NO3)2.6H20,

196 g KNO3 and 336 g of Ca(NO3)2.4H20 in 10 liters of water to

give a 500 ppm N, 500 ppm K and 50 ppm Mg solution. The solutions

were applied with a HyponexR

Repotting. When the seedlings had reached the first 2-true

leaf stage, they were removed and repotted in 6-inch pots. One

seedling was planted in each pot if the plant was to be utilized in

leafmier-parasitoid density experiments. Four seedlings were

planted in each pot for maintaining the insect colonies as well as

for rearing insects for other experiments. Potted plants were

watered and fertilized the same as the seedlings.

Precautions and care. Chemical pesticides were not applied

as long as their use was avoidable. Plants were kept free of

insects by good sanitation practices, culling, mechanical control,

exploiting natural enemies and the appropiate placing of baits.

Most pest problems were of short duration; they included

cockroaches, aphids, Myzus persici (Sulzer), russett mites,

Aculops lycopersici (Massee), and on rare occasions, armyworms.

DursbanR baits were placed next to trays of younger seedling

plants that were preferred by cockroaches. Russett mite infested








59

plants were discarded and the bench holding infested plants was

drenched with KelthaneR (dicofol) and not used for 1 month.

Green peach aphids were killed by hand. Heavily infested plants

were discarded. On a few occasions, armyworm egg masses were

detected on plants inside the screenhouse. Leaves with these egg

masses were removed.

Trays and transplants were drenched with a solution of 6.75 g.

of TrubanR in 6 liters of water. Leafmold became a problem

during warm humid conditions. If leafmold was present, Dithane

M-45R was sprayed at 16.8 g/l on the plants weekly until the

leafmold disappeared. Leaves infested with leafminer larvae were

removed.

Rearing L. trifolii.

Equipment and procedure. A converted refrigerated truck

body housed the leafminer colony. A 121:12D photoperiod was

provided by fluorescent lighting. An air-conditioner and a space

heater were utilized to maintain the temperature at 22-32 C. The

humidity inside the truck body was uncontrolled and ranged from 40

to 80% RH.

Fifty-200 adult flies were held in a 61 x 61 x 61 cm. organdy

cloth-covered, screened, oviposition cage. Four plants were placed

in the cage and replaced daily between 0800 and 0900 h with fresh

plants. A filter paper smeared with honey was hung from the top of

the cage. The honey ensured a large numbers of eggs were laid

daily (Charlton and Allen 1981). New flies were added periodically

to maintain 50 to 200 flies in the cage.








60


On removal from the ovipositon cage, the plants were shaken to

dislodge any flies still on the plants and placed in holding cages.

Holding cages were similar to the oviposition cage. Plants were

watered daily.

Larvae took 5 days to reach the third-instar. On the sixth

day, the plants were removed from the holding cages. Leaves were
R
cut from the plants and placed in Tupperware (Superseal, 30 x 30

x 12 cm) plastic containers. The containers had screen-covered

holes for ventilation. A hardware cloth divider (11 x 11 mm mesh)

in the box kept the leaves elevated from the container bottoms.

Larvae exiting the leaves dropped to the container bottoms. The

bottoms were coated with a layer of TeflonR material to prevent

the puparia from sticking to the containers. Leaves were

discarded, the hardware screen was removed, and the puparia were

gently brushed off the bottoms with a fine camel-hair brush 3 days

after the leaves were placed in the containers.

About 50-200 puparia were collected and placed in a 30-g clear

plastic creamer cup. A filter paper strip smeared with honey was

placed inside the cup. Adults emerged in 4 to 5 days and fed on

the honey. A cup containing adult flies was placed inside the

oviposition cage and the lid removed so that the adults escaped

into the cage.

Leafminers in all stages of development were available daily.

It took 14 to 16 days for development of an egg to the adult stage

under the conditions prevalent in the rearing facilities









61

Precautions and care. All the cages and containers were

kept free of standing water to prevent leafminer larvae and puparia

from drowning. Plants were watered prior to being placed in the

oviposition cage. Since adult longevity is considerably shortened

by very high temperatures and the egg laying activity is reduced by

temperatures below 17 C, the temperature inside the truck body was

monitored with a 7-day hygrothermograph recorder and failures in

the air-conditioner and heating unit were corrected as soon as

possible after they were noticed. Transfers of exposed plants into

holding cages were recorded daily to monitor the age of larvae.

Puparia transferred were similarly recorded. Extra leafminer

puparia were refrigerated at 5 C and later removed when needed for

the leafminer colony on the few occasions that there were not

enough adults from the regular colony. Puparia were discarded 1

week after being placed in the refrigerator if they were not needed

in the colony.

To maintain vigor of the colony, tomato foliage infested with

leafminers was collected from the field approximately every 6

months, and the insects were reared to the adult stage. Fifty to

200 adults were transferred to a new oviposition cage and

assimilated into the colony.

Too many ovipositing flies (>200) resulted in too many eggs.

Under such overcowded conditions, plants could be totally

defoliated by the developing larvae. The larvae that were able to

pupate were smaller. It was, therefore, necessary to ensure that

no more than 200 adults were present in the oviposition cage.









62

Rearing D. intermedius

Equipment and procedure. The parasitoid colony was kept in

a separate room with temperature, lighting, and humidity conditions

similar to those in the converted refrigerated truck body that

housed the leafminer colony.

One or 2 tomato plants containing 5-day-old leafminer larvae

were removed daily from the truck body. These plants were placed

inside a parasitoid oviposition cage similar to the other cages.

Approximately 20 to 100 adults were introduced into the oviposition

cage every week. Plants were removed from the cage after 1 day's

exposure to the parasitoid adults and replaced with other plants

containing 5-day-old leafminer larvae.

Plants with parasitized host larvae were not caged after

removal from the oviposition cage to make watering easier.

Parasitoid larvae turned bluish-green after 5 to 6 days. At this

point, leaflets were stripped from the plants and put in 1-quart

ice-cream cartons which were capped by clear plastic bags held to

the open top of the cartons by a rubber band.

Adult parasitoids emerged after another 5 to 8 days. A simple

aspirator made of flexible rubber tubing and a glass eye-dropper

plugged with cotton wool was connected to a small vacuum pump to

remove the parasitoid adults from the cartons. Strips of filter

paper smeared with honey were placed in 30-g creamer cups and adult

parasitoids were put into these cups. Approximately 20 to 100

adults were introduced into the oviposition cage every week. The









63
remaining parasitoids were refrigerated at 5 C. These could be

used up to 3 weeks after being placed in the refrigerator. After 3

weeks, the insects were discarded if they were not needed for the

colony.

Precautions and care. Plants with more than 400 leafminer

larvae were not utilized for the parasitoid colony. It was not

easy to maintain these plants because the larvae fed into the leaf

petioles causing early leaf drop. Also, it was necessary to water

the plants daily since they were under greater stress.

Not all larvae were parasitized. Many dropped onto the cage

floor and pupated. Cages were therefore vacuumed every 4 to 5 days

to prevent leafminers from becoming adults and ovipositing on the

plants.














CHAPTER IV
FECUNDITY AND LONGEVITY OF Diglyphus intermedius
AND ITS CONTRIBUTION TO HOST MORTALITY

Introduction

Patel and Schuster (1983) had demonstrated that the rates of

development of the life stages of D. intermedius were functions

of temperature. It is logical to hypothetize that, as with other

poikilothermic organisms, fecundity and adult longevity of D.

intermedius may also be functions of temperature. This was

recognized by Smerage et al. (1980). In formulating the leafminer

model those authors stated that the rates leafminer and parasitoid

ovipositions may be functions of temperature. However, for

simulation purposes, they assigned constant values for oviposition

parameters independent of temperature. They had done this for lack

of information on the quantitative description of the relationship

between temperature and oviposition rates for leafminers and its

parasitoids. Parasitoid fecundity was assumed by those authors to

be 2 eggs/parasitoid-day with each host killed yielding 1 parasitoid

egg. Parasitoid adults were assumed, for lack of better information,

to have a lifespan of 70 degree-days.

It was necessary to obtain better estimates of parameters for

parasitoid oviposition and longevity as functions of temperature to

represent more accurately the real life situation. Experiments to

estimate parameters for parasitoid oviposition and adult female

longevity as functions of temperature are reported in this chapter.



64









65

In addition, parasitoid-induced host mortality during the adult

female lifespan were also measured. To achieve these objectives,

experiments were done in a laboratory utilizing D. intermedius

adults and 3rd-instar L. trifolii larvae reared on tomato

foliage.

Materials and Methods

Five incubators were utilized to obtain 5 constant temperature

conditions. A sixth unit was available as a backup in case a

principal unit failed. The temperatures were 15.6, 19.4, 23.3,

27.2, and 31.1 C. Fluorescent and incandescent lighting provided a

12 h photoperiod beginning at 0800 h each day. To record

temperature and humidity, a 7-day hygrothermograph recorder was

placed on the lower of the 2 shelves in each incubator.

A moistened filter paper was placed in a 150 x 15 mm. clear

plastic petri dish. Twenty tomato leaflet pieces, each containing

a single 3rd-instar leafminer larva, were arranged in a single

layer and placed on the filter paper. A pair of 1-day-old

parasitoids were released into each of 4 petri dishes per each of

the 5 incubators. The parasitoids were transferred every day

between 0800 and 0900 h to new petri dishes containing 20

additional 3rd-instar leafminer larvae. Dead males were replaced

with 1-day-old males. The filter paper inside the dish was kept

moist to prevent leaflet pieces from drying. No honey or other

carbohydrate source was provided.

Live leafminer larvae, empty mines, and dead larvae within mines

were counted every day after the parasitoids were transferred to









66

new petri dishes. Mines containing dead larvae were dissected

under a dissecting microscope and larvae with and without parasitoid

eggs were counted. The number of eggs laid on each host, as well as

parasitoid female longevity, were also recorded.

The experiment was replicated 3 times. Temperatures were

assigned randomly to the incubators for the first replicate.

Temperature assignment was rotated for 2 additional replicates so

that an incubator was not assigned the same temperature again.

Data were averaged over replicates prior to analyses with the

SAS-GLM procedures (SAS 1982).

Results and Discussion

The mean values for host mortality, parasitoid fecundity, and

parasitoid longevity are reported in table 4-1.


Table 4-1. Number of L. trifolii larvae killed by and
the fecundity and longevity of D. intermedius at
different constant temperatures. (+ 1 SD)


Temp (C) No. larvae Fecundity Longevity
killed (days)


15.6 406.6 (168.4) 194.7 (113.3) 43.5 (8.4)
19.4 387.8 (153.0) 221.7 (105.5) 34.4 (6.9)
23.3 281.8 (93.5) 200.2 (107.7) 22.7 (7.7)
27.2 187.8 (55.0) 126.3 (84.4) 16.1 (3.5)
31.1 128.2 (40.5) 67.4 (50.3) 12.4 (3.4)



Host mortality and longevity of D. intermedius decreased with

an increase in temperature. The mean fecundity of D. intermedius

increased from 194.7 at 15.6 C to 221.7 at 19.4 C. and then









67

decreased at temperatures greater than 19.4 C. The graphs of

regression equations obtained by from mean values for parasitoid

fecundity, parasitoid longevity, and parasitoid-induced host

mortality for each replicate at each temperature are presented in

Figures 4-1, 4-2, and 4-3 respectively.

The low r2 of D. intermedius fecundity, even when a

quadratic model was fitted to the observed means, reflect the

variations among mean replicate values which in turn reflect

variations among individual observations. As there were only 4

observations (only three observations in one case since one female

had escaped) in each replicate-temperature combination, the low

r2 value was to be expected. R2 values of greater than .75

were obtained for D. intermedius longevity (Figure 4-2) and

parasitoid-induced host mortality (Figure 4-3) when linear

regression equations were fitted to the observed means. The

results of this experiment can only be utilized to interpret and

predict in the 15.6 to 31.1 C temperature range because no

measurements were made outside this range.

In the 15.6 to 23.3 C range, the parasitoid's fecundity was

higher than that of L. trifolii observed by Leibee (1984) on

celery. Leibee (1984) reported that L. trifolii laid

approximately 24 eggs at 15 C and 182 eggs at 20 C when the flies

were given a 10% honey solution as an additional food source. The

predicted leafminer fecundity ranged from 3 eggs at 15 C to

approximately 31 eggs at 30 C when provided with a 5% sugar water

solution (Patel 1981). As temperature increased to 31.1 C, D.










68




















300











200
@*










100





F= -19.11.+ 42.65T 1.1T 2 (r2= .65)






16.6 10.4 23.3 27.2 31.1
Tz TEMP (C)




Figure 4-1. Relationship between D. intermedius
fecundity, F, and temperature, T.
( : Observation means, n=4)









69


























40 -






320

S0

z
0 -

20








L: 74.32 2.09T (r2= .4)







15.0 19.4 23.3 27.2 31.1

T= TEMP (C)




Figure 4-2. Relationship between D. intermedius
longevity, L, and temperature, T.
( : Observation means, n=4)











70

























400











$ 300 a
*
w S








100








Hm= 721.07 19.1T (r2= .83)






15.6 19.4 23.3 27.2 31.1

Tz TEMP (C)





Figure 4-3. Relationship between # hosts killed,
-a

- 2



























Hm, and temperature, T.

(. : Observation means, n=4)
100
Him 721.97 1B.1T (r2 .83)






15.6 18.4 23.3 27.2 31.1

T. TEMP (C)





Figure 4-3. Relationship between # hosts killed,
Hm, and temperature, T.
( : Observation means, n=4)









71

intermedius fecundity declined while that of L. trifolii

increased.

The maximum parasitoid-induced host mortality and oviposition

rates as well as the highest ratio of eggs laid/day: hosts killed/

day occurred at 23.3 C (Figure 4-4). The ratio of parasitoid eggs

laid/day:hosts killed/day was 0.48, 0.57, 0.71, 0.67, and 0.53 at

15.6, 19.4, 23.3, 27.2, and 31.1 C, respectively. Parasitoid-

induced host mortality and oviposition rates at 15.6 C were

approximately 24 and 37% less than at 23.3 C. These rates declined

by 13.5 and 33.7% when the temperature was increased from 23.3 to

31.1 C. The rate of parasitoid-induced host mortality is apparently

less temperature dependent than is the rate of oviposition.

Parasitic hymenoptera have panoistic ovarioles and their eggs

develop throughout their adult lifespan. The presence of eggs at

all stages of development enables a female to oviposit if hosts are

available (Fisher 1971). If all the mature eggs are laid all at

once, the female can relatively quickly mature those eggs that are

only one stage behind (King and Richards 1969). The rate at which

these eggs mature would be expected to depend on temperature, since

most developmental phenomena are temperature dependent in

poikilothermic organisms.

Parasitoid-induced host mortality, on the other hand, is a

behavioral phenomenon. It may be temperature dependent to the

extent that the parasitoids may be more active and may take less

time to locate and kill hosts at higher temperatures. The actual






















(N







O


0
r r






U>
3 0


...... ................................................. C- -H *
.............................................








......................................... o -H
COM

cC,










co 0.









......................................................... "V r-1 44
m $


.. ............................................
(0















F141 0 4





r(








73

stinging and paralysis of a host takes only a few seconds, however

(Hendrickson and Barth 1978). The arena in which the hosts were

placed in the experiment was so small that a parasitoid would not

have to spend much time or effort in locating a host. The temperature

factor is therefore not likely to have influenced daily host mortality

rates as much as it would have affected the daily oviposition rates.

The maximum ovipositions/day of 8.5, 9.5, 13.5, 11.5, and 8.5

eggs/female were reached at 14, 11, 6, 5, and 4 day ages of the

parasitoids, respectively, at 15.6, 19.4, 23.3, 27.2, and 31.1 C

(Figure 4-5). The maximum parasitoid-induced host mortalities/day

were 14 hosts killed at age 7 days, 16 hosts killed at age 12 days,

18.5 hosts killed at age 6 days, 17.5 hosts killed at age 5 days,

and 16 hosts killed at age 5 days at 15.6, 19.4, 23.3, 27.2, and

31.1 C, respectively (Figure 4-6).

In comparison to these maximum ovipositions/day and parasitoid-

host mortalities/day for D. intermedius, L. trifolii attained

maximum daily ovipositions of 35 to 39 eggs on age 4 days at 25 C,

age 2 days at 30 C, and age 1 day at 35 C. The maximum L. trifolii

ovipositions of 15 eggs in a day occurred at age 5 days at 20 C.

Minimal leafminer oviposition occurs at 15 C with a maximum of 2

eggs/female aged 6 days (Leibee 1984). The respective predicted

average daily rates for D. intermedius at 15, 20, 25, and 30 C

are 5.2, 0.7, 0.3, and 0.34 for host mortality and 2.3, 0.4, 0.2,

and 0.2 for oviposition for every L. trifolii egg laid at the

maximum rates as reported by Leibee (1984).















74

















I
f 1








S00
....--......

.. .. co
0 0 0 0




I : a
. 5 S co v




0 a






..** )
. X..
o E)











..*. ..... .. 4





.. )
.'.. ,,..-U r 4







.. .. . *...
ftw .I.......
......... ...-.



















* . ". . ..- "- -" *O l
4" 9 4 E






-











75

















0o 0 U C





* 0 ***- -




.---- o






"/ o a
" <














.. -". S. .. I
.; ,*" ... "
't.* -, .... L** >

,. . I .
o... .... -C .
I -
e (



















"-. -., .. _...-N *-, -


IA I- .. I ii i
,1.."
















''*"I .O 4
-- --- * p -"
OI
0)-1
AV /03t11)l 61SOH4
*****2::












* a
AVI3.)I9LO









76
With average lifespan host mortality and oviposition rates as

high as 5.2 and 2.3 times that of the maximum leafminer oviposition

rates at 15 C, it would be expected that D. intermedius alone

might be adequate for regulating leafminer populations at low

temperatures. D. intermedius oviposition and host mortality

rates can be as low as 0.34 and 0.2 times the maximum L. trifolii

oviposition rates at temperatures greater than 20 C. D. intermedius

is at a disadvantage relative to L. trifolii with an increase in

temperature above 20 C.

Under these circumstances, it might be expected that L. trifolii

populations may increase under warming conditions. This has been

observed in the spring tomato season (Schuster personal

communication). The results of the experiments reported here may

partially explain why leafminers tend to increase in the spring

with a rise in temperature.















CHAPTER V
DIURNAL PATTERN IN OVIPOSITION AND HOST MORTALITY

Introduction

Zehnder and Trumble (1984b) showed that Liriomyza spp.

leafminers exhibited circadian patterns in flight activity and

larval emergence. Flight activity of leafminers peaked between

0700 and 1100 h. Most puparia and emerging larvae were also

collected in the same time period.

There have been no studies on the circadian patterns of

Diglyphus intermedius and patterns of only a few other

parasitic Hymenoptera have been studied. Sugimoto and Ishii (1979)

noted that daily host mortality from parasitization and host

feeding by Chrysocharis pentheus (Walker) fluctuated

periodically during its life when observed in the laboratory under

24 h light conditions. C. pentheus adults parasitized more

hosts during the first 6 h and fed on hosts more during the

subsequent 3 h in a 24 h period after the parasitoids had been

deprived of hosts for 1 day. Nyrop and Simmons (1986) trapped more

adults of Glypta fumiferanae (Viereck), a parasitoid that

attacks the spruce budworm Choristoneura fumiferana (Clemens),

during the 0800 to 2200 h than in any other time of a day.

If circadian acitivity patterns of D. intermedius were

known, sampling and monitoring for the leafminers and D.

intermedius could be optimized. Also, pesticides could be



77









78

applied at times that would conserve parasitoids and kill the

leafminers.

The experiments reported in this chapter were done to

determine the influence of the time of day on parasitoid

oviposition and parasitoid-induced host mortality.

Materials and Methods

Five-day-old females were utilized since oviposition and

parasitoid-induced host mortality rates peaked on the fifth day at

27.2 C (see chapter IV). Female parasitoids were isolated from the

colony upon emergence, paired with day-old males for 1 day, and

then placed singly in 150 x 15 mm petri dishes with leaflet pieces

containing 30 or more 3rd-instar L. trifolii larvae. Fresh

host larvae were provided every day. Females were removed at 0730 h

on the 5th day and placed in a new petri dish containing host

larvae.

For each female tested, 5 petri dishes were prepared. Each

petri dish had 20 leaflet pieces with each leaflet piece containing

a single 3rd-instar host larva. The petri dishes were held in a

refrigerator at 5 C to arrest leafminer development so that no

physiological age differences existed between host larvae offered

to the parasitoids at different times of the day. The petri dishes

were removed from the refrigerator and allowed to equalize to the

rearing room temperature for 5 minutes before being utilized for

the experiments.

The experiments were done in a rearing room maintained at

25-27 C. Fluorescent lighting inside the room was turned on at









79

0800 h and off at 2000 h. The parasitoids were transferred to

petri dishes with fresh host larvae at 0800 h, 1200 h, 1600 h and

2000 h. Fifteen females, 5 on 1 day, 4 on a 2nd day and 6 on a 3rd

day were observed. Killed hosts and parasitoid eggs were counted

in the manner described in Chapter IV.

Result and Discussion

The mean number of eggs laid/female in a 24 h period was 14.9

(Table 5-1). This is similar to 11.8 observed for 5-day-old

females at 27.2 C in the previous experiments (see Chapter IV).

The mean number of hosts killed/female was 14.0 compared to 17.5 as

described in chapter IV. The number of eggs laid/female varied

from 8 to 23 and the number of hosts killed/female varied from 11

to 24.


Table 5-1. Relationship between time of day and host mortality
and oviposition of 5-day-old female D. intermedius
at 25-27 C.(+ 1 SD).


TIME OF DAY OVIPOSITION HOST MORTALITY


0800-1200 10.3 (3.2) 7.9 (1.7)
1200-1600 2.1 (1.9) 2.4 (1.8)
1600-2000 2.4 (1.4) 3.6 (2.2)
2000-0800 0.1 (0.4) 0.1 (0.4)


TOTAL 14.9 (4.7) 14.0 (3.4)



Most of the parasitoid oviposition and parasitoid-induced host

mortality activity occurred between 0800 and 1200 h unlike the host

killing and ovipositional activity of C. pantheus which preferred









80

to oviposit in the first 6 hours and host-feed during the subsequent

3 hours in a 24-h period (Sugimoto and Ishii 1979).

The parasitoids were not active during the scotophase. Only

two of the females were even slighlty active at night. None of the

other parasitoids killed hosts or laid eggs at night. Parasitoids

were not moving and were usually found underneath the leaflet

pieces or away from the leaflets when the petri plates were removed

for observation at the end of the dark period. In either case, the

female parasitoids were at some distance from host larvae when

resting. They may, therefore, need vision to search and locate

hosts.

During the photophase, there was greater parasitoid

oviposition and parasitoid-induced host mortality during the 0800

to 1200 h compared to the 1200 to 1600h and the 1600 to 2000

period. The 5-day-old females had been previously conditioned in

the laboratory to have fresh host larvae available to them at the

beginning of the day and this may explain why more hosts were

killed during the first part of the photophase.

In the field, with most mature larvae emerging by 1100 h

(Zehnder and Trumble 1984b), it would be to the advantage of the

parasitoid to be most active during the early part of the day.

Charlton and Allen (1981) also showed that most mature L.

trifolii larvae exit from the leaves during the latter part of

the morning. Apparently, D. intermedius circadian behaviour

is linked to the timing of the emergence of mature host larvae from









81

the leaves. This type of behavior would enable the parasitoids to

utilize the most mature and, hence, the largest host larvae.

In addition, although presumably equivalent host larvae were

offered at 1200 and 1600 h, parasitoid oviposition and parasitoid-

induced host mortality activity were much reduced. Thus, laboratory

conditioning alone does not fully explain the observed behavior.

If chemical insecticides were to be utilized for leafminer

control, it may be best to do so during the period of least

parasitoid activity, i.e., during the night hours or in the

afternoon or evening. This would allow some adult parasitoids that

may be resting away from the plants to escape insecticidal

mortality. Spraying at these times may conserve parasitoids but

may also be less effective against leafminer adults since their

least active period coincides with that of D. intermedius.

Under such constraints, it may be advisable to use leafminer

larvicides. Schuster and Everett (1983) showed that cyromazine

(TrigardR) was an effective leafminer larvicide and that Trigard

residues on tomato were least lethal to leafminer adults.














CHAPTER VI
HOST STAGE PREFERENCE

Introduction

L. trifolii is attacked by different parasitoids in the

field, and it is important to know if any niche division occurs, at

least between the major parasitoids, so that competition among them

for hosts is understood. Lema and Poe (1979) showed that Opius

sp., a larval-pupal endoparasitoid and Chrysonotomyia sp., a

larval endoparasitoid, attacked 72-h-old and 120-h-old L. sativae

larvae, respectively. The hosts were reared at 20 + 7.9 C. The

L. trifolii larval stage lasts about 288 hours at 20 C (Leibee

1984). The larval period of L. sativae is probably comparable and

hence, 120-h-old larval L. sativae are likely to be in their

2nd-instar.

Hendrickson and Barth (1978) showed that D. intermedius

preferred to oviposit on 3rd-instar rather than 2nd-instar

Liriomyza trifoilearum Spencer larvae, although it fed on all

three instars under laboratory conditions. No parasitoids were

reared from 2nd-instar host larvae. Nearly 19% of field collected

larvae of the alfalfa blotch leafminer, Agromyza frontella

(Rondani), were parasitized by D. intermedius. This parasitoid

preferred 3rd-instar hosts although 23.7% of the parasitized hosts

were 2nd-instar larvae. Hendrickson and Barth (1978) attributed

the failure of D. intermedius in parasitizing 2nd-instar L.

trifoilearum larvae to the smaller host size.


82









83


The preference of a parasitoid for a particular host size is a

fairly common phenomenon. Hyposeter exigua (Viereck) laid 2.3

and 1.3 eggs per 24 h in 1st and 2nd instar, respectively, of

Trichoplusia ni (Hubner) larvae. Oviposition declined for each

subsequent instar (Smilowitz and Ivantsch 1975). Successful

parasitism by the braconid Cardiochiles nigriceps (Viereck) was

least in fifth-instar larvae of Heliothis virescens (F.) and

H. subflexa (Guernee) (Lewis and Vinson 1971).

The experiments reported here were done to test if different

aged L. trifolii larvae are equally acceptable to D. intermedius

for ovipositional and non-ovipositional purposes.

Materials and Methods

Leafminers and parasitoids were reared as previously described

in chapter III. Twenty 96-h-old L. trifolii larvae (referred

to as young larvae from hereon), and 20 120-h-old larvae (old

larvae), in cut tomato leaflet pieces, were arranged in each of 5

150x15 mm petri dishes containing water-moistened filter papers.

Five young larvae were alternated with 5 old larvae in each of the

radially sectored octants. Two additional plates, one with 20

3rd-instar larvae only, and another with 20 2nd-instar larvae only,

were also maintained. A single 5-day-old mated female parasitoid

was introduced into each petri dish. The adults were subsequently

removed after a period of less than 24 hours and the number of host

larvae killed and the number of parasitoid eggs laid were counted.

The experiment was replicated 4 times, once on each of 4 days. The









84

exposure period varied for the 4 replicates. The exposure time was

the same for each of the petri dishes placed on any particular day.

The experiments were done in the rearing room at the prevailing

conditions described in chapter V.

Results and Discussion

The results of the experiments are summarized in table

6-1. Significantly more old larvae were killed and parasitized



Table 6-1. Relationship between L. trifolii larva age
and mean no. of eggs laid and hosts killed by
D. intermedius (+ 1SD)


Choice or #Hosts Killed #Eggs Laid
No choice 96h 120h 96h 120h


Choice 3.2 (3.0) 6.0 (3.2) 1.3 (1.7) 5.8 (3.6)
No Choice 10.8 (6.6) 5.0 (5.3) 5.3 (5.1) 5.5 (5.5)



than were young larval hosts when the parasitoids were given a

choice (paired t-tests for pooled data, p >.001). Young larvae

were as acceptable as old larvae for oviposition and host mortality

was the same for larvae of both ages when the parasitoids did not

have a choice of different aged hosts.

D. intermedius could utilize synchronous as well as

asynchronous host populations under such circumstances. This fact

is important to recognize since the application of stage-specific

insecticides such as adulticides may tend to synchronize the

affected populations for at least a short period.









85


There may be long term effects on host and parasitoid

population densities with synchronous host populations, however.

If only small host larvae were present for a long period,

parasitoids emerging from small larvae may be smaller.

Trichogramma evanescens Westwood were smaller when reared on

smaller sized Sitrotoga cereallella (Olivier) host eggs (Salt

1940).

The advantage to D. intermedius of attacking larger hosts

can be understood if we consider the relationship between hosts and

all parasitoids of leafminers. Ectoparasitoids sting their hosts.

They cannot rely upon host growth to supply food for their

offspring. Endoparasitoids keep their hosts alive and can rely on

host growth for their offsprings' growth. Askew (1975) found no

early-attacking ectoparasitoids of Phyllonorycter spp. leafminers

while Syntomaspis spp. and Eurytoma spp. ectoparasitoids

attacked young hosts in oak galls. In oak galls, ectoparasiotids

could supplement their diet with gall material whereas there is no

provision for phytophagy in leafmines. Askew (1975) suggested

that, to a certain extent, the divergence in host selection, with

endoparasitoids attacking young leafminers and ectoparasitoids

attacking older leafminers, reduced the likelihood of ecto-

parasitoids attacking a host already containing an endoparasitoid.

The partitioning of hosts between ectoparasitoids and

endoparasitoids is not well defined, however. I have reared a

Chrysontomyia sp. adult from a field collected Diglyphus sp.









86
pupa. I have also observed D. intermedius parasitizing host

larvae that have already been parasitized by another individual

D. intermedius. This suggests that they may not be able to

distinguish between parasitized and nonparasitized host larvae or

may parasitize hosts even after recognizing the hosts as having

been parasitized by others. If this is true, they may just as

readily attack endoparasitized host larvae.

In summary, D. intermedius kills and oviposits on more

3rd-instar larvae than 2nd-instar laravae if both ages of larvae

are equally available to the parasitoid.













CHAPTER VII
IMPACT OF ALTERING HOST AND PARASITOID DENSITY
ON DAILY PARASITISM RATE

Introduction

The rate of parasitism in the Smerage et al. (1980) model was

a function of host and parasitoid densities. As mentioned earlier,

Figure 1-1 shows the assumed relationship between the rate of

parasitism and host and parasitoid densities at a constant

temperature. Equations (7) and (8) :


Y = C2.Kp.n2 (1 exp(n1/Kh)) (7)

Y = C2.Kp.n2 (nl/(n1 + Kl)) (8)


are alternative closed-form equations generating families of

curves, like Figure 1-1, as already mentioned in Chapter II.

In computer simulations of the model, Musgrave et al. (1980)

assumed Kh and Kp to be 5 larvae/m2 and 2 larvae/adult-day. The

experiments reported in this chapter were done to obtain a solid

description of the rate of parasitism as a function of leafminer

and parasitoid densities at a constant temperature. All experi-

ments were done in the laboratory and required the confinement

of parasitoids to cages in which leafminer infested plants were

kept.


Materials and Methods

Several limitations were encountered in the experimental

design. Ideally, observations on parasitoids should be made in a


87









88

large arena which allows the parasitoids to move naturally in and

out of the arena. This would enable the parasitoids to behave

naturally and to have random access to hosts, also, the likelihood

of encountering the same host again would not be due to confinement

to a restricted arena. Arena size would certainly be important if

the period of observation was large and parasitoids were confined

to a small region. Because of the small size of adult parasitoids,

it is not possible to monitor parasitoid densities in a large arena

with no confinement facilities, however. Nor is it possible to

maintain uniformity in leafminer and parasitoid densities between

observation plots. It is also not possible to maintain environ-

mental conditions constant.

To overcome these problems, a compromise was made. Female

parasitoids were confined in 67 x 67 x 67 cm cages within a

controlled environment room for 12 h observation periods. There

were several advantages to using cages. D. intermedius density

per cage was easy to record and manipulate, and other parasitoid

species were excluded. As the cages used were relatively small, it

was possible to keep them in a controlled environment room and

hence at constant temperature and uniform light conditions. One,

one-month-old greenhouse grown tomato plant, with approximately

2000 to 3000 sq cm leaf area, was placed in each cage. This amount

of leaf area was sufficient to sustain as many as 200 or more

leafminer larvae, and a wide range of leafminer densities was

possible. With cages, it was possible to alter leafminer densities









89

and maintain uniformity of larval age to a much greater extent than

if the experiments were done in the field.

Tomato host plants, cv. Hayslip, used in the experiments were

grown and maintained in the greenhouse as described in Chapter III.

The plants selected were approximatley 30 days post-transplant, 50

to 60 cm tall, had about 10 to 13 leaves, and were just beginning

to blossom. Plants of this size just fit into the 67 x 67 x 67 cm

cages and allow the maximum amount of leaf area possible on one

plant confined in the cage. The plants selected had approximately

2000 to 3000 sq cm leaf area, although on ocassions plants were

smaller or larger. Plants were moved from one holding area to

another with care to minimize the damage to leaves during handling.

After selection, 6 plants were moved to the refrigerated truck

body which housed the leafminer colony (see Chapter III). Three

tomato plants were exposed to leafminer adults confined in each of

two 67 x 67 x 67 cm cages. The number of leafminer adults confined

in the oviposition cage was varied from as few as 10 to as many as

50. The exposure period was also varied from as little as 1 minute

to 4 hours. It was possible to obtain various leafminer larval

densities by manipulating the number of adult flies and the

exposure period as just described. After exposure to leafminer

adults, the plants were moved to holding cages and maintained in

the manner described in Chapter III.

After 5 days, plants containing 3rd-instar larvae were removed

at 0800 h and taken to the insectary rearing room for exposure to

different densities of parasitoids. Of the 6 plants initially









90

exposed to the flies, 5 plants were selected for exposure to the

parasitoids. The plant most different from the others in height or

quality or leafminer density was discarded. One plant was placed

in each of five 67 x 67 x 67 cm cages.

The parasitoids used in these experiments were collected 4

days earlier on emergence and 40 females and 20 males were confined,

10 females and 5 males to a dish, in 150 X 15 mm petri dishes. Each

dish contained 40 to 60 3rd-instar leafminer larvae in tomato leaf-

lets and the larvae were replaced daily. On the 4th day, the

parasitoids were isolated singly in size 00 gelatin capsules and

sexed; males were discarded. One, 2, 3, 4, or 5 parasitoid females

were then released into each of the 5 cages holding the plants.

After 12 h plants were removed from the cages, shaken to dislodge

parasitoids and transferred to the laboratory adjoining the rearing

room at 2000 h.

Plant height, age, number of leaves and leaflets per plant

were counted. A LicorR leaf area meter was utilized to record

the total leaf area, the area of leaflets containing mines, and the

area of leaflets containing parasitized mines. A record was kept

of the number of empty, live and parasitized mines in each of 3

leaf categories. Leaves were classified as old (usually the 1st 3

leaves which showed signs of yellowing), fully expanded leaves (the

majority of the leaves) and unexpanded leaves (usually the top 3 to

leaves although occasionally there may have been several more in a

small, tight bundle at the plant apex). Mines containing paralyzed









91

larvae were dissected and the number of eggs laid on each larva was

recorded.

The experiment was repeated several times. The assignment of

parasitoids to the cages was rotated so that the same cage had the

same parasitoid density every 5th time. Because the plants varied

in leaf area, and leafminer oviposition could not be made uniform

on each plant, it was not possible to regulate leafminer density as

freely as parasitoid density. The number of mines on each plant,

and hence in each cage, could have been kept the same by destroying

some larvae on each plant. However, this would not have been

appropriate for 2 reasons. Parasitoids searching a greater leaf

area with the same number of mines/cage would take longer than for

the same number of mines on a smaller leaf area. Also, destroying

leafminers may elicit its own response from the parasitoids if the

parasitoids cue in to host plant damage in locating leafminers.

The SAS NLIN pocedure (SAS 1982) was utilized to decribe the

relationships between leafminer density and rate of parasitism at

different parasitoid densities. The same procedure was also

utilized to describe the relationship between leafminer density and

parasitoid-induced host mortality at different parasitoid densities.

Chi square (X2) tests were done to determine if there was any

effect of altering either leafminer or parasitoid density on the

numbers of parasitoid eggs on paralyzed host larvae.

Results and Discussion

Two rate of parasitism models, the rectangular hyperbola and

the negative exponential equations proposed earlier by Smerage et









92

al. (1908), fit equally well the observed values of rates of para-

sitism when plotted against leafminer densities in the 0.00-0.06

larvae/sq cm range. The exponential equation obtained using the SAS

nonlinear regression techniques was


Y = Kp.n2.(l exp(-n1/Kh)) (r2 = .79) (T = 25-27 C) (4)


where Y, the rate of parasitism, is the number of hosts parasitized/

day. With a value of 7.3908, Kp is a constant with the units,

number of hosts parasitized/parasitoid-day. Densities, n2 and n1

are, respectively, parasitoids/cage and leafminer larvae/leaf area

(cm2). Another constant, Kh, istleafminer hosts/leaf area (cm2).

Results of the experiments also were utilized to formulate the

similar but parametrically different equation (5) below expressing

parasitoid-induced host mortality as a function of host and para-

sitoid densities at 25-27 C:


Z = Cp.n2(1 -exp(-nl/Ch)) (r2 = .78) (T = 25-27 C) (5)


where Z, the parasitoid-induced host mortality rate, is the number

of hosts killed/day. With a value of 9.2064, Cp is a constant

with the units number of hosts killed/parasitoid-day and Ch is

0.0165 leafminer larvae/leaf area (cm2).

The predicted relationship between host density, parasitoid

density and the rate of parasitism utilizing the negative expo-

nential equation is depicted in Figure 7-1. The relationship

between rate of parasitoid-induced host mortality and host and




Full Text

PAGE 1

PARASITIZATION OF Liriomyza trifolii (Burgess) BY Diglyphus intermedius (Girault) By KIRTIKUMAR JASHBHAI PATEL 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 1987

PAGE 2

ACKNOWLEDGEMENTS My sincerest thanks to Dr. David J. Schuster, chairman of the supervisory committee for his financial support, guidance and advice I express my gratitude for the help given to me by Dr. Smerage and Dr. Kerr while they served on the supervisory committee. I am indebted to the late Dr. Sailer who had also served on the supervisory committee and for motivating me in my career. I would like to thank Dr. Portier and his graduate research assistants for making helpful suggestions for the statistical analysis of the experimental results. The staff at the Gulf Coast Research Station helped me in many ways while I was at the station doing the experiments. Their help is gratefully acknowleged. My colleagues and friends provided me with companionship and moral support during my graduate years. Their friendship will always be cherished. My family provided me with financial and moral support. I am grateful for the help I have received from them. Finally, my wife Falguni has been patient with me while I was busy with my studies and I wish to thank her for that. ii

PAGE 3

TABLE OF CONTENTS Pa^e ACKNOWLEDGEMENTS ii LIST OF TABLES v LIST OF FIGURES vi ABSTRACT vii CHAPTER I INTRODUCTION 1 II LITERATURE REVIEW 13 Tomato Production in Florida 13 Importance 13 Production Methods Relevant to Pest Management 14 Liriomyza trifolii (Burgess) 22 Description 22 Taxonomy 24 Distribution 24 Host Plant Range 25 Life History 34 Chemical Control of Leafminers on Tomato.... 39 Biological Control of Leafminers on Tomato.. 41 Diglyphus intermedius (Girault) 47 Taxonomic Status, Hosts and Distribution.... 47 Adult External Morphology 47 Parasitoid Abundance in Florida 48 Biology and Life History 48 Modeling Leaf miner-Parasitoid System 50 Introduction 50 The Smerage et al. Leaf miner-Parasitoid Model 51 III PLANT PRODUCTION AND REARING TECHNIQUES FOR Liriomyza trifolii (Burgess) AND Diglyphus intermedius (Girault) ON TOMATO 57 Introduction 57 Host PLant Production 57 Rearing L_. trifolii 59 Rearing D_. intermedius 62 iii

PAGE 4

Page IV FECUNDITY AND LONGEVITY OF Diglyphus intermedius (Girault) AND ITS CONTRIBUTION TO HOST MORTALITY 64 Introduction 64 Materials and Methods 65 Results and Discussion 66 V DIURNAL PATTERN IN OVIPOSITION AND HOST MORTALITY 77 Introduction 77 Materials and Methods 78 Results and Discussion 79 VI HOST STAGE PREFERENCE 82 Introduction 82 Materials and Methods 83 Results and Discussion 84 VII IMPACT OF ALTERING HOST AND PARASITOID DENSITY ON DAILY PARASITISM RATE 87 Introduction 87 Materials and Methods 87 Results and Discussion 91 VIII MULTIPLE OVIPOSITION AND ITS INFLUENCE ON PROGENY SURVIVAL RATE 98 Introduction 98 Materials and Methods 98 Results and Discussion 99 IX CONCLUSION 101 LITERATURE CITED 110 BIOGRAPHICAL SKETCH 124 iv

PAGE 5

LIST OF TABLES Page Table 2-1 Current known distribution of L. trif olii (Burgess) 26 Table 2-2 Host plant range of L_. trif olii 28 Table 2-3 Developmental times for L. trif olii immature stages on some crop plants 37 Table 4-1 Number of L. trif olii larvae killed by, and the fecundity and longevity of I), intermedius at different constant temperatures (+ 1SD) 66 Table 5-1 Relationship between time of day and host mortality and oviposition of 5-day old female D. intermedius at 25-27 C (+ 1SD) 79 Table 6-1 Relationship between L_. trif olii larva age and mean no. of eggs laid and hosts killed by D. intermedius (+ 1SD) 84 Table 7-1 Proportion of observations with 100% parasitization of killed hosts at different leafminer densities 95 Table 7-2 Frequency distribution of killed L. trif olii larvae in classes of different JJ. intermedius egg densities 96 Table 8-1 Survival of D. intermedius eggs to adult stage when placed at different densities on L. trif olii larvae 100 Table 9-1 Predicted number of parasitoid eggs reaching the adult stage for every 100 hosts killed 10 6 v

PAGE 6

LIST OF FIGURES Page Figure 1-1 Typical functional description of a single host-single parasitoid parasitism 4 Figure 2-1 Conceptual model of the leaf miner-parasitoid system from Smerage et al. 1980 52 Figure 4-1 Relationship between J), intermedins fecundity, F, and temperature, T 68 Figure 4-2 Relationship between I), intermedius longevity, L, and temperature, T 69 Figure 4-3 Relationship between # hosts killed, Hm, and temperature, T 70 Figure 4-4 Mean daily J), inter medius -induced host mortality rates and J), intermedius oviposition at different temperatures 72 Figure 4-5 Influence of temperature on the number of eggs laid daily by surviving J). intermedius females 74 Figure 4-6 Influence of temperature on the number of hosts killed by surviving J). intermedius females 75 Figure 7-1 Relationship between host density, n, and parasitism rates at different parasitoid densities, 93 Figure 7-2 Relationship between host density, n, and parasitoid-induced host mortality rate at different parasitoid densities, n„ 94 vi

PAGE 7

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 PARASITIZATION OF Liriomyza trifolii (Burgess) BY Diglyphus intermedins (Girault) By Kirtikumar Jashbhai Patel May 1987 Chairman: D. J. Schuster Major Department: Entomology and Nematology Diglyphus intermedius (Girault) is a commonly encountered parasitoid of Liriomyza species leafminers which are considered important pests of tomato and several other commercially grown crops in Florida. An understanding of the leaf miner-parasitoid interactions is important to any sound, integrated pest management system on tomato. Those aspects of the biology of Diglyphus intermedius that influence its parasitization of Liriomyza trifolii (Burgess), reared on tomato, were determined from laboratory experiments to obtain information on L. trif olii -D. intermedius interactions. The relationship between T, temperature in degrees Celsius, and parasitoid fecundity, F, longevity, L, and parasitoid-induced host mortality, Hm, were measured at 15.6, 19.4, 23.3, 27.2, and 31.1 C and are described by the equations: F = -196.11 + 42.65T 1.1T 2 (r 2 = .65), L = 74.32 2.09T (r 2 = .94), and Hm = 721.97 19. IT (r 2 = .83) vii

PAGE 8

The daily parasitism rate was determined to be a function of leafminer and parasitoid densities and is expressed by the equation: Y = Kp.n2(l -expC-nj/Kh) where Y is the number of hosts parasitized/ day, Kp is 7.3908 hosts parasitized/parasitoid-day and Kh is 0.0144 leafminer larvae/sq cm leaf area. Leafminer density, n^, ranged from 0 to 0.06 larvae/sq cm leaf area. Parasitoid density, 1X2$ ranged from 1 to 5/cage. The rate of daily host mortality is described by the equation Z = Cp.n2.(l -exp(-n2/Ch)) where Z is the number of hosts killed/ day, Cp is 9.2064 hosts killed/parasitoid-day and Ch is 0.0165 leafminer larvae/sq cm leaf area. I), in termedi us -induced host mortality and parasitization were highest in the first 4 h of a 24 h period, with 12 h of light followed by 12 h of dark. More 120-h-old larvae were parasitized and killed than were 96-h-old larvae when both ages of larvae were equally available to a female parasitoid. An average 0.8, 1.1, 1.35 and 0.98 parasitoid eggs became adults when 1, 2, 3, and 4 eggs were placed on each host larva, respectively. viii

PAGE 9

CHAPTER I INTRODUCTION The use of broad-spectrum insecticides on a large scale in the 1940s has led to the emergence of Liriomyza species leafrainers as important pests of tomato, chrysanthemum, celery and many other commercial crops grown in Florida (Schuster 1981). The worldwide spread of Liriomyza trif olii (Burgess) in the past decade, in addition, made it an international pest (Poe and Montz 1981a). The balance between leafminers and their natural enemies has presumably been upset by the use of broad-spectrum insecticides under Florida conditions. New or previously unused insecticides have been introduced whenever contemporary ones became ineffective (Leibee 1981b), but these have provided only short term relief. The period of effective use of a newly introduced, broad-spectrum insecticide has usually been followed by loss of control. The exact mechanisms of the breakdown in control are not known. Damage due to insecticide-induced outbreaks of leafminers can be reduced greatly if the insecticide is selected carefully along with population monitoring and the use of economic thresholds (Waddill 1981). Only recently have host specific insecticides become available for leafminers (Robb and Parrella 1984). Some of these chemicals are still awaiting clearance for commercial use on tomato. (Fluker personal communication 1986). As a result of the failure of long term control of the pest with insecticides, 1

PAGE 10

2 growers have embraced the current philosophy of integrated pest management (IPM) approach (Poe 1985). Action thresholds have been established for leafminers on tomato in Florida (Pohronezny and Waddill 1978, Schuster et al. 1980). The role that the natural enemies play in determining host population densities has subsequently become crucial to any IPM strategy for leafminer regulation. For example, only live leafrainer larvae are counted in determining if leafminers have reached the threshold level that requires insecticide application. The role of natural enemies is recognized and taken into consideration by excluding larvae killed by parasitoids. As a further refinement of IPM tactics, many current research programs are directed toward determining the role of natural enemies in regulating leafminer populations to make IPM programs more robust (Parrella et al. 1985, Schuster 1985, Zehnder and Trumble 1985, and Lindquist and Casey 1985). Leafminer-parasitoid interaction was a focus of the leafrainer population dynamics model for celery proposed by Smerage et al. (1980) and provides an excellent framework for the elucidation of the role of natural enemies in regulating leafminer populations. Their model pertained to within-field populations of eggs, larvae, pupae and adults of leafminers and its parasitoids, broadly expressed as processes that contribute to the overall dynamics of the model. Other processes, such as parasitism, were also incorporated into the model.

PAGE 11

3 One hypothesis about the rate of parasitism is that it is a function of host and parasitoid densities. Holling (1959), for example stated that parasitism rate is directly dependent on host and parasitoid densities. An increase in either density led to higher parasitism rate. This hypothesis motivated Smerage et al. (1980) to utilize the family of curves depicted in Figure 1-1 to illustrate the general relationship between the host and parasitoid densities and the number of hosts parasitized/day. The scale of the family of curves was assumed by Smerage et al. (1980) to fluctuate with temperature. Musgrave et al. (1980) made some computer simulations of the Smerage et al. (1980) model. Process and parameter descriptions were derived from hypothetical values and preliminary estimates. The results of the simulations were encouraging enough to warrant further research to obtain more complete data to improve the model. Several workers in Florida are striving for more complete quantitative descriptions of the biologies of leafminers and its parasitoids. This includes research in better estimates of parameters. The ultimate aim of this research is to incorporate a leafrainer model into IPM strategies. As results become available, they may be incorporated into the Smerage et al. (1980) model. An alternative model may become necessary if the new information can not be readily incorporated into the existing model without major structural changes.

PAGE 12

4K, Host Density Figure 1-1. Typical functional description of a single hostsingle parasitoid parasitism (figure obtained from Sraerage 1980).

PAGE 13

The specific needs being addressed currently are 1) complete quantitative information about the biologies of leafminers and their most common parasitoids and 2) sampling procedures to estimate populations of the leafminers and its major parasitoids. A major thrust of research efforts has been the measurement of fecundity, longevity and development of leafminers and its parasitoids on different crop and weed hosts and at different temperatures. Behavioral phenomena, such as peak periods of ovipositional and feeding activities, have been reported to follow circadian patterns. These are also being studied for the leafminer and its parasitoids Another focal point of leafminer research is leafminerparasitoid interactions. In particular, the extent to which leafminer and parasitoid densities determine the rate of leafminer parasitism is being investigated. Parasitization of a host refers to the act of utilizing a host for oviposition. Parasitoids may also kill hosts without ovipositing on them. Therefore, the effect of host and parasitoid density on the rate of parasitoid-induced host mortality is also being investigated. The terra parasitoidinduced host mortality is from hereon used to refer to the sum of hosts killed by parasitoids for ovipositional and non-ovipositional purposes. In other words, it is the number of successful attacks by parasitoids that lead to death of hosts. More than one egg of a parasitoid may be deposited when a host is parasitized. Multiple oviposition may lead to reduced survivorship of the progeny of parasitoids. In addition, not all

PAGE 14

6 host larvae may be equally acceptable to parasitoids for ovipositional or non-ovipositional purposes. Therefore, there are current investigations of multiple oviposition and parasitoid preference for a particular size of leafminer larvae. The objectives of the research presented here were 1) To determine the rates of parasitism and parasitoid-induced host mortality as functions of leafminer and parasitoid densities, with temperature being kept constant. 2) To determine the effect of temperature on parasitoid fecundity, female parasitoid longevity and parasitoid-induced host mortality 3) To determine the role of time of day on parasitoid oviposition and parasitoid-induced host mortality rates during different times within a day. 4) To determine the influence of host age on parasitoid oviposition and parasitoid-induced host mortality. 5) To determine the influence of multiple oviposition by the parasitoid on the survivorship of the parasitoid progeny. In the Smerage et al. model (1980), all parasitoid species were aggregated into a single equivalent parasitoid population to simplify the model. There are many parasitoids that attack Liriomyza trifolii however. In addition, the biologies of

PAGE 15

7 these different species vary. For the research presented here it was possible to investigate only one parasitoid species, Diglyphus intermedins (Girault), with the resources and time available I), intermedins was chosen as a representative natural enemy of the leafminers for two reasons. I), intermedius is one of the three most common parasitoids reared from leafminers on tomato in Florida (the other two are Chrysonotomyia sp. (Eulophidae) and Opius sp. (Braconidae) ) (Schuster 1985). _D. intermedius is an ectoparasitoid laying its eggs on or near the host larva (Hendrickson and Barth 1978). This makes I). intermedius easier to observe than either Chrysonotomyia sp. or Opius sp., which are endoparasitoids (Lema 1976). Before initiating any experimental work, the literature on leafminer and J), intermedius was reviewed (see Chapter II). To study the biology of the parasitoid and its interaction with leafminers, experiments should ideally be done in the field. It is not possible to vary a single factor and keep all others constant in the field, however. All experiments were, therefore, done in the laboratory. L. trifolii and J3. intermedius were initially collected from tomato fields and subsequently reared on tomato in controlled environmental conditions to ensure an adequate supply for use in the experiments. The insect rearing techniques are described in Chapter III. The first series of replicated experiments determined the influence of temperature on D. intermedius fecundity, longevity

PAGE 16

8 and I), intermediusinduced host mortality (Chapter IV). Hendrickson and Barth (1978) had shown that I), intermedins usually oviposited on 3rd-instar hosts although it killed all three instars. Leafminer larvae in their 3rd-instar were kept in petri dishes and placed in 5 incubators each set at a different constant temperature in the 15.6 to 31.1 C range. A pair of parasitoids was introduced into each petri dish. Killed hosts and parasitized hosts were counted daily. Leafminer larvae from the previous day were replaced with more 3rd-instar larvae. The petri plates were maintained until the female parasitoids died. In the temperature range studied, the temperature dependence of F, the average fecundity per adult female lifespan, was described by the equation: F = 196.11 + 42.65T 1.1T 2 (r 2 = .65) (1) When the number of hosts was a non-limiting resource, the temperature dependence of Hm, the average number of hosts killed/adult female lifespan, was described by the equation: Hm = 721.97 19. IT (r 2 = .83) (2) Finally, the temperature dependence of L, the average female parasitoid longevity in days, was described by the equation: L 74.32 2.09T (r 2 = .94) (3)

PAGE 17

9 In these and all subsequent equations, T denotes temperature in degrees Celsius. As a female parasitoid aged, daily parasitism and parasitoid-induced host mortality rates declined. Temperature determined the adult female parasitoid age at which oviposition and parasitoid-induced host mortality rates peak. With an increase in temperature, these peaks are reached earlier. Experiments to determine the effect of time of day on parasitoid-induced host mortality and parasitization were done in a controlled environment room (Chapter V). The room was maintained at 25-27 C. Fluorescent lighting was turned on at 0800 h and off at 2000 h. A 5-day-old female parasitoid female was introduced into a petri dish containing 20 3rd-instar host larvae. Every 4 hours, for the first 12 h of a 24 h day, host larvae were replaced with 20 more 3rd-instar larvae. The numbers of hosts killed and parasitized were counted for each time period. I), intermedius -induced host mortality and parasitization were highest between 0800 and 1200 h. There was an almost complete absence of parasitism and parasitoid-induced host mortality in the absence of light. The data obtained in this experiment could be utilized in determining, for example, when to spray during the day to produce the least effect on parasitoid densities. Whether JJ. intermedius preferred 3rd-instar larvae or 2ndinstar larvae for oviposition was also determined (Chapter VI). An equal number of 2ndand 3rd-instar host larvae were made available to females singly confined to petri dishes. Significantly more 3rdinstar larvae were parasitized and killed than were 2nd-instar larvae.

PAGE 18

10 Experiments done to determine rates of parasitism and parasitoid-induced host mortality, as functions of host and parasitoid densities, are reported in Chapter VII. Female I). intermedins along with tomato plants containing 3rd-instar leafminer larvae, were confined to cages in a controlled environment room. The temperature in the room was maintained within the 25-27 C range. Each cage contained 1, 2, 3, 4, or 5 4-day-old female parasitoids. The density of leafminer larvae was altered by exposing clean tomato plants to different densities of leafminer adults for different time intervals. At the end of 12 h of exposure to parasitoids, plants were removed, leaf area was measured, and the numbers of hosts parasitized and killed were recorded. The data obtained from the experiments were utilized to obtain estimates of parameters permitting description of the relationship between the rate of parasitism and the densities of L. trif olii and J), intermedius by the equation: Y = Kp.n 2 (l -exp(ni /Kh) (r 2 = .79) (T = 25-27C) (4) where Y, the parasitism rate, is the number of hosts parasitized per day. In (4), Kp, with a value of 7.3908, is a constant with the units number of hosts parasitized/parasitoid-day Densities n 2 and Dj are, respectively, parasitoids/cage and leafminers/

PAGE 19

11 2 leaf area (cm ) and Kh, another constant, is 0.0144 leafminer hosts/leaf area. Results of the experiments also were utilized to formulate the similar but parametrically different equation (5) below that expresses parasitoid-induced host mortality as a function of host and parasitoid densities at 25-27C: Z = Cp.n 2 (l -expC-nj/Ch) (r 2 = .78) (T = 25-27) (5) where Z, the parasitoid-induced host mortality rate, is the number of hosts killed per day. With a value of 9.2064, Cp is a constant with the units number of hosts killed/parasitoid-day and Ch is 0.0165 leafminer hosts/leaf area. The effect of multiple oviposition on survival of progeny is important in estimating the parasitoid density in the following generation. To study this effect, parasitoid eggs were kept at 4 different densities on host larvae in the experiment reported in Chapter VIII. An average 0.8, 1.1, 1.35 and 0.98 eggs became adults when 1, 2, 3, and 4 eggs, respectively, were kept on each host larva. The results of all experiments reported in this dissertation are summarized in Chapter IX. In this final chapter, I have also hypothetized that, Ke, the number of parasitoid eggs reaching the adult stage/100 host larvae killed may be described by the equation: Ke = 93 -800n, f ^

PAGE 20

12 In (6), is the number of leaf miners/leaf area (sq cm). The relationship would be true at 25-27 C temperature range only since survivorsip of eggs to adult stage was not reported at other temperatures. This equation was obtained from the observed distribution of parasitoid eggs densities on host larvae at different host densities as reported in Chapter VII, and the observed survival rates of different densities of parasitoid eggs on host larvae as reported in Chapter VIII. Although it is not a substitute for a comprehensive leaf miner-parasitoid model which has provisions for stage specific mortality, Ke provides a crude estimate of the number of parasitoids expected in the next generation. This could be a useful tool in the field if it is necessary to obtain a quick estimate of the density of parasitoids to be expected in the next generation. In addition to summarizing the results and suggesting the utility of Ke, I have also suggested that the classical biological approach should not be neglected. Exploration of other regions for leafminer parasitoids not found in Florida, and subsequent importation and establishment of these in Florida should also be seriously considered in the tomato IPM program.

PAGE 21

CHAPTER II LITERATURE REVIEW Tomato Production in Florida Importance In monetary terms, the tomato is the most important vegetable crop grown in Florida (Cantliffe 1985). Total value of the crop increased from $122.3 million in the 1973/74 season to $390.6 million in the 1982/83 season (Van Sickle and Belibasis 1985). Tomato acreage has increased from 31,500 acres in the 1974/75 season to 47,600 acres in the 1983/84 season. Manatee and Hillsborough counties and adjacent areas accounted for 17,500 harvested acres in the 1983/84 season while Dade county harvested 12,800 acres in the same season and Collier and Hendry counties harvested 9,375 acres. These three regions accounted for 40,075 acres of the 47,600 total acres harvested in Florida (Van Sickle and Belibasis 1985). The yield of tomatoes has increased from an average 796 25-pound boxes per acre in the 1973/74 season to an average 1,250 boxes in the 1981/82 season (Van Sickle and Belibasis 1985). An important aspect of tomato production in Southern Florida is the perceived competition from West Mexico producers for the U.S. domestic market. Both regions supply a large part of the Eastern U.S. winter demand for fresh tomatoes. Van Sickle and Belibasis (1985) have shown that Florida's share of the tomato market has increased over the past 10 years, however. The 13

PAGE 22

14 widespread use of hybrid varieties such as 'FTE-12', x Duke' and *Sunny' has contributed to Florida's corapetetive edge over West Mexico, according to those authors. These hybrids have higher yields, firmer fruits and concentrated production compared to the traditional varieties. Fields are now picked only 2 or 3 times, whereas the previous practice was to pick as many as 5 times. In addition, the cost of marketing and shipping Florida tomatoes is less than that for Mexican tomatoes (Van Sickle and Belibasis 1985). The competitive edge has also been sustained by keeping down the costs of pest control. Twice weekly sprays just to control leafminers was the recommendation to growers by the University of Florida's Institute of Food and Agricultural Sciences (IFAS) in the 1960's (Brogdon et al. 1970). Spray applications by a typical grower for control of all pests were reduced to 14 per season by 1980 (Prevatt, personal communication). The costs (in 1978/1979 prices) of pest control ranged from $228.79 to $438.20 per acre in 1978/1979 season (Brooke 1980). In 1984/85, these costs (in 1984/1985 prices) ranged from $453.43 to $501.96 per acre (Van Sickle and Belibasis 1985). Production Methods Relevant to Pest Management There are two tomato growing seasons a year in the ManateeHillsborough area of Florida. The 110-day fall season starts in late August and ends in December. The 120-day winter season lasts from January to April. In the Dade, Collier and Hendry counties,

PAGE 23

15 tomatoes are grown once a year in a growing season extending from approximately November to February. In northern regions of the state, tomatoes are grown during the spring months. Pest management decisions are made throughout the growing season. These decisions are discussed below. Land selection Most growers use the same field only in alternate or every fourth season for growing tomato. Some growers avoid using fields with a previous history of consistently bad yields attributable to soil pathogens or nematodes (personal communications) Land preparation Rotovating, forming raised beds, preparing irrigation and drainage ditches, fertilizing, fumigating and covering with stripor full-bed plastic mulch are the cultural practices commonly employed in preparing the land for planting. The soil pH should ideally be maintained in the 6.0 to 6.5 range (Hochmuth 1985). Crop losses due to Fusarium wilt ( Fusarium oxysporum f S P lycopersici (Saccardo) Snyder and Hensen races 1, 2, and 3), Fusarium crown rot (F. oxysporum f sp. radicis-lycopersici Jarvis & Shoemaker), southern blight ( Sclerotinium rolfsii Saccardo), and and many nematode-incited diseases can be reduced by adjusting the pH to within the 6.5 to 7.5 range (Jones and Overman 1985). Other soil pathogens are not easily managed by simply regulating the soil pH. These organisms include Verticillium wilt (Verticillium albo-atrum Reinke and Berthold races 1 and 2), the damping-of f disease caused by Rhizoctonia solani Kuhn and Pythium spp., Pyrenochaeta brown-rot ( Pyrenochaeta lycopersici Schneider and

PAGE 24

16 Geralach) and bacterial wilt ( Pseudomonas solanacearum E. F. Smith). These organisms are controlled with soil fumigants (Jones and Overman 1985). Hochmuth (1985) recommends that at least 50% of the nitrogen should be in the nitrate form. Blossom-end rot and soil conditions favorable to fusarium wilt can occur as a result of reduced calcium uptake if much of the nitrogen is in the ammoniacal form. He also points out that some pesticides contain elements that are essential micro-nutrients for the tomato. These micronutrients should not be given separately if previous soil analyses indicated that they were lacking and if pesticides containing the micronutrients are to be used. Toxicity could occur as a result of build up of some elements. Excess copper could induce iron deficiency, for example. Tomato yields can be increased greatly by reducing the diseases caused by soil-borne pathogens by using other methods in addition to proper liming and fertilizing regimens. This involves the use of grower-imposed quarantines, exclusion of infected materials, removal and incineration of crop residues, the use of resistant cultivars, and soil fumigation. Resistant cultivars 'Sunny', 'Duke', 'FTE 12', 'Hayslip', and *Floradade' tomato cultivars accounted for 96% of the total acreage in Florida in the 1984/85 season (Hawkins 1985). Most cultivars in current use in Florida are resistant to race 1 verticillium wilt but not to race 2 (Jones and Overman 1985). They are also resistant to F. oxysporum f sp. lycopersici races 1 and 2

PAGE 25

17 and to gray leaf spot ( Stemphylium solani Weber) (Maynard 1986). Scott (1985) has shown true resistance to Fusariura wilt race 3 in LA716 ( Lycopersicon penellii (Correll)D' Arcy) accession. Advanced testing of lines resistant to bacterial spot ( Xanthomonas campestris pv. vesicatoria (Doidge) Dye), bacterial wilt ( Pseudomonas solanacearum E. F. Smith) or fusarium crown-rot should occur in the next two years (Scott 1985). Soil fumigation It is essential to use broad-spectrum fumigants even when all the other tactics mentioned above are used to control the soil-borne pathogens. The fumigants include methylbromide + chloropicrin, chloropicrin + nematicides, and methylisothiocyanate + nematicides (Jones and Overman 1985). Weed control Nightshade ( Solanum americanum Mill.) is the most important weed in southwest Florida (Gilreath 1985). It belongs to the family Solanaceae as does the tomato. Because of the close genetic relationship between the two plant species, tomato would likely be injured if herbicides were used to control nightshade. Herbicides can be used to control nightshade if the tomato is grown on mulched, seepage irrigated beds and the spray is directed to the row middle so that it does not contact the tomato plant (Gilreath 1985). Gilreath (1985) found that paraquat (Paraquat R ) R R metribuzin (Sencor /Lexone ), and other labeled herbicides will control nightshade if applied to nightshade in the 2 to 4 true leaf stage of development. Older plants and those hardened by cold or other factors would not be adequately controlled by labeled hebicides. Gilreath (1985) has evaluated several new currently

PAGE 26

18 unregistered herbicides for their efficacy against nightshade and other weeds in tomato. He found that a combination of oxyfluorfen R R (Goal ) and flauzifop (Fusilade 2000 ) provided the best control of many weeds, including nightshade and grasses, and that there was a rapid kill of emerged weeds. It is his hope that Goal will be registered for use in mulched tomato middle in the next two years. Other weeds of tomato include the nutsedge ( Cyperus spp.), ragweed ( Ambrosia artemesif olia L.), Medicago spp., common beggartick (Bidens pilosa L.), and downy ground cherry Physalis pubescens L.). Almost all weeds other than nightshade are easily controlled by currently registered herbicides. Nutsedge is controlled by methylbromide, for example (Dunn 1985). Foliar disease management Regular maneb/mancozeb and copper applications are made to control bacterial infections. They have to be used carefully. Certain copper and mancozeb combinations may lead to higher incidence of target spot (Corynespora cassiicola (Berk. & Curt)), than the use of chlorothalonil (Jones and Jones 1984). Conover and Gerhold (1981) showed that raaneb or mancozeb alone were more effective than a combination of basic copper sulfate with either maneb or mancozeb in controlling late blight ( Phytophthora infestans (Montagne) de Bary) and gray leaf spot. Metalaxyl, which is now registered for use on tomato, gives excellent control of the tomato late blight, and related fungi (Pohronezny 1985). Samoucha and Cohen (1984)

PAGE 27

19 have shown a high degree of cross-resistance to other compounds by fungus biotypes that have resistance to metalaxyl, however. A weather-based system, BLITECAST, predicts outbreaks of late blight and is used in many parts of the country. It has not been very useful under Florida conditions, however. First outbreaks of late blight occur much later than predicted by BLITECAST (Pohronezny 1985). Insect pest management The tomato is damaged by direct and indirect insect pests. Direct pests attack the marketed product, the fruit. The armyworms ( Spodoptera eridania (Cramer), S^. exigua Hubner), and S. dolichos (F.)), the tomato pinworm Keif eria lycopersicella (Walsinghara) ) the tomato fruitworm ( Heliothis zea (Boddie)), the tobacco budworm (H. virescens (F.)), and the southern green stink bug ( Nezara viridula (L.)) are some of the direct insect pests of tomato. Indirect insect pests of tomato include leafminers as well as most of the pests listed above when they damage plant parts other than the fruit. Defoliation could result in reduced photosynthetic activity and reduced fruit size. The sun may scorch fruit on the vine if leaf shade is removed by defoliation (Musgrave et al. 1975). Pathogens may exploit damaged parts for invasion of the plant. Alternaria alternata (Fries) Keisler, which is weakly parasitic, and Xanthomonas campestris pv. vesicatoria enter leafmines to invade the tomato (Keularts 1980). Chemical control Several effective insecticides are available and in current use by growers (Johnson 1985). Permethrin (Ambush R ),

PAGE 28

20 R R methomyl (Lannate ), endosulfan (Thiodan ), and raetharaidophos (Monitor ) are some of the popular insecticides for control of R R tomato insect pests. Avermectin (Avid ) and cyromazine (Trigard ) are new insecticides that are effective against leafminers (Schuster and Everett 1983), but they are currently not registered for use against leafminers on tomato in Florida. The effect of insecticides on leafminers and their natural enemies is described in detail in the section on leafminer control in this chapter. Integrated pest management There are essentially two types of strategies for insect management available to growers on tomato. Some growers treat their crops when needed. Other growers spray the crop on a regular basis whether they know pests are present in the field or not. Both strategists employ resistant cultivars, intensive pre-plant treatment of soil with nutrients and multi-purpose fumigation, hebicides and other practices such as staking and mulching. Growers who treat their crops for insects only when needed, make their decisions on recommendations from professional scouting consultants or use their own judgement. The employment of professional scouts is now widespread (Pohronezny 1985). Integrated Pest Management (IPM) programs were initiated in the Homestead region in 1976 (Pohronezny and Waddill 1978) and, with modifications, in the Manatee-Ruskin area in 1978 (Schuster et al. 1980). Funding for the programs was from state and federal sources but scouts were encouraged to form private firms and solicit

PAGE 29

21 contracts from the growers. There are now at least two tomato IPM consultant companies in the Manatee-Ruskin area, and one in the Homestead region. Sampling methods remain very similar to those initially developed hy Phoronezny and Waddill (1978). Scouts in both areas spend most of their time in the field sampling leafminers and pinworms. All other scouting activity is secondary in that observations of other pests are made at the same sampling stations and while walking from one station to the next. Schuster et al. (1980) described the procedure, technique and leafrainer thresholds for the Manatee-Ruskin area. Each field is scouted twice weekly throughout the growing season. More stations are allocated to field boundaries than to the middle of a field, as pinworm and leafminer populations seem to be distributed in clumps, with higher densities of these insects being found near boundaries. There is one sampling station for every 2.5 acres of a field. Plants are sampled for leafminers, pinworms, and other pests. Live leafminer larvae are now counted instead of the earlier method of counting total mines in deciding whether or not leafminers have reached a population level that required chemical treatment. In considering only live larvae and excluding larvae killed by parasitoids, the role of natural enemies in regulating leafminer populations is recognized (Pohronezny et al. 1984). The threshold for treating lepidopteran pests has been altered from 1 larva/field to 1 egg/field after fruiting (Schuster et al. 1980). Pena (1983) revised the tomato pinworm threshold to 0.67

PAGE 30

22 larva/plant or 0.83 foliar injury/plant. To save time on the laborious task of counting mines, Schuster and Beck (1983) established a visual rating system for assessing total leafraines. Liriomyza trifolii (Burgess) Description Adult L. trifolii are 1.3 mm long. Their wing length is 1.5 mm (Burgess 1880). A full description of a neotypic male specimen reared from alfalfa was given by Spencer (1965). Spencer (1973) also illustrated male genitalia, a character which he uses to differentiate closely related Agromyzidae. L. sativae Blanchard is often confused with L. trifolii These species can be readily identified by differences in four features. The curvature of the aedeagus is more pronounced in L. trifolii than in L. sativae The .L. trifolii mesonotum is distinctly greyishand matte while that of L. sativae is shining black. The area of the L. trifolii upper orbits and much of the hind margin of the eye is yellow and both vertical bristles are on a yellow background. The upper orbits and much of the hind margin in L. sativae is dark and the vertical bristles are on a dark background (Spencer 1981a). Knodel-Montz and Poe (1982) used scanning electron microscopy to study female adults. They showed that L. sativae and L. trifolii females can be distinguished by the denticles and the egg guides on the genitalia. The denticles are angular and the egg guide is V-shaped in L. trifolii In L. sativae the denticles are elongate and the egg guide is acutely angled. Zehnder

PAGE 31

23 and Trumble (1983) utilized electrophoretic and scanning electron microscopy techniques to separate the two species. L. trif olii eggs are "oval, creamy or translucent and approximately 0.2 x 0.1 mm (E. A. Mortimer personal communication). The larva, which is initially colorless, darkens to yellow as it matures and the pupa is orange-yellow" (Bartlett and Powell 1981, p. 186). A distinguishing characteristic of agromyzid larvae is that the anterior pair of spiracles is located dorsally, adjacent to the dorsomeson. The anterior spiracles are laterally located in other cyclorrhaphan larvae (Peterson 1979). The number of bulbs on the posterior is sometimes useful in separating agromyzid species. There are 3 spiracular bulbs in L. sativae while 6 bulbs occur in k* huidobrensis Blanchard, another closely related species (Spencer 1981a). The larva has 12 segments (the head, 3 thoracic segments, and 8 abdominal segments). Most, and sometimes all, segments have distinctive minute tubercles in bands that are mainly confined to the segmental boundaries. Muscle scar patterns on the body are also found. The body tends to be cylindrically uniform but the anterior and posterior taper (Allen 1956). Agromyzid larvae have 2 mandibles The mandibles and the rest of the much reduced mouth parts form the cephalopharyngeal skeleton.

PAGE 32

24 Taxonomy "Article 35 of the International Commission on Zoological Nomenclature states that a name, once placed in homonymy, must be rejected. As a result, the economically important and well-known Liriomyza trif olii (Burgess) requires a new name" (Zoebisch 1984, p. 5). Zoebisch (1984) has shown from his review of literature on L_. trif olii taxonomy that the species ought to be renamed since it is a homonym of Agromyza trif olii Kaltenbach. _L. trif olii (Burgess) was first described and named by Burgess (1880) as Oscinis trifolii Coquillet (1898) placed 0. trifolii in the genus Agromyza after having synonymized the species with Agromyza diminuta 0. trifolii was placed in the genus Liriomyza by de Meijere (1925). He was also the first to note the homonymy of A. trifolii (Burgess) and A_. trifolii Kaltenbach. Spencer (1981a) synonymized L. alliovora of Frick (1955) and also L. archboldi of Frost (1962) with L. trifolii Distribution L. trifolii is of nearctic origin. Several authors put its origin in Florida (Spencer 1981b; Parrella and Keil 1984) but the first description of the species made by Burgess (1880) was of a specimen collected from white clover in the District of Columbia. This is important to note as many of the authors who place the origin of the species in Florida also claim that the worldwide spread of the species resulted from the shipping of infested plant material from Florida during the last decade. The species was known to exist in Oregon, California, Indiana as well as Florida

PAGE 33

25 and was suspected to occur throughout the U.S. as early as the 1950's (Frick 1959). This is much earlier than the reported spread into the rest of the U.S. from Florida as claimed by Parrella and Keil (1984). Irrespective of the contention that Florida is the origin of L_. trifolii the species is now distributed worldwide. It has been recorded in N. America, S. America, Africa, Europe, and Asia (Table 2-1). A distribution map of L_. trifolii is available from the Commonwealth Institute of Entomology (CIE) in London, England (CIE 1984). Host Plant Range L. trifolii is a polyphagous species (Spencer 1964). The reported host plant range of L_. trifolii is presented in Table 2-2. Only literature sources published since 1981 have been utilized in the table because there was much confusion about the leaf miner species involved prior to Spencer's (1981a) clarification on the differences between L_. trifolii and L_. sativae. Hence, the citation of a host plant does not necessarily mean a first record. It is interesting to observe that no author notes more than 100 species as hosts while the total worldwide host plant range is 148 species in 31 families. For example, Stegmaier (1981) does not record watermelon as a host of L. trifolii in Florida, while Poe and Montz (1981a, 1982) record that the insect was reared from watermelon in Texas. Fagoonee and Toory (1984) also recovered Ij. trifolii from watermelon in Mauritius.

PAGE 34

26 Table 2-1 Current Known Distribution of L_. trif olii (Burgess). Region Area Literature source N. America Califonia Florida Georgia Hawaii Indiana Iowa Massachussetts Mexico Minnesota North Carolina Nova Scotia, Can, Ohio Ontario, Can. Oregon Pennsylvania South Carolina Texas Washington Wisconsin Poe and Montz 1982 tt Frick 1959 Stegmaier 1968 Vittum 1982 Poe and Montz 1982 it it it ii Frick, 1959 Poe and Montz, 1982 Frick 1959 Price 1981 S. America Colombia Carribean Europe US Virgin Is. Bahamas Barbados Denmark Finland France Holland Italy Malta Spain Sweden U.K. W. Germany Poe and Montz 1982 Poe and Montz 1982 Spencer and Stegmaier 1973 Spencer 1963 Bartlett and Powell 1981 Tuovinen and Aapro 1981 d'Aguillar and Martinez 1979 Poe and Montz 1982 Arzone 1979 Bartlett and Powell 1981 it Nedstam 1981 Anon 1977 Bartlett and Powell 1981 continued

PAGE 35

Table 2-1 continued. Region Area Literature source Africa Canary Is, Kenya Mauritius Reunion Senegal Tanzania Poe and Montz 1982 de Lima 1979 Fagoonee and Toory 1983 Vercambre 1980 Spencer 1985 Spencer 1985 Asia Israel Japan Spencer 1985 Minkenberg and Lenteren 1986

PAGE 36

28 Table 2-2 Host plant range of L_. trif olii Botanical name Literature source Acanthaceae Thunbergia sp. Amaryllidaceae Alstromeria sp. Amaranthaceae Amaranthus retrof lexus L. A_. viridis L. Celosia sp. Apiaceae Apium graveolens var. dulce (Mill.) Pers. Coriandrum sativum L. Daucus carota L. J), carota var. sativae L. Hydrocotyle bonariensis Comm. ex Lam II. umbellata L. Pastinaca sativa L. Asclepiadaceae Asclepsias syriaca Asteraceae AReratum conyzoides L. Ambrosia artemisiif olia L. Arctium sp. Aster cordif olius L. Aster sp. Baccharis halimif olia L. Bellis perennis L. Bidens alba L. 15. pilosa L. Calendula officinalis L. Callistephus chinensis (L.) Nees Centaurea cyanus L. Chrysanthemum leucanthemum Chrysanthemum X morif olium Dahlia sp. Eclipta prostrata L. Poe and Montz 1982 Bartlett and Powell 1981 Fagoonee & Toory 1984 Zoebisch & Schuster 1984 Poe and Montz 1982 Stegmaier 1981 Stegmaier 1981 Fagoonee & Toory 1984 Broadbent 1982 Stegmaier 1981 Fagoonee & Toory 1984 Stegmaier 1981 Powell 1981 Broadbent 1982 Fagoonee & Toory 1984 Broadbent 1984 Broadbent 1982 Smith and Broadman 1986 Stegmaier 1981 Stegmaier 1981 Fagoonee & Toory 1984 Zoebisch & Schuster 1984 Stegmaier 1981 Powell 1981 Stegmaier 1981 Powell 1981 Smith and Broadman 1986 Ramat Poe and Montz 1981a Stegmaier 1981 Fagoonee & Toory 1984 continued

PAGE 37

29 Table 2-2 continued Botanical name Literature source Erechititis heiracifolia (L.) Raf. ex DC Eupatorium coelostinium L. E. serotinum Mich X. Gaillardia aristata Pursh Galinsooa ciliata (Raf.)Blake £. quadriradiata Ruiz.S Pav. Gazania sp Gerbera sp. Gerbera jamesoni Bolus ex Hook Gamochaeta pennsylvanica (Willd )Cabrera ( Gnaphalium spathalium Lam.) Stegmaier Helianthus annus L. Stegmaier Smith and Broadman Stegmaier Stegmaier Stegmaier Stegmaier Stegmaier Stegmaier Smith and Broadman Bartlett and Powell Fagoonee & Toory Stegmaier H_. bipinnatus Cav. Hymenopappus scaboseasus L'Her Lactuca canadensis L. L. sativa L. L,. serrioala L. Launaea cornuta (Oliv. & Hearn) Jeffrey Melanthera aspera Mich X Mikania scandens (L.) Willd. Senecio glabellus Poir. _S. jacobaea L. S_. vulgaris L. Solidago sp. Sonchus asper (L.) Hill S_. oleraceous L. Synedrella nodiflora (L.) Gaertn. Tagetes erecta L. T. indica L. Tagetes sp. Taraxacum officinale Weber Broadbent Tithonia diversifolia (Hemsl.) Gray Spencer Tridax procumbens L. Stegmaier Xanthium sp. Stegmaier Zinnia S P Stegmaier Stegmaier Stegmaier Broadbent Spencer Stegmaier Genung Stegmaier Powell Broadbent Broadbent Stegmaier Genung Stegmaier Stegmaier Fagoonee & Toory Fagoonee & Toory Balsaminaceae Impatiens sp. 1981 1981 1981 1981 1981 1986 1981 1984 1981 1981 1981 1986 1981 1981 1981 1984 1985 1981 1981 1981 1981 1982 1982 1981 1981 1981 1981 1984 1984 1984 1985 1981 1981 1981 Poe and Montz 1982

PAGE 38

Table 2-2 continued. 30 Botanical name Literature source Boranginaceae Cordia myxa L. Brassicaceae Brassica campestris var. rapa L. _B. chinensisL J3. juncea (L.) Czern. _B. oleraceaea var. botry tis L. B_. oleraceaea var. capitata L. B_. oleraceaea var. gemmif erae L. Capsella bursa-captoris (L, Raphanus raphanistrum L. 11 sativus L. Thlaspi arvense L. Caryophyllaceae Dianthus sp. Gypsophila sp. Stellaria media (L.) Vill Chenopodiaceae Beta vulgaris L. Chenopodium album L. Spinacea oleracea L. Convolvulaceae Ipomoea batatas (L.) Lam. Cucurbitaceae Citrullus lanatus (Thunb.) Matsumura & Nakai Cucumis melo L. sativus L. Cucurbita mixima Duchesne C. pepo L. Lagenaria siceraria (Molina) Standi. Melothria pendula L. Fagoonee & Toory 1984 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Fagoonee & Toory 1984 ) Medic. Powell 1981 Broadbent 1982 Fagoonee & Toory 1984 Broadbent 1984 Poe and Montz 1982 Poe and Montz 1981a Broadbent 1982 Stegmaier 1981 Stegmaier 1981 Stegmaier 1981 Fagoonee & Toory 1984 Poe and Montz 1981a Stegmaier 1981 Stegmaier 1981 Fagoonee & Toory 1984 Stegmaier 1981 Fagoonee & Toory 1984 Genung 1981 continued.

PAGE 39

Table 2-2 continued. 31 Botanical name Literature source Euphorbiaceae Ricinus communis L. Fabaceae Arachis hypogea L. Cajanus cajan (L.) Druce Cicer arietinum L. Crotolaria incana L. Glycine max (L.) Merr. Lathyrus odoratus L. Medicago sativa L. M. lupulina L. Phaseolus aureus Roxb. P_. coccineus L. JP. lunatus L. ( limensis McFad.?) P_. vulgaris L. Pisum sativum L. Trif olium repens L. Trigonella f oenum-graecum L. Vicia f aba L. V. angustif olia Reichard Vigna luteola (Jacq.) Benth. V. unguiculata (L.) Walp ( repens ?) Lamiaceae Ajuga remota Benth. Lamium amplexicaule L. Salvia splendens Ker-Gawl. Lillaceae Allium cepa L. onion A^. sativum L. A^. porrum L. _A. schoenoparsum L. Malvaceae Abelmoschus esculentus J. C. Malva moschata L. M. neglecta Wallr. P. Parkman, personal communication 1986 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Stegmaier 1981 Fagoonee & Toory 1984 Powell 1981 Stegmaier 1981 Smith and Broadman 1986 Stegmaier 1981 Powell 1981 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Stegmaier 1981 Stegmaier 1981 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Smith and Broadman 1986 Genung 1981 Fagoonee & Toory 1984 Spencer 1985 Broadbent 1984 Poe and Montz 1982 Stegmaier 1981 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Poe and Montz 1982 Wendl. Stegmaier 1981 Smith and Broadman 1986 Broadbent 1982

PAGE 40

32 Table 2-2 continued. Botanical Literature name source Oxalidaceae Oxalis corniculata L. 0. corymbosa DC. 0. latifolia HBK. Plantaginaceae Plantago lanceolota L. P_. ma jor L. Polemoniaceae Phloxdrum mondii L. Polygonaceae Polygonum convolvulus P_. persicaria L. Primulaceae Primula sp. Ranunculaceae Ranunculus repens L. R. sp Rosaceae Crataegus monogyna Jacq. Scrophulariaceae Antirrhinum ma jus L. Linaria canadensis (L.) DC. Solanaceae Capsicum sp. Lycopersicon esculentum Mill. Petunia sp. Physalis sp. P_. pubescens L. Solanum mauritianum Scop. S_. dulcamara L. S_. indicum L. S.. melongena L. S. americanum L. S_. nigrum L. S_. tuberosum L. Tropeolaceae Tropaeolum ma jus L. T. peregrinum L. Fagoonee & Toory 1984 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Broadbent 1984 Broadbent 1984 Fagoonee & Toory 1984 Smith and Broadman 1986 Broadbent 1984 Bartlett and Powell 1981 Powell 1981 Powell 1981 Powell 1981 Fagoonee & Toory 1984 Fagoonee & Toory 1984 Stegmaier 1981 Stegmaier 1981 Stegmaier 1981 Stegmaier 1981 Zoebisch & Schuster 1984 Fagoonee & Toory 1984 Powell 1981 Fagoonee & Toory 1984 Stegmaier 1981 Zoebisch & Schuster 1984 Fagoonee & Toory 1984 Spencer 1985 Powell 1981 Powell 1981 continued

PAGE 41

33 Table 2-2 continued. Botanical Literature name source Turneraceae Piriqueta caroliniana (Walt.) Urban Stegraaier 1981 Verbenaceae Lantana camara L. Fagoonee & Toory 1984 Verbena officinalis L. Fagoonee & Toory 1984 Zygophyllaceae Kallstroemia mixima (L.) Hook.S Arn. Stegraaier 1981 Tribulus terrestris L. Poe and Montz 1982

PAGE 42

34 Life History Eggs are laid singly in the leaf meosophyll by a female inserting her ovipositor through the upper epidermis and creating a depression or N stipple' which varies in size and shape according to the age of the female (Personal communication Dong 1980). The larva feeds on the mesophyll, creating a serpentine mine. A trail of fecal material, alternating from one side to another, is deposited in the mine. The mature third-instar larva makes a semicircular hole in the epidermis and drops to the soil. As with other Agromyzidae, L. trif olii pupates within the cuticle of the last larval instar. Adults emerge from the puparia and can mate on the first day (personal observation). After mating, female begin ovipositing in leaves. Females also feed from puctures made in the leaves. These punctures are larger than oviposition punctures (Zoebisch 1984). Longevity Adult lifespan varies with environmental conditions. Temperature and food sources are especially important regulators of longevity. Charlton and Allen (1981) found that at 23.8 C females lived 22.7 days and males lived 13.9 days when the the flies were given honey in addition to the host plant, blackeyed pea. Interestingly, when only honey was provided, males lived an additional 1.9 days; the female lifespan, on the other hand, was shortened by 6.4 days. Supposedly, males also feed from the punctures made by females (Musgrave et al. 1975); however, if this were the case, males would live longer when the host plant was made available in addition to honey.

PAGE 43

35 Leibee (1981a) found that temperature influenced leafrainer longevity on celery. The females lived 27.7, 28.3, 16.8, and 14.6 days at 15, 20, 25, and 30 C respectively when a 10% honey solution was provided with celery. Parrella et al. (1983b) showed that female lifespans were 14, 12, and 10 days on chrysanthemum, celery, and tomato, respectively, when honey was also provided (the authors do not report at what temperature the flies were maintained). Fecundity Like longevity, the egg laying capacity of leafminers also varies. Once again, temperature, host plant and food supplements influence the fecundity of flies. L_. trif olii laid 239.67, 405.67, 288.25, 182.33, and 24.33 eggs at 35, 30, 25, 20, and 15 C, respectively, on celery when the flies were provided with a 10% honey solution (Leibee 1981a). Flies maintained at 23.8 C on black-eyed pea laid 177 eggs (Charlton and Allen 1981). An average 439 eggs was deposited when the flies were given honey also. The mean numbers of viable eggs were 298, 212, and 39 on chrysanthemum, celery and tomato when honey was also given to the flies (Parrella et al. 1983b). The daily oviposition rates peaked on day 1, 2, and 4 at 35, 30, and 25 C, respectively, on celery. Also, oviposition was much reduced at lower temperatures and almost stopped at 15 C (Leibee 1981a). Even though the flies lived much longer at the lower temperatures, they cannot be very damaging to the host plants since fewer eggs are laid. Provision of additional carbohydrate sources results in increased fecundity. Naturally occurring carbohydrate sources

PAGE 44

include floral and extra-floral nectars, and aphid and other homopteran honeydew secretions. If such sources were exploited by the flies, their fecundity would be increased. Zoebisch and Schuster (in press) have demonstrated that potato aphid, Macrosiphum euphorbiae (Thomas), honeydew on tomato leaflets increased the fecundity and longevity of L_. trif olii under laboratory conditions. Developmental times for L. trifolii on some crop plants Temperature regulates the rate at which L_. trifolii develop. Host plants also affect developmental times (Table 2-3). It is interesting to note that Leibee (1981a) reports much longer developmental times for the species on celery than other authors report on chrysanthemum, pink bean and tomato. For example, at 15.0 C, larvae on celery mature after 25.8 days while larvae pupate in 12.6 days at 14.8 C on pink bean and in 10.1 days at 15.6 C on tomato. Why development takes much longer on celery is not clear. Celery is not a "bad" host plant for L. trifolii since the fly lays comparable numbers of eggs on celery and favored host plants, such as the chrysanthemum (Parrella et al. 1983b). Mortality Temperature also influences leafminer mortality. Only 9.4% of the pupae survived at 35 C while almost 10 times as many pupae survived to the adult stage at temperatures ranging from 15 to 30 C (Leibee 1981a). Parrella et al. (1983b) found that at 37.8 C 100% pupae died on White Hurricane cv. chrysanthemum. The host plant also influences mortality of the immature stages. Charlton and Allen (1981) reported that total larval and pupal

PAGE 45

37 Table 2-3 Developmental times for L_. trif olii immature stages on some crop plants. Temp #days as Host Literature C egg larva pupa plant source 14.8 10.7 12.6 28.0 pink bean Charlton and Allen 1981 15.0 10.0 25.8 — celery Leibee 1981a 15.6 — 10.1 — tomato Schuster and Patel 1985 20.0 3.0 8.0 10.6 mum Charlton and Allen 1981 4.2 o. 7 9.4 pink bean Charlton and Allen 1981 4.4 12.0 celery Leibee 1981a 21.1 7.1 tomato Schuster and Patel 1985 25.0 2.2 4.7 8.2 mum Charlton and Allen 1981 2.9 4.5 8.5 pink bean Charlton and Allen 1981 2.3 8.0 celery Leibee 1981a 26.7 4.4 tomato Schuster and Patel 1985 30.0 3.8 5.1 6.7 mum Charlton and Allen 1981 2.2 3.7 6.8 pink bean Charlton and Allen 1981 6.8 celery Leibee 1981a 32.2 3.5 tomato Schuster and Patel 1985 32.5 5.2 4.5 7.6 mum Charlton and Allen 1981 2.0 3.4 6.9 pink bean Charlton and Allen 1981 2.0 5.4 celery Leibee 1981a

PAGE 46

38 mortality was 26.9, 26.8, 52.6, and 99.0% on pink bean, black-eyed pea, *show-off mum and ''yellow knight' mum, respectively. As humidity is increased, the percentage of pupae becoming adults also increases. More than 60% of the pupae held at any humidity higher than 50% RH survived to become adults. Circadian behavior L. trif olii exhibits diurnal patterns in many of its primary behavioral phenomena including feeding, oviposition, larval emergence, and adult emergence (Charlton and Allen 1981). Most of these activities occurred primarily in the late morning hours. Most L. trif olii larvae emerged from black-eyed pea leaves between 0830 and 1330 h (Charlton and Allen 1981). No larvae exited mines between 1830 and 0630 h. Adult emergence was concentrated between 0930 and 1230 h and no adults emerged from 1530 to 0730 h. Oviposition and feeding continued throughout the daylight hours, but the period of greatest oviposition activity was from 1130 to 1430 h. In this same period, the ratio of punctures: eggs was also at its lowest. Feeding occurred throughout the day, although the number of feeding punctures was the lowest during 0630 to 0830 h and after 2030 h. Zehnder and Trumble (1984b) used yellow sticky trap cards to monitor adult leafminer activity in tomato. They found that, in 1981, more L. trifolii were trapped between 1500 and 2000 h. But in 1982, equal numbers were trapped between 0700-1100 h and 1100-1500 h with fewer adults trapped during 1500-2000 h. It is interesting to note that the period of greatest oviposition

PAGE 47

39 activity recorded by Charlton and Allen (1981) coincides with the lower daytime sticky trap counts monitored by Zehnder and Trumble (1982) in 1982. Zehnder and Trumble (1982) contend that the high trap count periods are associated with greater female oviposition activity, arguing that the more active females find the best oviposition sites and increase the time available for oviposition. Diapause Attempts by Charlton and Allen (1981) to induce diapause in Californian L. trif olii failed. Larew et al. (1986) have shown that JL. trif olii can overwinter outdoors in Maryland. They suspect that the insect probably diapauses in the pupal stage. Chemical Control of Leafminers On Tomato Leibee (1981b) discussed past and contemporary methods of control, the importance of leaf miner parasites, and the development of insecticide resistance in Liriomyza spp. leafminers on vegetables in Florida. He discussed the effectiveness or noneffectiveness of insecticides used in Florida until 1981 but did not specifically name the leafminer species involved because of the difficulty in verification since voucher specimens were not retained. Several authors have reported on the efficacy of various chemicals against L. trif olii on tomato since 1981. In a detailed study, Schuster and Everett (1983) examined the effects of avermectin, cyromazine, permethrin, fenvalerate (Pydrin R ), and methamidophos on oviposition and survival of L. trif olii Avermectin and cyromazine were effective in controlling the insect on tomato in the field. In the laboratory, avermectin killed L. trifolii larvae and pupae and also inhibited oviposition.

PAGE 48

40 Cyromazine had no effect on oviposition. Very few adults emerged from foliage treated with averraectin and none emerged from cyromazine treated foliage because no larvae from this treatment pupated successfully. Methamidophos resulted in 80% of the adults dying within 24 hours after treatment in the laboratory. Permethrin, cyromazine, or avermectin did not induce as high a mortality as did methamidophos. Avermectin is a macrocyclic lactone natural product isolated from the soil microorganism Streptomyces avermitilis n. sp. (Brown and Dybas 1982). It is effective against insects from several orders and against mites and nematodes. Its mode of action involves inhibition of gamma amino-butyric acid (GABA) mediated neuromuscular junction impulse transmission. Although the chemical has no apparent direct effect on adult mortality, it apparently inhibits the use of the ovipositor. Females are, thus, unable to make oviposition or feeding punctures (Schuster and Everett 1983). Cyromazine is an insect growth regulator (IGR) with a mode of action that is suspected to be hormonal (Parrella et al. 1983a). Leibee (1985) examined the dosage-response relationship of cyromazine on the development of Ij. trif olii Increasing dosage led to higher larval mortality as well as increased numbers of larviform and other abnormally developed pupae. Such pupae failed to become adults. Lindquist and Casey (1985) collected more pupae from methomyl treated greenhouse tomato plants than from cyromazine treated plants. Their study showed that methomyl had an adverse effect on the parasite population density while neither

PAGE 49

41 cyromazine nor parasitoid introduction altered the parasitoid population density. Neem seed extract has been tested for activity against L. trif olii The leaves and seeds of the neem tree, Azadirachta indica Juss. have been used in India for hundreds of years as sources of insect repellents and insecticides (Jacobson 1981). It is believed that the principle insecticidal compound, azadirachtin, has IGR properties (Rembold et al. 1982). The chemical is absorbed by roots and translocated to the leaves (Larew et al. 1984, Webb et al. 1984). Those authors also report the efficacy of neem seed extract against the puparia when used as a soil drench, suggesting that the active ingredient can be absorbed by the pupae through the puparial cuticle. Webb et al. (1984) have also shown that neem seed extract and the purified azadirachtin are highly active against L^. trif olii Intensive control methods, including the use of several insecticides, are employed whenever outbreaks of L. trif olii occur in the greenhouses in England (Powell 1981). The methods include fumigation of empty greenhouses with methyl bromide, twice weekly applications of heptenophos (applied as a high volume, HV, foliar spray), oxamyl applied in granules, permethrin (HV foliar spray), and resmethrin (ultra low volume spray) application. Biological Control of Leafminers on Tomato Poe and Montz (1981b) summarized the parasitoids recovered from L. trifolii L. sativae, L. huidobrensis and L brassicae All parasitoids recovered so far were Hymenoptera and included

PAGE 50

42 species in the families Braconidae, Cynipidae, Eucoilidae, Eulophidae, and Pteromalidae. Some species, including Opius dimidiatus Ashmead (Braconidae), Ganaspidium sp. (Eucoilidae), Halticoptera aenea (Walker) (Pteromalidae), and the eulophids, Chrysocharis ainsliei Crawford, Closterocerus cinctipennis (Ashmead) Diglyphus intermedius and _D. begini (Ashmead) have been recovered from more than one species of leafminers. Schuster (1985) reared the following parasitoids from leafminers infesting tomato in the Bradenton area during 1980 to 1984: Diglyphus intermedius Diglyphus sp., Chrysonotomyia punctiventris (Crawford), Diaulinopsis callichroma Crawford, Chrysocharis parksi Crawford, Opius sp., Halticoptera circulus (Walker), and Ganaspidium sp. Zehnder and Trumble (1984a) indicated that L. sativae JL. trif olii and some of their natural enemies could discriminate between host plant species of adjacently grown celery and tomato. Diglyphus begini D. intermedius Chrysocharis parksi C. ainsliei Halticoptera circulus and Chrysyonotomyia punctiventris were the most abundant parasitoids. C. parksi was more numerous on tomato, which was a favored host of L. sativae I), intermedius was more active in celery where L. trif olii was the predominant leaf miner species. The host plant, or the host leafminer species, or both may be involved in observed differential densities of the parasitoid species. The impact of insecticides on parasitoids has been studied by several workers. Hills and Taylor (1951) and Shorey and Hall (1963)

PAGE 51

43 demonstrated that DDT led to an increase in leafrainer populations due to a reduction in parasitoid density. The subsequent switch to methoxychlor, dieldrin, endrin, and lindane (Wene 1955) and also to parathion, ethion, and diazinon (Getzin 1960) showed the same results. Synthetic pyrethroid residues were found to be less toxic to I), intermedius than was methomyl when pyrethroids were first introduced to control leafminers (Waddill 1978). Fenvalerate was the least toxic pyrethroid evaluated. A combination of leptophos R T? (Phosvel ) and endosulfan (Thiodan ) sprayed on tomato foliage killed fewer parasitoids than the control (water spray) or any other insecticide tested (Poe et al. 1978). Weekly applications of oxamyl (Vydate ) on tomato foliage reduced the number of parasitoids reared from excised tomato foliage compared to the water control (Schuster et al. 1979). A single application of oxamyl at rates ranging from 30g Al (active ingredient)/100 1 to as high as 119. 8g AI/100 1 or methamidophos at 89. 9g AI/100 1 did not reduce parasitoid emergence. Getzin (1960) recommended integration of chemical and biological control methods by selecting chemicals that would kill leafminers but not the parasitoids. Such chemicals have not been available until recently. Most insecticides used for leafminer control have been broad-spectrum and have induced mortality in parasitoid and leafminer populations. Schuster and Price (1985) noted that Opius sp. and D. intermedius were proportionately greater in nonsprayed tomatoes

PAGE 52

44 than in sprayed tomatoes where Chrysonotomyia punctiventris was more abundant. Organophosphates depressed Diglyphus spp. while favoring Chrysonotomyia spp. in celery (Zehnder and Trumble 1985). The pyrethroid permethrin, on the other hand, seemed to favor Diglyphus species. Schuster (1985) noted that the proportions of different parasitoid species varied from season to season. Methomyl and permethrin application reduced population densities of all parasitoids including I), intermedins when compared with a water check (Schuster and Price 1985). Endosulfan in combination with Bacillus thuringiensis var kurstaki Berliner did not significantly reduce parasitoid numbers after 2 applications while endosulfan alone did reduce parasitoid numbers, d. intermedins did not decrease significantly even when endosulfan was applied alone. This suggests a differential activity of endosulfan against the various parasitoids of the leafminers. Such differential activity was also noted with methomyl which was more injurious to D. interemdius and Qpius spp. compared to Chrysonotomyia punctiventris Methomyl also reduced the percentage of parasitism of leafminers on celery in California as a result of a reduced rate of adult parasitoid survival indicated by the increased number of dead adult parasitoids collected from trays placed under the plants (Trumble 1985). The ratio of leafminers to parasitoids reared was lowest for cyromazine and avermectin treated plants (1.18:1 and 1.41:1 respectively) in 1982.

PAGE 53

45 In 1983, avermectin and a water check had the lowest leafrainer to parasitoids ratios (3.80:1 and 4.16:1 respectively). These results suggest that the effects of insecticides could not be readily predicted from one year to the next. Also they do not prove conclusively that there is differential susceptibility of leafminers and parasitoids to cyromazine. Application of avermectin as well as other chemicals altered the composition of the parasitoid complex. Avermectin reduced the numbers of J). intermedius reared, but at the same time increased the proportion of I), begini in comparison to other treatments. In 1983, no reduction in JJ. intermedius populations was observed in the avermectin treatment. Although Trumble (1985) stated that changing the pesticide shifted the composition of the parasitoid complex, he failed to qualify his conclusions by not emphasizing the differences observed in the composition of the parasitoid complex during the two years in which in the experiments were done. From a single season of experiments done in 1984, Zehnder and Trumble (1985) reported that out of six insecticides (permethrin, diazinon, naled (Dibrom R ), methamidophos mevinphos (Phosdrin R ) and endosulfan) applied to celery, permethrin produced greater parasitoid mortality than any other treatment. The authors also reported that the composition of the parasitoid complex varied with the treatments. Organophosphates favored C. punctiventris emergence while permethrin treatment led to recovery of more Diglyphus species.

PAGE 54

46 Parasitoid population levels are monitored in a field by recording numbers of live leafminer larvae. Spraying decisions are based on live larvae counts rather than the earlier practice of using total mine counts, which ignored the contribution of parasitoids to regulating leafminer populations (Pohronezny et al. 1984). Most workers (e.g. Trumble and Nakakihara 1983) report parasitoid abundance in fields in terras of % parasitism. This method is faulty because no correlation between % parasitism and actual parasitoid densities in the field is established. Schuster and Beck (unpublished) have formulated a sampling plan for estimating leafminer densities. This plan could be incorporated into a program for correlating % parasitism and parasitoid— host densities in conjunction with measurements made by Marlow et al. (1983) of plant area at different growth stages of the plant. Utilization of parasitoids is more intense in greenhouse tomato cultivation. Success in regulating leafminer populations with releases of parasitoids into greenhouses has been documented in several parts of the world. Lindquist and Casey (1985) released Diglyphus into a greenhouse after an initial population of L. trifolii had been established. This treatment gave the same tomato yields as alternative treatments where either cyromazine or methomyl was used as the control agent. Petitt (1984) managed L. sativae infestations on greenhouse tomato by vacuuming the plastic mulch, increasing parasitoid populations, particularly of Opius S P> a d by not using any chemical insecticides in the greenhouse. Lyon (1984) obtained greater success in regulating

PAGE 55

47 JL. trif olii populations on greenhouse tomato in France by initially establishing populations of Diglyphus isaea Walker and Dacnusa sibirica Telenga in the greenhouse and the surrounding vegetation rather than waiting to release the parasitoids into infested greenhouses. Diglyphus intermedius (Girault) Taxonomic Status, Hosts and Distribution I), intermedius is one of eighteen species belonging to this genus in the family Eulophidae. The species was described by Girault from one female reared in 1916 from Phytomyza chrysanthemi Kowarz (Gordh and Hendrickson 1979). Hosts of Jj. intermedius include Agromyza frontella (Rondani), Liriomyza brassicae (Riley) L^. sativae L. trif oliearum Spencer, L. trif olii Phytomyza atricornis Meigen, P. chrysanthemi and P. nigra Meigen (Kamm 1977; and Gordh and Hendrickson 1979). Gordh and Hendrickson (1979) consider the species to be neartic and neotropical in distribution. Adult External Morphology The general trend of sexual dimorphism in the subfamily Eulophinae is expressed by the presence of dorsal rami on the funicular segments of the male antennae. This is not evident in Diglyphus the male of which is distinguished from the female by smaller size and longer marginal fringe on the forewings (Gordh and Hendrickson 1979). The genus Diglyphus has two-segmented funicles and parallel longitudinal grooves on the scutellura. D. intermedius and D. isaea are the only two species in this genus that closely

PAGE 56

48 resemble each other. In the female J), intermedius only the basal 0.25-0.35 of the hind tibia is metalic colored with coloration fading along the middle 0.30. In J), isaea the hind tibia is predominantly metallic colored and the metallic color does not fade distally. I), begini has pale markings on its antennal scapes the middle and hind tibiae are without apical duskiness and the basal cell of the forewing is sparsely or moderately setose (Gordh and Hendrickson 1979). V. isaea an introduced palearctic species, has not been recorded from Florida. Parasitoid Abundance in Florida D. intermedius was at least as numerous as any of the 5 other parasitoids reared during the spring of 1975 from excised tomato foliage from Immokalee (Poe et al. 1978). D. intermedius accounted for 46.6% and Chrysonotomyia formosa Westwood accounted for 49.7% of all parasitoids reared from excised tomato foliage during the winter and spring of 1975 (Schuster et al. 1979). Schuster and Price (In Press) observed that D. intermedius continues to be one of the most common parasitoids found on tomato in Florida. Biology and Life History Biological notes on a related species, D. begini were first made by Webster and Parks (1913). Hendrickson and Barth (1978) studied the life history and biology of D. intermedius in the laboratory. They reported that D. intermedius was a solitary ectoparasitoid of leafminers, although occasionally 2 to 5 parasitoids developed on a single Liriomyza trifoliearum larva.

PAGE 57

49 It paralyzed and killed all 3 larval instars, but it usually deposited eggs on 3rd-instars only. Incomplete larval development ocurred when eggs were deposited on 2nd-instar hosts, possibly because of an inadequate food supply. Developmental times were recorded for parasitoids reared at a constant temperature of 25.5 C and 60% RH and a photoperiod of 16L:8D. Hosts were reared on bush snap bean, Phaseolus vulgaris cv Bountiful and lima bean _P. limensis cv Thaxter. As it was difficult to observe parasitoids inside leaf tissue, artificial mines were constructed with small cardboard rings sandwiched between a glass microscope slide and coverslip. The egg, larval, and pupal stages lasted 1, 4, and 6 days, respectively, under the conditions provided. Patel and Schuster (1983), using a method similar to that of Hendrickson and Barth (1978), measured developmental rates of D. intermedins at 15.6, 21.1, 26.7, and 32.3 C constant temperatures. The quadratic regression equation Y = -0.2028 + 0.0214T 0.0004T 2 described the relation found between temperature (T) and the developmental rate/day (Y). At 32.2 C, parasitoid larval mortality averaged 88.3% which was twice the mortality at other temperatures. The manner of pupation in I), intermedins is unusual. The larva changes from light yelow to lime bluish green and constructs 6-8 pillars made from meconium voided at the time of pupation. The pillars extend vertically between the two epidermal layers and are usually laid in pairs along the length of the pupa. It has been suggested that the pillars protect the pupa from being crushed if

PAGE 58

50 the leaf dries out and also prevents the pupa from rolling about in its mine. Hendrickson and Barth (1978) noted that a female D. intermedius (N = 6) produced an average 40.2 progeny and lived from 3 to 4 weeks. Modeling the Leaf miner-Parasitoids System Introduction Modeling is fast gaining acceptance as a major tool in life system studies, including insect population dynamics research (Ruesink 1975, Huffaker et al. 1976, Barfield and Jones 1979, Shoemaker 1980, Getz and Gutierrez 1982, Stimac 1982). A system can be defined as a composition of interacting objects (Smerage 1980). A model is simply a description of a system and may take many forms. Two factors have contributed to the popularity of the modeling approach: 1) Computers have made it possible, at a rapid pace, to realistically simulate and predict the behavior of real life systems using models. 2) It is becoming increasingly cheaper to simulate rather than establish experiments to develop and test system behavior hypotheses. The leafminer model of Smerage et al. (1980), for example, can be utilized in predicting leafminer and parasitoid population dynamics much more cheaply and quickly than a field experiment ever could. Models such as that by Smerage et al. also provide a framework in which research towards better understanding of a system can be rationally organized, expressed and analyzed. The model of Smerage et al. has been a motivating factor for much of the research

PAGE 59

51 done on leafminers in Florida. It is therefore appropriate to review briefly here this model. The Smerage et al. Leaf miner-Parasitoid Model This model represents insect population systems in general, although it specifically described the leaf miner-parasitoid population system. Figure 2-1 illustrates a. conceptual model of the leaf miner-parasitoid system. The model pertained to the coupling of within-field egg, larval, pupal and adult populations of leafminers, labeled Cj, C 2 C 3 and C^, respectively, with the egg, larval, pupal, and adult populations of parasitoids, C 5 C^, Cy, and Cg, respectively, by the parasitism process Hp. The densities of each class of populations is labeled n^ where the subscript i, is denoted by the same number as the population. For example leafminer egg population, C^, has density n,. Transformations Hg represent development and physiological mortality. The flow of an individual, from one store to the next, resulted from the individual developing into the next stage. For example, upon completion of the egg stage, a leafminer individual flowed from the leafminer egg store to the leafminer larval store through Hj. Flow into the egg store resulted from oviposition, Hq. Mortality of an individual in a particular store, for example in the leafminer egg store, resulted in the flow of that individual into a 'sink' for dead individuals. Sinks are denoted by triangular

PAGE 60

52 I) CO M u 0) E W s 0 s 0) -J X w o iJ c •H O CO -H CO J-> Ui 10 co c a. co I rU Q. 0) X c
PAGE 61

53 symbols in Figure 2-1. Temperature dependence of a process is indicated by the terminal labeled T. Insecticide mortalities affecting larvae and adults were incorporated as extrinsic factors. In Figure 2-1, flow sources, ^12* ^14' ^16' anc ^18 remove individuals from their respective age classes at prescribed rates that reflect insecticide induced mortality. Population density sources n g 2> n s 4 an( n 8 re P r esent external sources of leafminer larvae, leafrainer adults, and parasitoid adults, respectively. Movement of adults between the field is in response to a gradient in an environmental variable. and Gg denote the processes that regulate movement of leafminer and parasitoid adults, respectively. Process G^ may be a function of the quality of vegetation V" F in the external reservoir, and in the field to which leaf miners respond. Parasitism, Hp, was incorporated as the link between leafrainer and parasitoid populations, a process by which attacked hosts were killed and a parasitoid egg was simultaneously oviposited on each host larva. In this model, the rate of parasitism was a function of host and parasitoid densities and environmental variables. The exponential and the rectangular hyperbolic functions represented by equations (7) and (8) are alternative closed-form descriptions of the family of curves for the rate of parasitism when plotted as in Figure 1-1. Y = C p .Kp.n 2 (1 expCnj/Kh)) (7)

PAGE 62

54 Y = C p .Kp.n 2 (n l /(n 1 + Kl)) (8) In (7) and (8), Y is the number of hosts parasitized/day and Kp is the specific parasitism rate (hosts/adult-day). In (7) and (8) Kh and Kl represent the critical host density above which Y is nearly independent of host density, n^. Parasitoid density is denoted by n^. The capacitance, or the ground area or volume occupied by the parasitoids is represented by Cp. The mathematical model of the system is not reviewed here. It is sufficient to report here that a mathematical model results in the system being described as a single entity by a set of complex equations systematically obtained from the descriptions of the processes and structure (Smerage 1980). The reader is referred to the chapter on mathematical model in Smerage et al. (1980) if interested in more details or mathematical description of the model. Several simplifying assumptions were made in formulating the model: 1) the celery crop was a non-limiting host resource for leafminer reproduction and development, 2) all parasitoid species were aggregated into an equivalent single species population, 3) development rates of each species were linear functions of temperature, 4) oviposition rates were constant and independent of temperature, 5) both populations were uniformly distributed over the field, 6) parasitism was an exponential function of host and parasitoid densities, and 7) movement of both populations was by diffusion or drift.

PAGE 63

55 A combination of hypothetical and preliminary estimates of parameter values obtained from data collected from greenhouse, laboratory, and field studies were utilized to run computer simulations of the Smerage et al. model (Musgrave et al. 1980). These parameters included durations of 40, AO, 90, and 35 degreedays for the leafminer egg, larval, pupal, and adult stages. Corresponding parasitoid durations were 25, AO, 1A0, and 70 degreedays. Leafminer sex-ratio was observed to be 1:1 with each female laying A eggs/day. The parasitoid sex ratio was not measured but a parasitoid was assumed to lay 1 egg per host, killing 2 hosts/day. Natural mortalities, other than mortality due to parasitization for the egg, larval, pupal, and adult stages for each egg laid, respectively, were assumed to be 0.A, 0.2, 0.2, and 1.0 for the leafminer, and 0.2, 0.2, 0.1, and 1.0, respectively, for the parasitoid. The critical host density (number of hosts/area), Kh, and the specific parasitism rate (number of hosts parasitized/ parasitoid adult-day), Kp, as defined in equation (A) in Chapter I, were assumed to be 5 larvae/sq m and 2 larvae/adult-day, respectively. Simulations using several initial conditions provided considerable insight into the influence of various factors on the leafminer-parasitoid population dynamics. Important findings included the observation that sudden outbreaks of leafminers could only be suppressed by the parasitoids 20 to 30 days after the outbreak started. If external reservoirs of leafminers were large, chemical control of leafminers inside the field was ineffective and

PAGE 64

56 required frequent and routine applications, and that under these conditions could even be considered wasteful (Musgrave et al. 1980). More importantly, the work of Smerage et al. (1980) in formulating the leafminer model, and that of Musgrave et al. (1980) in simulating the model has been a significant factor in determining the direction of research on leafminers in Florida. This is especially the case with the work presented in Chapters IV through IX of this dissertation. As with other research currently ongoing in Florida, these chapters describe the work done to obtain better estimates of parameters and descriptions of the processes involved in the leaf miner-parasitoid system.

PAGE 65

CHAPTER III PLANT PRODUCTION AND REARING TECHNIQUES FOR Liriomyza trif olii (Burgess) AND Diglyphus intermedius (Girault) Introduction Colonies of L_. trif olii and J), intermedius were maintained to insure that healthy, homogeneous populations of insects were always available. The techniques used for rearing the insects as well as the host plants are described in this chapter. As many as 1000-2000 adults per day of each insect species were produced by using this method. Several plant species can be used for rearing _L. trif olii (see section on host plant range in chapter II). Perhaps the fastest and easiest plants to grow are the bean plants, which have been used by several researchers working on L. trif olii (e.g. Charlton and Allen 1981). Tomato cultivars "Walter' and x Hayslip' were used because the results of the experiments were to be used to understand the leaf miner-parasitoid relationship on the tomato. Host Plant Production p Planting Tomato seeds were planted in Speedling cellular styrofoam trays. In each cell, 4 to 8 seeds were placed 1/2 cm deep in Speedling soil mix. The soil pH was maintained between 5.5 and 6.5. Acidic soil mixes were corrected to ideal pH by adding lime or sodium hydroxide. Two trays were planted every Monday, Wednesday, and Friday. As many as 12 trays were planted 57

PAGE 66

58 per week when more plants were needed for the experiments. Seed germination took 3 to 10 days, depending on the temperature. Trays were watered gently every day, although twice daily watering was required on very hot days. Plants were fertilized twice weekly, 379 g of 20-20-20 (N, P, K) Miller's brand nutrileaf solution was alternated with a solution of 80 g MgCNO^^^^O, 196 g KN0 3 and 336 g of Ca(N0 3 ) 2 .4H 2 0 in 10 liters of water to give a 500 ppm N, 500 ppm K and 50 ppm Mg solution. The solutions p were applied with a Hyponex Repotting When the seedlings had reached the first 2-true leaf stage, they were removed and repotted in 6-inch pots. One seedling was planted in each pot if the plant was to be utilized in leaf mier-parasitoid density experiments. Four seedlings were planted in each pot for maintaining the insect colonies as well as for rearing insects for other experiments. Potted plants were watered and fertilized the same as the seedlings. Precautions and care Chemical pesticides were not applied as long as their use was avoidable. Plants were kept free of insects by good sanitation practices, culling, mechanical control, exploiting natural enemies and the appropiate placing of baits. Most pest problems were of short duration; they included cockroaches, aphids, Myzus persici (Sulzer), russett mites, Aculops lycopersici (Massee), and on rare occasions, armyworms. Dursban baits were placed next to trays of younger seedling plants that were preferred by cockroaches. Russett mite infested

PAGE 67

59 plants were discarded and the bench holding infested plants was drenched with Kelthane (dicofol) and not used for 1 month. Green peach aphids were killed by hand. Heavily infested plants were discarded. On a few occasions, armyworm egg masses were detected on plants inside the screenhouse. Leaves with these egg masses were removed. Trays and transplants were drenched with a solution of 6.75 g. of Truban in 6 liters of water. Leaf mold became a problem during warm humid conditions. If leafmold was present, Dithane M-45 was sprayed at 16.8 g/1 on the plants weekly until the leafmold disappeared. Leaves infested with leafrainer larvae were removed Rearing L. trifolii Equipment and procedure A converted refrigerated truck body housed the leafminer colony. A 121:12D photoperiod was provided by fluorescent lighting. An air-conditioner and a space heater were utilized to maintain the temperature at 22-32 C. The humidity inside the truck body was uncontrolled and ranged from 40 to 80% RH. Fifty-200 adult flies were held in a 61 x 61 x 61 cm. organdy cloth-covered, screened, oviposition cage. Four plants were placed in the cage and replaced daily between 0800 and 0900 h with fresh plants. A filter paper smeared with honey was hung from the top of the cage. The honey ensured a large numbers of eggs were laid daily (Charlton and Allen 1981). New flies were added periodically to maintain 50 to 200 flies in the cage.

PAGE 68

60 On removal from the ovipositon cage, the plants were shaken to dislodge any flies still on the plants and placed in holding cages. Holding cages were similar to the oviposition cage. Plants were watered daily. Larvae took 5 days to reach the third-instar On the sixth day, the plants were removed from the holding cages. Leaves were p cut from the plants and placed in Tupperware (Superseal, 30 x 30 x 12 cm) plastic containers. The containers had screen-covered holes for ventilation. A hardware cloth divider (11 x 11 mm mesh) in the box kept the leaves elevated from the container bottoms. Larvae exiting the leaves dropped to the container bottoms. The p bottoms were coated with a layer of Teflon material to prevent the puparia from sticking to the containers. Leaves were discarded, the hardware screen was removed, and the puparia were gently brushed off the bottoms with a fine camel-hair brush 3 days after the leaves were placed in the containers. About 50-200 puparia were collected and placed in a 30-g clear plastic creamer cup. A filter paper strip smeared with honey was placed inside the cup. Adults emerged in 4 to 5 days and fed on the honey. A cup containing adult flies was placed inside the oviposition cage and the lid removed so that the adults escaped into the cage. Leafminers in all stages of development were available daily. It took 14 to 16 days for development of an egg to the adult stage under the conditions prevalent in the rearing facilities

PAGE 69

61 Precautions and care All the cages and containers were kept free of standing water to prevent leafminer larvae and puparia from drowning. Plants were watered prior to being placed in the oviposition cage. Since adult longevity is considerably shortened by very high temperatures and the egg laying activity is reduced by temperatures below 17 C, the temperature inside the truck body was monitored with a 7-day hygrothermograph recorder and failures in the air-conditioner and heating unit were corrected as soon as possible after they were noticed. Transfers of exposed plants into holding cages were recorded daily to monitor the age of larvae. Puparia transferred were similarly recorded. Extra leafminer puparia were refrigerated at 5 C and later removed when needed for the leafminer colony on the few occasions that there were not enough adults from the regular colony. Puparia were discarded 1 week after being placed in the refrigerator if they were not needed in the colony. To maintain vigor of the colony, tomato foliage infested with leafminers was collected from the field approximately every 6 months, and the insects were reared to the adult stage. Fifty to 200 adults were transferred to a new oviposition cage and assimilated into the colony. Too many ovipositing flies (>200) resulted in too many eggs. Under such overcowded conditions, plants could be totally defoliated by the developing larvae. The larvae that were able to pupate were smaller. It was, therefore, necessary to ensure that no more than 200 adults were present in the oviposition cage.

PAGE 70

62 Rearing D. intermedius Equipment and procedure The parasitoid colony was kept in a separate room with temperature, lighting, and humidity conditions similar to those in the converted refrigerated truck body that housed the leafminer colony. One or 2 tomato plants containing 5-day-old leafminer larvae were removed daily from the truck body. These plants were placed inside a parasitoid oviposition cage similar to the other cages. Approximately 20 to 100 adults were introduced into the oviposition cage every week. Plants were removed from the cage after 1 day's exposure to the parasitoid adults and replaced with other plants containing 5-day-old leafminer larvae. Plants with parasitized host larvae were not caged after removal from the oviposition cage to make watering easier. Parasitoid larvae turned bluish-green after 5 to 6 days. At this point, leaflets were stripped from the plants and put in 1-quart ice-cream cartons which were capped by clear plastic bags held to the open top of the cartons by a rubber band. Adult parasitoids emerged after another 5 to 8 days. A simple aspirator made of flexible rubber tubing and a glass eye-dropper plugged with cotton wool was connected to a small vacuum pump to remove the parasitoid adults from the cartons. Strips of filter paper smeared with honey were placed in 30-g creamer cups and adult parasitoids were put into these cups. Approximately 20 to 100 adults were introduced into the oviposition cage every week. The

PAGE 71

63 remaining parasitoids were refrigerated at 5 C. These could be used up to 3 weeks after being placed in the refrigerator. After 3 weeks, the insects were discarded if they were not needed for the colony Precautions and care Plants with more than 400 leafminer larvae were not utilized for the parasitoid colony. It was not easy to maintain these plants because the larvae fed into the leaf petioles causing early leaf drop. Also, it was necessary to water the plants daily since they were under greater stress. Not all larvae were parasitized. Many dropped onto the cage floor and pupated. Cages were therefore vacuumed every 4 to 5 days to prevent leafminers from becoming adults and ovipositing on the plants.

PAGE 72

CHAPTER IV FECUNDITY AND LONGEVITY OF Diglyphus intermedius AND ITS CONTRIBUTION TO HOST MORTALITY Introduction Patel and Schuster (1983) had demonstrated that the rates of development of the life stages of I), intermedius were functions of temperature. It is logical to hypothetize that, as with other poikilothermic organisms, fecundity and adult longevity of J). intermedius may also be functions of temperature. This was recognized by Smerage et al. (1980). In formulating the leafminer model those authors stated that the rates leafminer and parasitoid ovipositions may be functions of temperature. However, for simulation purposes, they assigned constant values for oviposition parameters independent of temperature. They had done this for lack of information on the quantitative description of the relationship between temperature and oviposition rates for leafrainers and its parasitoids. Parasitoid fecundity was assumed by those authors to be 2 eggs/parasitoid-day with each host killed yielding 1 parasitoid egg. Parasitoid adults were assumed, for lack of better information, to have a lifespan of 70 degree-days. It was necessary to obtain better estimates of parameters for parasitoid oviposition and longevity as functions of temperature to represent more accurately the real life situation. Experiments to estimate parameters for parasitoid oviposition and adult female longevity as functions of temperature are reported in this chapter. 64

PAGE 73

65 In addition, parasitoid-induced host mortality during the adult female lifespan were also measured. To achieve these objectives, experiments were done in a laboratory utilizing I), intermedius adults and 3rd-instar L_. trif olii larvae reared on tomato foliage. Materials and Methods Five incubators were utilized to obtain 5 constant temperature conditions. A sixth unit was available as a backup in case a principal unit failed. The temperatures were 15.6, 19.4, 23.3, 27.2, and 31.1 C. Fluorescent and incandescent lighting provided a 12 h photoperiod beginning at 0800 h each day. To record temperature and humidity, a 7-day hygrothermograph recorder was placed on the lower of the 2 shelves in each incubator. A moistened filter paper was placed in a 150 x 15 mm. clear plastic petri dish. Twenty tomato leaflet pieces, each containing a single 3rd-instar leafminer larva, were arranged in a single layer and placed on the filter paper. A pair of 1-day-old parasitoids were released into each of 4 petri dishes per each of the 5 incubators. The parasitoids were transferred every day between 0800 and 0900 h to new petri dishes containing 20 additional 3rd-instar leafminer larvae. Dead males were replaced with 1-day-old males. The filter paper inside the dish was kept moist to prevent leaflet pieces from drying. No honey or other carbohydrate source was provided. Live leafminer larvae, empty mines, and dead larvae within mines were counted every day after the parasitoids were transferred to

PAGE 74

66 new petri dishes. Mines containing dead larvae were dissected under a dissecting microscope and larvae with and without parasitoid eggs were counted. The number of eggs laid on each host, as well as parasitoid female longevity, were also recorded. The experiment was replicated 3 times. Temperatures were assigned randomly to the incubators for the first replicate. Temperature assignment was rotated for 2 additional replicates so that an incubator was not assigned the same temperature again. Data were averaged over replicates prior to analyses with the SAS-GLM procedures (SAS 1982). Results and Discussion The mean values for host mortality, parasitoid fecundity, and parasitoid longevity are reported in table 4-1. Table 4-1 Number of L. trif olii larvae killed by and the fecundity and longevity of I), intermedius at different constant temperatures. (+ 1 SD) Temp (C) No. larvae Fecundity Longevity killed (days) 15.6 406.6 (168.4) 194.7 (113.3) 43.5 (8.4) 19.4 387.8 (153.0) 221.7 (105.5) 34.4 (6.9) 23.3 281.8 (93.5) 200.2 (107.7) 22.7 (7.7) 27.2 187.8 (55.0) 126.3 (84.4) 16.1 (3.5) 31-1 128.2 (40.5) 67.4 (50.3) 12.4 (3.4) Host mortality and longevity of D. intermedius decreased with an increase in temperature. The mean fecundity of D. intermedius increased from 194.7 at 15.6 C to 221.7 at 19.4 C. and then

PAGE 75

67 decreased at temperatures greater than 19.4 C. The graphs of regression equations obtained by from mean values for parasitoid fecundity, parasitoid longevity, and parasitoid-induced host mortality for each replicate at each temperature are presented in Figures 4-1, 4-2, and 4-3 respectively. 2 The low r of I), intermedius fecundity, even when a quadratic model was fitted to the observed means, reflect the variations among mean replicate values which in turn reflect variations among individual observations. As there were only 4 observations (only three observations in one case since one female had escaped) in each replicate-temperature combination, the low 2 2 r value was to be expected. R values of greater than .75 were obtained for I), intermedius longevity (Figure 4-2) and parasitoid-induced host mortality (Figure 4-3) when linear regression equations were fitted to the observed means. The results of this experiment can only be utilized to interpret and predict in the 15.6 to 31.1 C temperature range because no measurements were made outside this range. In the 15.6 to 23.3 C range, the parasitoid's fecundity was higher than that of L. trif olii observed by Leibee (1984) on celery. Leibee (1984) reported that L. trifolii laid approximately 24 eggs at 15 C and 182 eggs at 20 C when the flies were given a 10% honey solution as an additional food source. The predicted leafminer fecundity ranged from 3 eggs at 15 C to approximately 31 eggs at 30 C when provided with a 5% sugar water solution (Patel 1981). As temperature increased to 31.1 C, D.

PAGE 76

Figure 4-1 Relationship between JD. intermedins fecundity, F, and temperature, T. ( • : Observation means, n=4)

PAGE 77

69 Figure 4-2. Relationship between D. intermedius longevity, L, and temperature, T. ( • : Observation means, n=4)

PAGE 78

Figure 4-3 Relationship between # hosts killed, Hm, and temperature, T. ( • : Observation means, n=A)

PAGE 79

71 intermedius fecundity declined while that of L. trif olii increased The maximum parasitoid-induced host mortality and oviposition rates as well as the highest ratio of eggs laid/day: hosts killed/ day occurred at 23.3 C (Figure 4-4). The ratio of parasitoid eggs laid/day:hosts killed/day was 0.48, 0.57, 0.71, 0.67, and 0.53 at 15.6, 19.4, 23.3, 27.2, and 31.1 C, respectively. Parasitoidinduced host mortality and oviposition rates at 15.6 C were approximately 24 and 37% less than at 23.3 C. These rates declined by 13.5 and 33.7% when the temperature was increased from 23.3 to 31.1 C. The rate of parasitoid-induced host mortality is apparently less temperature dependent than is the rate of oviposition. Parasitic hymenoptera have panoistic ovarioles and their eggs develop throughout their adult lifespan. The presence of eggs at all stages of development enables a female to oviposit if hosts are available (Fisher 1971). If all the mature eggs are laid all at once, the female can relatively quickly mature those eggs that are only one stage behind (King and Richards 1969). The rate at which these eggs mature would be expected to depend on temperature, since most developmental phenomena are temperature dependent in poikilotherraic organisms. Parasitoid-induced host mortality, on the other hand, is a behavioral phenomenon. It may be temperature dependent to the extent that the parasitoids may be more active and may take less time to locate and kill hosts at higher temperatures. The actual

PAGE 80

15.6 19.4 23.3 27.2 31.1 TEMP (C) figure 4-4 Mean daily D. intermedius -induced host mortality and J), intermedius oviposition at different temperatures.

PAGE 81

73 stinging and paralysis of a host takes only a few seconds, however (Hendrickson and Barth 1978). The arena in which the hosts were placed in the experiment was so small that a parasitoid would not have to spend much time or effort in locating a host. The temperature factor is therefore not likely to have influenced daily host mortality rates as much as it would have affected the daily oviposition rates. The maximum ovipositions/day of 8.5, 9.5, 13.5, 11.5, and 8.5 eggs/female were reached at 14, 11, 6, 5, and 4 day ages of the parasitoids, respectively, at 15.6, 19.4, 23.3, 27.2, and 31.1 C (Figure 4-5). The maximum parasitoid-induced host mortalities/day were 14 hosts killed at age 7 days, 16 hosts killed at age 12 days, 18.5 hosts killed at age 6 days, 17.5 hosts killed at age 5 days, and 16 hosts killed at age 5 days at 15.6, 19.4, 23.3, 27.2, and 31.1 C, respectively (Figure 4-6). In comparison to these maximum ovipositions/day and parasitoidhost mortalities/day for D. intermedius L. trifolii attained maximum daily ovipositions of 35 to 39 eggs on age 4 days at 25 C, age 2 days at 30 C, and age 1 day at 35 C. The maximum L. trifolii ovipositions of 15 eggs in a day occurred at age 5 days at 20 C. Minimal leafminer oviposition occurs at 15 C with a maximum of 2 eggs/female aged 6 days (Leibee 1984). The respective predicted average daily rates for D. intermedius at 15, 20, 25, and 30 C are 5.2, 0.7, 0.3, and 0.34 for host mortality and 2.3, 0.4, 0.2, and 0.2 for oviposition for every L. trifolii egg laid at the maximum rates as reported by Leibee (1984).

PAGE 82

74 ••:: CO -a •H cc o u o o o s I* • e at •> k ^ o. tn — ;• I 3 o o PI < ***.. y a v o /atvn sooa en CO 0) u — • E 3 d> C i-< CO 0> E j= a; 4-1 U -H C > 0) U 3 3 iH CO c >, I sr 1) ki 3 a H

PAGE 83

75 o o o o o 3 a — n < CO T3 0) n CO o jf.--'' : 3 o Q < 'OSt. Ui V -o • e co 3 CD C rH CO CU E .e cu c co O 3 •H CU T3 U CU 3 E -J U CO u I 3 Of A V Q / 03111)1 S1SOH

PAGE 84

76 With average lifespan host mortality and oviposition rates as high as 5.2 and 2.3 times that of the maximum leafrainer oviposition rates at 15 C, it would be expected that I), intermedius alone might be adequate for regulating leafminer populations at low temperatures. I), intermedius oviposition and host mortality rates can be as low as 0.34 and 0.2 times the maximum L^. trif olii oviposition rates at temperatures greater than 20 C. _D. intermedius is at a disadvantage relative to L. trif olii with an increase in temperature above 20 C. Under these circumstances, it might be expected that L^. trif olii populations may increase under warming conditions. This has been observed in the spring tomato season (Schuster personal communication). The results of the experiments reported here may partially explain why leafminers tend to increase in the spring with a rise in temperature.

PAGE 85

CHAPTER V DIURNAL PATTERN IN OVIPOSITION AND HOST MORTALITY Introduction Zehnder and Trumble (1984b) showed that Liriomyza spp. leafminers exhibited circadian patterns in flight activity and larval emergence. Flight activity of leafminers peaked between 0700 and 1100 h. Most puparia and emerging larvae were also collected in the same time period. There have been no studies on the circadian patterns of Diglyphus intermedius and patterns of only a few other parasitic Hymenoptera have been studied. Sugimoto and Ishii (1979) noted that daily host mortality from parasitization and host feeding by Chrysocharis pentheus (Walker) fluctuated periodically during its life when observed in the laboratory under 24 h light conditions. C. pentheus adults parasitized more hosts during the first 6 h and fed on hosts more during the subsequent 3 h in a 24 h period after the parasitoids had been deprived of hosts for 1 day. Nyrop and Simmons (1986) trapped more adults of Glypta fumiferanae (Viereck), a parasitoid that attacks the spruce budworm Choristoneura fumiferana (Clemens), during the 0800 to 2200 h than in any other time of a day. If circadian acitivity patterns of I), intermedius were known, sampling and monitoring for the leafminers and J). intermedius could be optimized. Also, pesticides could be 77

PAGE 86

78 applied at times that would conserve pardsitoids and kill the leaf miners The experiments reported in this chapter were done to determine the influence of the time of day on parasitoid oviposition and parasitoid-induced host mortality. Materials and Methods Five-day-old females were utilized since oviposition and parasitoid-induced host mortality rates peaked on the fifth day at 27.2 C (see chapter IV). Female parasitoids were isolated from the colony upon emergence, paired with day-old males for 1 day, and then placed singly in 150 x 15 mm petri dishes with leaflet pieces containing 30 or more 3rd-instar L. trif olii larvae. Fresh host larvae were provided every day. Females were removed at 0730 h on the 5th day and placed in a new petri dish containing host larvae. For each female tested, 5 petri dishes were prepared. Each petri dish had 20 leaflet pieces with each leaflet piece containing a single 3rd-instar host larva. The petri dishes were held in a refrigerator at 5 C to arrest leafminer development so that no physiological age differences existed between host larvae offered to the parasitoids at different times of the day. The petri dishes were removed from the refrigerator and allowed to equalize to the rearing room temperature for 5 minutes before being utilized for the experiments. The experiments were done in a rearing room maintained at 25-27 C. Fluorescent lighting inside the room was turned on at

PAGE 87

79 0800 h and off at 2000 h. The parasitoids were transferred to petri dishes with fresh host larvae at 0800 h, 1200 h, 1600 h and 2000 h. Fifteen females, 5 on 1 day, 4 on a 2nd day and 6 on a 3rd day were observed. Killed hosts and parasitoid eggs were counted in the manner described in Chapter IV. Result and Discussion The mean number of eggs laid/female in a 24 h period was 14.9 (Table 5-1). This is similar to 11.8 observed for 5-day-old females at 27.2 C in the previous experiments (see Chapter IV). The mean number of hosts killed/female was 14.0 compared to 17.5 as described in chapter IV. The number of eggs laid/female varied from 8 to 23 and the number of hosts killed/female varied from 11 to 24. Table 5-1 Relationship between time of day and host mortality and oviposition of 5-day-old female I), intermedius at 25-27 C.(+ 1 SD) TIME OF DAY OVIPOSITION HOST MORTALITY 0800-1200 10.3 (3.2) 7.9 (1.7) 1200-1600 2.1 (1.9) 2.4 (1.8) 1600-2000 2.4 (1.4) 3.6 (2.2) 2000-0800 0.1 (0.4) 0.1 (0.4) TOTAL 14.9 (4.7) 14.0 (3.4) Most of the parasitoid oviposition and parasitoid-induced host mortality activity occurred between 0800 and 1200 h unlike the host killing and ovipositional activity of C_. pantheus which preferred

PAGE 88

80 to oviposit in the first 6 hours and host-feed during the subsequent 3 hours in a 24-h period (Sugimoto and Ishii 1979). The parasitoids were not active during the scotophase. Only two of the females were even slighlty active at night. None of the other parasitoids killed hosts or laid eggs at night. Parasitoids were not moving and were usually found underneath the leaflet pieces or away from the leaflets when the petri plates were removed for observation at the end of the dark period. In either case, the female parasitoids were at some distance from host larvae when resting. They may, therefore, need vision to search and locate hosts. During the photophase, there was greater parasitoid oviposition and parasitoid-induced host mortality during the 0800 to 1200 h compared to the 1200 to 1600h and the 1600 to 2000 period. The 5-day-old females had been previously conditioned in the laboratory to have fresh host larvae available to them at the beginning of the day and this may explain why more hosts were killed during the first part of the photophase. In the field, with most mature larvae emerging by 1100 h (Zehnder and Trumble 1984b), it would be to the advantage of the parasitoid to be most active during the early part of the day. Charlton and Allen (1981) also showed that most mature L. trifolii larvae exit from the leaves during the latter part of the morning. Apparently, I), intermedins circadian behaviour is linked to the timing of the emergence of mature host larvae from

PAGE 89

81 the leaves. This type of behavior would enable the parasitoids to utilize the most mature and, hence, the largest host larvae. In addition, although presumably equivalent host larvae were offered at 1200 and 1600 h, parasitoid oviposition and parasitoidinduced host mortality activity were much reduced. Thus, laboratory conditioning alone does not fully explain the observed behavior. If chemical insecticides were to be utilized for leafminer control, it may be best to do so during the period of least parasitoid activity, i.e., during the night hours or in the afternoon or evening. This would allow some adult parasitoids that may be resting away from the plants to escape insecticidal mortality. Spraying at these times may conserve parasitoids but may also be less effective against leafminer adults since their least active period coincides with that of J), intermedins Under such constraints, it may be advisable to use leafminer larvicides. Schuster and Everett (1983) showed that cyromazine (Trigard ) was an effective leafminer larvicide and that Trigard residues on tomato were least lethal to leafminer adults.

PAGE 90

CHAPTER VI HOST STAGE PREFERENCE Introduction L. trif olii is attacked by different parasitoids in the field, and it is important to know if any niche division occurs, at least between the major parasitoids, so that competition among them for hosts is understood. Leraa and Poe (1979) showed that Opius sp., a larval-pupal endoparasitoid and Chrysonotomyia sp., a larval endoparasitoid, attacked 72-h-old and 120-h-old L_. sativae larvae, respectively. The hosts were reared at 20 7.9 C. The L_. trif olii larval stage lasts about 288 hours at 20 C (Leibee 1984). The larval period of _L. sativae is probably comparable and hence, 120-h-old larval L,. sativae are likely to be in their 2nd-instar Hendrickson and Barth (1978) showed that D. intermedius preferred to oviposit on 3rd-instar rather than 2nd-instar Liriomyza trif oilearum Spencer larvae, although it fed on all three instars under laboratory conditions. No parasitoids were reared from 2nd-instar host larvae. Nearly 19% of field collected larvae of the alfalfa blotch leaf miner, Agromyza f rontella (Rondani), were parasitized by D. intermedius This parasitoid preferred 3rd-instar hosts although 23.7% of the parasitized hosts were 2nd-instar larvae. Hendrickson and Barth (1978) attributed the failure of I), intermedius in parasitizing 2nd-instar L. trif oilearum larvae to the smaller host size. 82

PAGE 91

33 The preference of a parasitoid for a particular host size is a fairly common phenomenon. Hyposeter exigua (Viereck) laid 2.3 and 1.3 eggs per 24 h in 1st and 2nd instar, respectively, of Trichoplusia ni (Hubner) larvae. Oviposition declined for each subsequent instar (Smilowitz and Ivantsch 1975). Successful parasitism by the braconid Cardiochiles nigriceps (Viereck) was least in fifth-instar larvae of Heliothis virescens (F.) and H.. subf lexa (Guernee) (Lewis and Vinson 1971). The experiments reported here were done to test if different aged L. trif olii larvae are equally acceptable to JD. intermedius for ovipositional and non-ovipositional purposes. Materials and Methods Leafminers and parasitoids were reared as previously described in chapter III. Twenty 96-h-old L. trif olii larvae (referred to as young larvae from hereon), and 20 120-h-old larvae (old larvae), in cut tomato leaflet pieces, were arranged in each of 5 150x15 mm petri dishes containing water-moistened filter papers. Five young larvae were alternated with 5 old larvae in each of the radially sectored octants. Two additional plates, one with 20 3rd-instar larvae only, and another with 20 2nd-instar larvae only, were also maintained. A single 5-day-old mated female parasitoid was introduced into each petri dish. The adults were subsequently removed after a period of less than 24 hours and the number of host larvae killed and the number of parasitoid eggs laid were counted. The experiment was replicated 4 times, once on each of 4 days. The

PAGE 92

84 exposure period varied for the 4 replicates. The exposure time was the same for each of the petri dishes placed on any particular day. The experiments were done in the rearing room at the prevailing conditions described in chapter V. Results and Discussion The results of the experiments are summarized in table 6-1. Significantly more old larvae were killed and parasitized Table 6-1 Relationship between L_. trif olii larva age and mean no. of eggs laid and hosts killed by JJ. intermedius (+ 1SD) Choice or #Hosts Killed #Eggs Laid No choice 96h 120h 96h 120h Choice No Choice 3.2 (3.0) 10.8 (6.6) 6.0 (3.2) 5.0 (5.3) 1.3 (1.7) 5.3 (5.1) 5.8 (3.6) 5.5 (5.5) than were young larval hosts when the parasitoids were given a choice (paired t-tests for pooled data, p >.001). Young larvae were as acceptable as old larvae for oviposition and host mortality was the same for larvae of both ages when the parasitoids did not have a choice of different aged hosts. J), intermedius could utilize synchronous as well as asynchronous host populations under such circumstances. This fact is important to recognize since the application of stage-specific insecticides such as adulticides may tend to synchronize the affected populations for at least a short period.

PAGE 93

85 There may be long term effects on host and parasitoid population densities with synchronous host populations, however. If only small host larvae were present for a long period, parasitoids emerging from small larvae may be smaller. Trichogramma evanescens Westwood were smaller when reared on smaller sized Sitrotoga cereallella (Olivier) host eggs (Salt 1940). The advantage to J), intermedins of attacking larger hosts can be understood if we consider the relationship between hosts and all parasitoids of leafminers. Ectoparasitoids sting their hosts. They cannot rely upon host growth to supply food for their offspring. Endoparasitoids keep their hosts alive and can rely on host growth for their offsprings' growth. Askew (1975) found no early-attacking ectoparasitoids of Phyllonorycter spp. leafminers while Syntomaspis spp. and Eurytoma spp. ectoparasitoids attacked young hosts in oak galls. In oak galls, ectoparasiotids could supplement their diet with gall material whereas there is no provision for phytophagy in leafmines. Askew (1975) suggested that, to a certain extent, the divergence in host selection, with endoparasitoids attacking young leafminers and ectoparasitoids attacking older leafminers, reduced the likelihood of ectoparasitoids attacking a host already containing an endoparasitoid The partitioning of hosts between ectoparasitoids and endoparasitoids is not well defined, however. I have reared a Chrysontomvia sp. adult from a field collected Diglyphus sp.

PAGE 94

86 pupa. I have also observed I), intermedius parasitizing host larvae that have already been parasitized by another individual D. intermedius This suggests that they may not be able to distinguish between parasitized and nonparasitized host larvae or may parasitize hosts even after recognizing the hosts as having been parasitized by others. If this is true, they may just as readily attack endoparasitized host larvae. In summary, jD. intermedius kills and oviposits on more 3rd-instar larvae than 2nd-instar laravae if both ages of larvae are equally available to the parasitoid.

PAGE 95

CHAPTER VII IMPACT OF ALTERING HOST AND PARASITOID DENSITY ON DAILY PARASITISM RATE Introduction The rate of parasitism in the Sraerage et al. (1980) model was a function of host and parasitoid densities. As mentioned earlier, Figure 1-1 shows the assumed relationship between the rate of parasitism and host and parasitoid densities at a constant temperature. Equations (7) and (8) : Y = C 2 .Kp.n 2 (1 expCnj/Kh)) (7) Y = C 2 .Kp.n 2 (nj/Cnj + Kl)) (8) are alternative closed-form equations generating families of curves, like Figure 1-1, as already mentioned in Chapter II. In computer simulations of the model, Musgrave et al. (1980) 2 assumed Kh and Kp to be 5 larvae/m and 2 larvae/adult-day. The experiments reported in this chapter were done to obtain a solid description of the rate of parasitism as a function of leafminer and parasitoid densities at a constant temperature. All experiments were done in the laboratory and required the confinement of parasitoids to cages in which leafminer infested plants were kept. Materials and Methods Several limitations were encountered in the experimental design. Ideally, observations on parasitoids should be made in a 87

PAGE 96

88 large arena which allows the parasitoids to move naturally in and out of the arena. This would enable the parasitoids to behave naturally and to have random access to hosts, also, the likelihood of encountering the same host again would not be due to confinement to a restricted arena. Arena size would certainly be important if the period of observation was large and parasitoids were confined to a small region. Because of the small size of adult parasitoids, it is not possible to monitor parasitoid densities in a large arena with no confinement facilities, however. Nor is it possible to maintain uniformity in leafminer and parasitoid densities between observation plots. It is also not possible to maintain environmental conditions constant. To overcome these problems, a compromise was made. Female parasitoids were confined in 67 x 67 x 67 cm cages within a controlled environment room for 12 h observation periods. There were several advantages to using cages. I), intermedius density per cage was easy to record and manipulate, and other parasitoid species were excluded. As the cages used were relatively small, it was possible to keep them in a controlled environment room and hence at constant temperature and uniform light conditions. One, one-month-old greenhouse grown tomato plant, with approximately 2000 to 3000 sq cm leaf area, was placed in each cage. This amount of leaf area was sufficient to sustain as many as 200 or more leafminer larvae, and a wide range of leafminer densities was possible. With cages, it was possible to alter leafminer densities

PAGE 97

89 and maintain uniformity of larval age to a much greater extent than if the experiments were done in the field. Tomato host plants, cv. Hayslip, used in the experiments were grown and maintained in the greenhouse as described in Chapter III. The plants selected were approximatley 30 days post-transplant, 50 to 60 cm tall, had about 10 to 13 leaves, and were just beginning to blossom. Plants of this size just fit into the 67 x 67 x 67 cm cages and allow the maximum amount of leaf area possible on one plant confined in the cage. The plants selected had approximately 2000 to 3000 sq cm leaf area, although on ocassions plants were smaller or larger. Plants were moved from one holding area to another with care to minimize the damage to leaves during handling. After selection, 6 plants were moved to the refrigerated truck body which housed the leafrainer colony (see Chapter III). Three tomato plants were exposed to leafminer adults confined in each of two 67 x 67 x 67 cm cages. The number of leafminer adults confined in the oviposition cage was varied from as few as 10 to as many as 50. The exposure period was also varied from as little as 1 minute to 4 hours. It was possible to obtain various leafminer larval densities by manipulating the number of adult flies and the exposure period as just described. After exposure to leafrainer adults, the plants were moved to holding cages and maintained in the manner described in Chapter III. After 5 days, plants containing 3rd-instar larvae were removed at 0800 h and taken to the insectary rearing room for exposure to different densities of parasitoids. Of the 6 plants initially

PAGE 98

90 exposed to the flies, 5 plants were selected for exposure to the parasitoids. The plant most different from the others in height or quality or leafrainer density was discarded. One plant was placed in each of five 67 x 67 x 67 cm cages. The parasitoids used in these experiments were collected 4 days earlier on emergence and 40 females and 20 males were confined, 10 females and 5 males to a dish, in 150 X 15 mm petri dishes. Each dish contained 40 to 60 3rd-instar leafrainer larvae in tomato leaflets and the larvae were replaced daily. On the 4th day, the parasitoids were isolated singly in size 00 gelatin capsules and sexed; males were discarded. One, 2, 3, 4, or 5 parasitoid females were then released into each of the 5 cages holding the plants. After 12 h plants were removed from the cages, shaken to dislodge parasitoids and transferred to the laboratory adjoining the rearing room at 2000 h. Plant height, age, number of leaves and leaflets per plant were counted. A Licor leaf area meter was utilized to record the total leaf area, the area of leaflets containing mines, and the area of leaflets containing parasitized mines. A record was kept of the number of empty, live and parasitized mines in each of 3 leaf categories. Leaves were classified as old (usually the 1st 3 leaves which showed signs of yellowing), fully expanded leaves (the majority of the leaves) and unexpanded leaves (usually the top 3 to leaves although occasionally there may have been several more in a small, tight bundle at the plant apex). Mines containing paralyzed

PAGE 99

91 larvae were dissected and the number of eggs laid on each larva was recorded The experiment was repeated several times. The assignment of parasitoids to the cages was rotated so that the same cage had the same parasitoid density every 5th time. Because the plants varied in leaf area, and leafminer oviposition could not be made uniform on each plant, it was not possible to regulate leafminer density as freely as parasitoid density. The number of mines on each plant, and hence in each cage, could have been kept the same by destroying some larvae on each plant. However, this would not have been appropriate for 2 reasons. Parasitoids searching a greater leaf area with the same number of mines/cage would take longer than for the same number of mines on a smaller leaf area. Also, destroying leafminers may elicit its own response from the parasitoids if the parasitoids cue in to host plant damage in locating leafminers. The SAS NLIN pocedure (SAS 1982) was utilized to decribe the relationships between leafminer density and rate of parasitism at different parasitoid densities. The same procedure was also utilized to describe the relationship between leafminer density and parasitoid-induced host mortality at different parasitoid densities. 2 Chi square (X ) tests were done to determine if there was any effect of altering either leafminer or parasitoid density on the numbers of parasitoid eggs on paralyzed host larvae. Results and Discussion Two rate of parasitism models, the rectangular hyperbola and the negative exponential equations proposed earlier by Smerage et

PAGE 100

92 al. (1908), fit equally well the observed values of rates of parasitism when plotted against leafminer densities in the 0.00-0.06 larvae/sq cm range. The exponential equation obtained using the SAS nonlinear regression techniques was Y = Kp.n 2 .(l exp(ni /Kh)) (r 2 = .79) (T = 25-27 C) (4) where Y, the rate of parasitism, is the number of hosts parasitized/ day. With a value of 7.3908, Kp is a constant with the units, number of hosts parasitized/parasitoid-day Densities, and n^ are, respectively, parasitoids/cage and leafminer larvae/leaf area (cm ). Another constant, Kh, is,' leaf miner hosts/leaf area (cm ). Results of the experiments also were utilized to formulate the similar but parametrically different equation (5) below expressing parasitoid-induced host mortality as a function of host and parasitoid densities at 25-27 C: Z = Cp.n 2 (l -exp(ni /Ch)) (r 2 = .78) (T = 25-27 C) (5) where Z, the parasitoid-induced host mortality rate, is the number of hosts killed/day. With a value of 9.2064, Cp is a constant with the units number of hosts killed/parasitoid-day and Ch is 0.0165 leafminer larvae/leaf area (cm 2 ). The predicted relationship between host density, parasitoid density and the rate of parasitism utilizing the negative exponential equation is depicted in Figure 7-1. The relationship between rate of parasitoid-induced host mortality and host and

PAGE 101

93 u a
PAGE 102

94 >>> 4J •H c O 4J 09 O EC -a 0J U a • T3 CN C C H 1 -c M •H CJ C •H 4J u H •H 0) tn (0 c u CJ a a c a a H c 0 CD 4J H • B) c u a • a. >* , 4J Q. H H — J= cn U c U o 0 •H E u ta 4J r— | n V c Xep/paiipi s^soq # rsi I r-
PAGE 103

95 parasitoid densities is illustrated by the family of curves in Figure 7-2. The predicted rate of parasitism at a particular parasitoid density is always lower than the predicted rate of parasitoidinduced host mortality at that parasitoid density. This confirms the previous observation that I), intermedins killed more hosts than it parasitized. However, the proportion of hosts killed that are also parasitized is not the same at all leafminer densities (Table 7-1). Table 7-1 Proportion of observations with 100% parasitization of killed hosts at different leafminer densities. Leafminer larvae/ # of observations # of observations sq cm of leaf area with host mortality with 100% parasitization of killed hosts 0.00 0.0099 38 24 0.01 0.0199 19 5 0.02 0.0600 33 3 In the 0-0.0099 host larvae/sq cm range, 63% of all observations had a 100% parasitization rate of killed hosts while only 26% and 10% of the observations had 100% parasitization of the killed hosts when the leafminer densities were 0.01-0.0199 and 2 0.02-0.06 larvae/leaf area (cm ), respectively. I), intermedius does not always lay a single egg on each host, as was assumed by Smerage et al. (1980). The observed

PAGE 104

96 distribution of parasitoid eggs on host larvae at different leaminer densities is shown in Table 7-2. Table 7-2 Frequency distribution of killed L. trif olii in classes of different I), intermedins egg densities Leafrainer # Eggs/killed host larva Total density/sq cm leaf area 0 1 2 3 4 4+ 0.00 0.0099 19 108 60 19 11 16 233 0.01 0.0199 42 189 74 31 8 6 350 0.02 0.0299 49 163 50 11 10 3 286 0.03 0.0399 45 180 46 10 2 2 285 0.04 0.0499 43 135 43 9 1 0 231 0.05 0.0600 16 40 13 5 0 1 75 0.00 0.0600 214 815 286 85 32 28 1460 Strong evidence in favor of a relationship between the number of parasitoid eggs per host larva and host density was shown when a 2 2 X test was done using the data in Table 7-2 (X = 67.62, 2 significant at p < 0.001, df = 25). A X test done with observations grouped according to the parasitoid density was not significant at p < 0.01 level (X 2 = 15.6, df = 20). It was possible to make some deductions regarding the effect of spatial heterogeneity of host larvae on parasitism rate although this experiment was not designed to specifically test for it. In 68 of 79 observations, the leaflet with the greatest number of mines was encountered by parasitoids as evidenced by the parasitization of at least 1 larva on that leaflet.

PAGE 105

97 In another 4 observations, the parasitoids were similarly attracted to the leaflet with the next highest leafrainer density. In the remaining 7 observations, the maximum number of mines on the leaflet was 4. The parasitoids were perhaps equally attracted to leaflets with 4 or fewer larvae at these densities, particularly if many leafrainer infested leaflets were close to each other. This was not tested, however, as it would have required determining the area of each leaflet and measuring distance between individual larvae A small leaflet with 10 larvae may not have the same attraction to a parasitoid as a large leaflet with the same number of larvae. If the cue, to which the parasitoids responsed, is a volatile chemical, there would be greater concentration of the chemical around a smaller leaf area than around a bigger area. There are many instances of parasitoids that have been reported to be strongly attracted by chemicals produced by their hosts (e.g Monteith 1956) or by the damage caused to the plants by the hosts (e.g Vinson 1975). In this chapter, the rates of parasitism and parasitoidinduced host mortality were shown to be functions of parasitoid and host density. These relationships are represented by equations (4) and (5) for parasitism and parasitoid-induced host mortality, respectively

PAGE 106

CHAPTER VIII MULTIPLE OVIPOSITION AND ITS INFLUENCE ON PROGENY SURVIVAL RATE Introduction It was shown in the experiments reported in the previous chapters that I), intermedius does not always lay just a single egg on a host. On one occassion, there were AO parasitiod eggs on a host. Not all those eggs could possibly survive to the adult stage. Hendrickson and Barth (1978) reported 2 to 5 eggs occasionally developing on a single host larva. I have reared 5 parasitoid adults from a parasitized leafminer larva collected in the field. The number of parasitoid eggs that successfully reach the adult stage may depend on their density on a host larva. The number of surviving progeny of parasitoids may influence the host and parasitoid population densities in subsequent generations. In this chapter, experiments done to determine the survival of progeny of parasitoids, as influenced by the density of parasitoid eggs on 3rd-instar host larvae, are described. Materials and Methods The leafminers and parasitoids used in this experiment were reared in the manner described earlier in Chapter III. Parasitized 3rd-instar leafminer larvae were dissected from mines obtained from the colony and placed in artificial mines constructed in the manner described by Patel and Schuster (1983). Parasitoid eggs, less than 24 h old, were dissected from mines containing parasitized hosts 98

PAGE 107

99 and placed next to the larvae in the artificial mines. One, 2, 3, or 4 parasitoid eggs were placed next to each host larva. A cover slip was placed on the mine and the entire mine was kept in a 100 x 15 mm petri dish containing a water-moistened filter paper. The water-moistened filter paper provided relatively high humidity to prevent the eggs and subsequent stages of the parasitoid from dessicating. The number of host larvae utilized was 12 with 1 egg/ larva, 6 with 2 eggs/larva, 4 with 3 eggs/larva, and 3 with 4 eggs/ larva. There were 12 eggs at each egg density. The petri dishes were maintained in the insectary rearing room with temperature, lighting and humidity conditions as previously described in Chapter V. Filter papers in the petri dishes were moistened daily. The petri dishes were kept until no more adult parasitoids emerged from the mines. The experiment was replicated 5 times. The number of parasitoids emerging from each mine was recorded. SAS ANOVA pocedure (SAS 1982) was utilized to determine if any relationship existed between the initial total number of eggs placed at each egg density and the total number of eggs reaching the adult stage at each density. Results and Discussion Results of the experiment are presented in table 8-1. The initial density of eggs/larva is in the 1st column of the table. The mean number of eggs that reached the adult stage, at each egg density, are in the 2nd column. The data in the 2nd column was

PAGE 108

100 transformed to obtain the survivorship of eggs to adults/host larva and is shown in the final column. Table 8-1 Survival of D. intermedius eggs to adult stage when placed at different densities on L. trif olii larvae Initial egg density/ Mean # eggs surviving Survivorship host larva (12 eggs to adult stage of eggs/larva at each density) Note: means in a column followed by the same letter are not significantly different (Duncan's multiple range test). At the densities studied, observed survivorship of eggs/larva reached a maximum of 1.35 when 3 eggs were initially placed on a host larva and declined to lower survivorship when the initial egg density was either higher or lower than 3 eggs/larva. Some of the mortality may have resulted from the parasitoids being kept in artificial mines but this would occur at all egg densities. It would not be appropriate to formulate a regression equation to mathematically describe the relationship between parasitoid survivorship and initial parasitoid egg density based on the limited data reported here if the survivorship of parasitoids is affected by the type of mines. In natural mines, higher survivorship might be expected at all egg densities. 1 2 3 4 9.6a 6.6b 6.0b 3.0c 0.80b l.lOab 1.35a 0.98ab

PAGE 109

CHAPTER IX CONCLUSION Hendrickson and Barth (1978) showed that JD. intermedius females produced an average 40.2 progeny on L_. trif oliearum larvae and lived from 3 to 4 weeks. In the experiments described in chapter IV, I have shown that the mean values of I), intermedius fecundity, and longevity and of I), intermedius induced mortality of L_. trif olii are temperature dependent. I), intermedius fecundity increased from 194.7 eggs at 15.6 C to 221.7 eggs at 19.4 C and then declined to 126.3 eggs at 27.2 C and 67.4 eggs at 31.1 C. The maximum number of host larvae killed was at 15.6 C, the lowest temperature studied. The longevity of female I), intermedius was also greatest at 15.6 C, when the females lived for 43.5 days. At 31.1 C, the highest temperature studied, females lived 12.4 days. Equations describing the relationship between temperature and d. intermedius fecundity, longevity, and J), intermedius induced host mortality are reported in Chapter IV. _D. intermedius fecundity is much higher than leafminer fecundity on tomato reported by Patel (1981) and observed by Leibee (1984) on celery in the 15.6 to 23.3 C range. At temperatures greater than 23.3 C, D. intermedius fecundity declined while that of L. trif olii on celery and tomato continued to increase. The average daily rates of J), intermedius -induced host mortality and JJ. intermedius fecundity on leafminers reared on tomato were 5.2 and 2.3 times, respectively, those of the fecundity of L,. 101

PAGE 110

102 trif olii on celery as calculated from Leibee's (1984) data. The peak daily rate of oviposition by L. trif olii was between 35 and 39 eggs on celery in the 25 to 30 C temperature range and this peak daily rate was realized in the first 1 to 4 days after the flies emerged (Leibee 1984). In comparison, peak J), intermedins oviposition rate ranged from 8.5 eggs/female-day to 13.5 eggs/female-day and host mortality ranged from 14 to 18.5 larvae/parasitoid female-day in the 19.4 to 31.1 C temperature range. Not only are peak I), intermedins oviposition and host mortality rates lower, they also occur later than the peak of leafminer oviposition. Peak rates of D. intermedins oviposition and host mortality do not occur until at least the 4th day after the parasitoid adults emerge. At 15 C, L. trifolii peak oviposition rate reported by Leibee (1984) was 2 eggs/femaleday while a D. intermedius female laid 8.5 eggs/day and each female laid a maximum of 14 eggs/day at 15.6 C on leafminers reared on tomato. If temperature were the only determining factor, D. intermedius should be able to limit leafminer populations at low temperatures even on host plants, such as celery, that allow high leafminer population growth rates. At higher temperatures, particularly at 31.1 C, leafminer populations can be expected to increase at a faster rate than D. intermedius populations. D. intermedius was primarily active in killing hosts as well as ovipositing on them during the first 4 h of a 24 h period

PAGE 111

103 with 12 h of light and 12 h of dark. There was an almost complete absence of oviposition and host killing activity during the scotophase Twice as many 120-h-old JL. trif olli larvae as 96-h-old larvae were killed by I), intermedins when larvae of both ages were made available to D. intermedius at the same times and 4 times more eggs were laid on 120-h-old larvae than on 96-h-old larvae. When larvae of either age group were available alone, there was no difference between the two groups in the number of hosts killed or the number of parasitoid eggs laid. Beddington et al. (1978) proposed that good biological control agents should impose density dependent host mortality. While most researchers report a density dependent relationship between the rate of parasitism and host density (e.g., Hassell 1968), others found alternative relationships between parasitism rate and host and parasitoid densities. Waage (1983) found that among patches of different densities of the host, Plutella xylostella (L.), Diadegma spp. ichnuemonids made the most frequent and longest visits to crop plants with the highest densities of the host insects but the increased total host handling time limited the oviposition rate at high host densities. Parasitism/patch as a consequence was density independent. Morrison and Strong (1981) showed that the aggregation of Cepahaloleia consanguinea Baley (Chrysomelidae) eggs on Heliconia imbricata (Kuntze) (Heliconiaceae) leaves improved their chance for escaping attack by eulophids and trichograramatids to the extent that

PAGE 112

104 an inverse relationship existed between the overall probability of parasitism and the number of hosts on each leaf. I found that the relationship between parasitism/day and the densities of L. trif olii and I), intermedius can be described by the equation: Y = Kp n 2 (1 expC-nj/Kh) (r 2 = .79) (T = 25-27 C) (4) where Y, the parasitism rate, is the number of hosts parasitized/day. With a value of 7.3908, Kp is a constant with the units number of hosts parasitized/parasitoid-day Densities and are, respectively, parasitoids/cage and leafminer larvae/leaf area (sq cm). Another constant, Kh, is 0.0144 leafminer larvae/leaf area. In the equation : Z = Cp n 2 (1exp(-n 1 /Ch) (r 2 = .78) (T = 25-27 C) (5) Z is the parasitoid-induced host mortality rate, the number of hosts killed/day. The constant Cp is 9.2064 hosts killed/parasitoid-day and Ch is 0.0165 leafminer larvae/leaf area. These results were obtained with data from laboratory studies involving caged insects and only included parasitoid densities ranging from 1 to 5 females/cage and leafminer larval densities ranging from 0 to 0.06 larvae/sq cm leaf area. I had observed that, as the leafminer density was increased, the number of host larvae that were killed but not subsequently parasitized also increased

PAGE 113

105 There was a trend toward more J), intermedius eggs/killed L,. trif olii larva with decreasing host density. The number of parasitoid eggs/killed host larva increased from 1.14 eggs/larva in the 0.05 to 0.06 leafminer larvae/sq cm leaf area range to 1.82 eggs/larva in the 0 to 0.0099 leafminer larvae/sq cm leaf area range At low leafminer densities (0 to 0.0099 larvae/sq cm leaf area), approximately 20% of killed host larvae had 3 or more parasitoid eggs. The number of parasitoid eggs/killed host larva did not seem to depend on the density of searching parasitoids, as a X 2 test showed no significant relationship between these variables. The frequency of survival to adulthood of parasitoid eggs depended on their density on host larvae. An average 0.8, 1.1, 1.35 and 0.98 eggs became adults when 1, 2, 3, and 4 eggs were placed on each host larva respectively. Multiplying the frequency of different egg densities on host larvae, as shown in Table 7-2, with the eggs survivorship shown in Table 8-1 would give an estimate of the number of parasitoid eggs reaching the adult stage at various leafminer larval densities. For example, with 0 to 0.0099 host larvae/sq cm leaf area, 8 out of every 100 hosts killed had 0 parasitoid eggs, 46 had 1 egg each, 26 had 2 eggs each, 8 had 3 eggs each and the remaining 12 hosts had 4 or more eggs deposited on each of them. If the number of eggs at each egg density is multiplied with suvivorship of the eggs at that density, 89 parasitoid adults/initial 100 hosts can be estimated to emerge. The number of parasitoids expected to emerge

PAGE 114

106 from 100 host larvae killed at various leafminer densities is shown in Table 9-1. Table 9-1 Predicted number of parasitoid eggs Ke reaching the adult stage for every 100 hosts killed. Leafminer density (# larvae/sq cm leaf area) # Parasitoids reaching the adult stage = Ke 0.00 0.0099 0.01 0.0199 0.02 0.0299 0.03 0.0399 0.04 0.0499 0.05 0.0600 89 82 74 64 61 57 Plotting a graph of estimated K g values versus mid-range values of the leafminer density classes was used to estimate the linear equation Ke = 93 SOOn^ It should be noted that leafminer density indirectly determines the number of parasitoids in the next generation. Patel (1981) showed that temperature also influenced the number of parasitoid eggs reaching the adult stage. At 32.2 C, 88.3 % of the eggs failed to reach the adult stage while at 15.5, 21.1, and 26.7 C, about 50 % of the eggs failed to become adults when these eggs were placed on leafminer larvae in artificial mines. Survivorship is expected to be higher in natural mines, but mortality is still to be expected to be dependent on temperature. Parasitism rate was studied in a rearing room where the temperature was maintained at 25-27 C and using 4-day-old parasitoids only.

PAGE 115

107 These are ideal conditions of temperature and age of parasitoids to estimate maximum rate of parasitism. Parasitoid age and temperature affect the rate of parasitism as described in Chapter IV. Altering temperature and utilizing parasitoid females younger or older than 4 days may reduce the observed parasitism rates. It is, therefore, necessary to establish, through further experimental work, the effects of temperature and parasitoid age on daily parasitism rate as a function of host and parastoid densities. As a next step, to better understanding JJ. intermedins and L. trif olii population dynamics, sex ratio studies need to be done. A potentially important component is excluded if the effects of host and parasitoid interactions on parasitoid sex ratio are not incorporated into host-parasitoid interaction studies (Hassell and Waage 1984). Sex ratios of the progeny of hymenopteran parasitoids have been shown to be influenced by factors such as host species, host size, host densities, and parasitoid denities (e.g. Flanders 1942, Kochetova 1978, Waage 1982). I found that virgin JD. intermedius females produced only male progeny while mated females (paired singly with males for 1 day before testing) produced approximately equal numbers of males and females when each female was provided with 20 3rd-instar host larvae. D. intermedius can thus be said to be arrhenotokous and capable of producing male offspring from unfertilized eggs and females from fertilized eggs. Consequently, they can exert considerable control over the sex ratio of their progeny.

PAGE 116

108 I), intermedius is only one of 3 common parasitoids of leafminers on tomato in Florida. Studies similar to those described for I), intermedius should also be made for Opius spp. and Chrysonotomyia spp. parasitoids of leafminers. In addition, field studies should be made to validate or repudiate the results obtained in the laboratory studies. Only then could an IPM program based on host-parasitoid interactions be considered feasible The effect of Opius spp. parasitoids on leafminer populations and methods for sampling leafminer and parasitoid populations are presently being investigated in Florida by other workers. These studies would compliment the current work on I), intermedius Further areas of investigation should include field validation of J), intermedius and Opius spp. laboratory studies, laboratory and field studies on Chrysonotomyia spp., and investigations of the mechanisms and factors influencing dispersal of these insects in crop fields and surrounding areas. Because of the dynamic nature of interactions between organisms and the ability of organisms to respond to environmental changes through competitive displacement, natural selection, evolution, dispersal and other biological phenomena, results of studies covering a short period of time and utilizing a narrow gene pool may not be universally valid. Leafminers have been shown to be susceptible to many insecticides only to become resistant to them after a period of continued use (see Chapter II). With the current

PAGE 117

109 state of IPM on tomato, it is unlikely that the use of chemical pesticides will be eliminated. The need for continued use of chemical pesticide and the realization that additional studies are required to describe more accurately the host-parasitoid interactions, neccesitates consideration of alternative leafminer management strategies. It is possible, that under Florida conditions, leafminer regulation by D. intermedius is ineffective at high temperatures which persist for a long period during the growing season. If this is indeed the case, then exploration of warmer climates for leafminer parasitoids may provide alternative parasitoid species or strains that could be introduced into Florida and could give better leafminer population regulation at higher temperatures.

PAGE 118

LITERATURE CITED Allen, P. 1956. Observations of the biology of some Agromyzidae. Proc. R. Entomol. Soc. 31:117-31. Arzone, A. 1979. L Agromyzidae nearctico Liriomyza trif olii (Burgess) nuovo nemico di Gerbera in Italia. Inform. Fitopatol. 29(3):3-6. Askew, R. R. 1975. The organization of chalcid-dominated parasitoid communities centered upon endophytic hosts. Pages 130-154 in P. W. Price, ed. Evolutionary stratgies of parasitic insects and mites. Plenum Press, New York. Barfield, C. S., and Jones, J. W. 1979. Research needs for modeling pest management systems involving defoliators in agronomic crop systems. Fla. Entomol. 6:133-7. Bartlett, P. W., and Powell, D. F. 1981. Introduction of American serpentine leaf miner, Liriomyza trif olii into England and Wales and its eradication from commercial nurseries, 1977-81. Plant Pathol. 30:185-93. Beddington, J. R., Free, C. A., and Lawton, J. H. 1978. Modeling biological control: On the characteristics of successful natural enemies. Nature. 273:513-19. Broadbent, A. B. 1982. Liriomyza trif olii on chrysanthemums in Ontario greenhouses. Pages 90-100 in S. L. Poe, ed. Proc. 3rd. Ann. Indus. Conf. on the Leafminer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria, VA. 216pp. Broadbent, A. B. 1984. Liriomyza trifolii on chrysanthemum in Ontario: Research update. Pages 41-49 .in Proc. 4th. Ann. Indus. Conf. on the leafminer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria. VA. 191pp. Brogdon, J. E., Marvel, M. E., and Mullin, R. S. 1970. Commercial vegetable insect and disease control guide. Univ. Fla. IFAS Veg. Crops. Ext. Ser. Circ. 193G. 44pp. Brooke, D. L. 1980. Costs and returns from vegetable crops in Florida. Season 1978-79 with comparisons. Univ. Fla. Economic Information Report No. 127. 110

PAGE 119

Ill Brown, R. D., and Dybas, R. A. 1982. MK-936: A novel miticide/ insecticide for control of Liriomyza leafminers. Pages 59-61 in_ S. L. Poe, ed. Proc. 3rd. Ann. Indus. Conf on the Leafrainer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria, VA. 216pp. Burgess, E. 1880. The clover Oscinis ( Oscinis trifolii Burgess (n. sp.)). Report of the Commissioner of Agriculture for the year 1879. USDA, Washington. p200-201. Cantliffe, D. J. 1985. Introductory remarks. Page 1 iri Florida Tomato Institute. Univ. Fla. Veg. Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2. Charlton, C. A., and Allen, W. W. 1981. The biology of Liriomyza trifolii on beans and chrysanthemums. Pages 42-49 in D. J. Schuster, ed. IFAS-Ind. Conf. Biol, and Control of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, Fl. 205pp. CLE. 1984. Distribution maps of pests, no. 450. Commonwealth Inst. Entomol. London. Conover, R. A., and Gerhold, N. R. 1981. Mixtures of copper and maneb or mancozeb for control of bacterial spot on tomato and their compatibility for control of fungus diseases. Proc. Fla. State. Hort. Soc. 94:154-56. Coquillet, D. W. 1898. On the habits of Oscinidae and Agromyzidae, reared at the United States Department of Agriculture. Bull. 10, n. ser., Div. Entomol. p70-79. d'Aguilar, C. M., and Martinez, M. 1979. Sur la presence en France de Liriomyza trifolii (Burgess). Bull. Entomol. Soc. France. 84:143-46. de Lima, C. P. F. 1979. Liriomyza trifolii (Diptera: Agromyzidae), an important new leafminer pest in Kenya. Kenya Entomol. Newsletter. 10:8. de Meijere, J. C. H. 1925. Die larven der Agromyzinen. Tidschr. Entomol. 68:195-293. Dunn, R. A. 1985. Tomato nematicides for 1985 1986 in Florida. Pages 109-112 in Florida Tomato Institute. Univ. Fla. Veg. Crops. Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2.

PAGE 120

Fagoonee, I., and Toory, V. 1983. Preliminary investigation of host selection mechanisms by the leafminer Liriomyza trif olii Insect Sci. Application. 4(4) : 337-341 Fagoonee, I., and Toory, V. 1984. Contribution to the study of the biology and ecology of the leaf-miner Liriomyza trif olii and its control by neem. Insect Sci. Application. 5(l):23-30. Fisher, R. C. 1971. Aspects of the phsiology of endoparasitic Hymenoptera. Biol. Rev. 46:243-278. Flanders, S. E. 1942. Environmental control of sex in hymennopteran insects. Ann. Entomol. Soc. Amer. 35:251-66. F.A.O. 1977. New records. Quarterly Newsletter, FAO Plant Protection Committee for the SE Asian and Pacific Region. 20(4):5-7. Frick, K. E. 1955. Nearctic species in the Liriomyza pusilla complex, no. 3 L_. alliovora new name for the Iowa onion miner (Diptera:Agromyzidae) J. Kansas Entomol. Soc. 28(3):88-92. Frick, K. E. 1959. Synopsis of the species of Agromyzid leafminers described from North America. Proc. U. S. Natl. Mus. 108:347-465. Frost, S. W. 1962. Liriomyza archboldi a new species (Dipt., Agromyzidae) Entomol. News. 73(l):51-53. Genung, W. G. 1981. Weed hosts of Liriomyza and parasite incidence in the celery agro-ecosystem at Belle Glade, Florida. Pages 61-69 in D.J. Schuster, ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla., Gainesville, FL. 205pp. Getz, W. M., and Guiterrez, A. P. 1982. A perspective on systems analysis in crop production and insect pest management. Annu. Rev. Entomol. 27:447-66. Getzin, L. W. 1960. Selective insecticides for vegetable leafminer control and parasite survival. J. Econ. Entomol. 53(5):872-75. Gilreath, J. P. 1985. Advances in weed management in tomatoes. Pages 41-51 in Florida Tomato Institute. Univ. Veg. Crops. Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2.

PAGE 121

Gordh, G., and Hendrickson, R., Jr. 1979. New species of Diglyphus a world list of the species, taxonomic notes and a key to the new world species of Diglyphus and Diaulinopsis (HymenopterarEulophidae) Proc. Entomol. Soc. Wash. 81(4):666-684. Hassell, M. P. 1978. The dynamics of arthropod-predator prey systems. Princeton Univ. Press. Princeton, NJ. 237pp. Hassell, M. P., and Waage, J. K. 1984. Host-parasitoid population interactions. Annu. Rev. Entomol. 29:89-114. Hawkins, W. 1985. The 1984-85 tomato season. Univ. Fla. Pages 100-105 in Florida Tomato Institute. Univ. Fla. Veg Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext Report VEC 85-2. Hendrickson, R. M. Jr., and Barth, S. E. 1978. Notes on the biology of Diglyphus intermedius (Hymenoptera : Eulophidae) a parasite of the alfalfa blotch leafminer, ARromyza f rontella (Diptera : Agromyzidae) Proc. Entomol. Soc. Wash 80(2):210-15. Hills, 0. A., and Taylor, E. A. 1951. Parasitization of Dipterous leafminers in cantaloupe and lettuce in the Salt River Valley, Arizona. J. Econ. Entomol. 44(5) : 759-62. Hochmuth, G. J. 1985. Fertilizer management: Back to basics. Pages 2-9 in_ Florida Tomato Institute. Univ. Fla. Veg. Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2. Holling, C. S. 1959. Some characteristics of simple types of predation and parasitism. Can. Entomol. 91:385-98. Huffaker, C. B., Smith, R. F., and Guiterrez, A. P. 1976. The need for systems analysis and its use in the US/IBP integrated pest management. Pages 209-16 iji R. L. Turamala, D. L. Haynes, and B. A. Croft, eds. Modeling for pest management: concepts, techniques, and applications. Mich. State Univ. East Lansing, MI. 247pp. Jacobson, M. 1981. Neem research in the USDA: Chemical, biological and cultural aspects. Pages 33-42 in Schmutterer, H., Ascher, K. R. S., and Rembold, H. eds. Proc. 1st Int. Neem Conf, Rottach-Egern. 1980.

PAGE 122

114 Johnson, J. A. 1985. Legal insecticides for control of insects on tomatoes. Pages 119-139 iji Florida Tomato Institute. Univ. Fla. Veg. Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2. Jones, J. P., and Jones, J. B. 1984. Target spot of tomato: Epidemiology and control. Proc. Fla. State Hort. Soc. 97:216-18. Jones, J. P., and Overman, A. J. 1985. Soil-borne diseases of tomato and their control. Pages 10-11 iji Florida Tomato Institute. Univ. Fla. Veg. Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops. Ext. Report VEC 85-2. Kamm, J. A. 1977. Seasonal reproduction and parasitism of a leafminer, Phytomyza nigra Meigen. Environ. Entomol. 6(4):592-95. Keularts, J. 1980. Effect of the vegetable leafminer Liriomyza sativae and the associated plant pathogens on yield and quality of the tomato, Lycopersicon esculentum cv. Walter. Ph.D dissertation. University of Florida. 154pp. King, P. E., and Richards, J. G. 1969. Oogenesis in Nasonia vitripennis (Walker) (Hymenoptera:Pteromalidae) Proc. R. Entomol. Soc. London Ser. A. 44:143-57. Knodel-Montz, J. J., and Poe, S. L. 1982. Ovipositor morphology of three economically important Liriomyza species (Diptera: Agromyzidae) Pages 186-195 iji S. L. Poe, ed. Proc. 3rd. Ann. Indus. Conf. on the Leafminer. The Center for Commercial Floriculture, Growers Division, Alexandria, VA. 216pp. Kochetova, N. I. 1978. Factors determining the sex ratio in some entomophagous Hymenoptera. Entomol. Rev. 60:1-5. Larew, H. G., Knodel-Montz, J. J., and Poe, S. L. 1986. Liriomyza trif olii (Burgess) (Diptera: Agromyzidae) overwinters outdoors in Maryland. Proc. Entomol. Soc. Washington. 88(l):p 189. Larew, H. G., Webb, R. E., and Warthon, Jr., J. D. 1984. Leafminer controlled on chrysanthemum by neem seed extract applied to potting. Pages 108-117 in_ S. L. Poe, ed. Proc. 4th. Ann. Indus. Conf. on the Leafminer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria, VA. 191pp.

PAGE 123

115 Leibee, G. L. 1981a. Development of Liriomyza trif olii on celery. Pages 35-42 jin D. J. Schuster, ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Leibee, G. L. 1981b. Insecticidal control of Liriomyza spp. on vegetables. Pages 216-20 in_ D. J. Schuster, ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Leibee, G. L. 198A. Influence of temperature on development and fecundity of Liriomyza trifolii (Burgess) (Diptera: Agromyzidae) on celery. Environ. Entomol. 13( 2) : 497-501 TM Leibee, G. L. 1985. Effects of cyromazine (Trigard ) on Liriomyza trifolii (Burgess). Pages 45-48 i_n an informal conf. on Liriomyza leafminers. USDA ARS Technical Information Bulletin, issued June 1985 by National Technical Information Services, Springfield, VA. 75pp. Lema, K. M. 1976. Cultural and insecticidal effect on Liriomyza sativae Blanchard (Diptera: Agromyzidae) and its parasites in celery. M. S. Thesis. University of Florida. 82pp. Lema, K. M., and Poe, S. L. 1979. Age specific mortality of Liriomyza sativae due to Chrysonotomyia f ormosa and parasitization by Opius dimidiatus and Chrysonotomyia f ormosa Environ. Entomol. 8(5):935-37. Lewis, W. J., and Vinson, S. B. 1971. Suitability of certain Heliothis (Lepidoptera: Noctuidae) as hosts for the parasite Cardiochiles nigriceps Ann. Entomol. Soc. Amer. 64:970-72. Lindquist, R. K., and Casey, M. L. 1985. Progress toward development of an IPM program for Liriomyza trifolii on greenhouse tomatoes in Ohio. page 28 in_ an informal conf. on Liriomyza leafminers. USDA ARS Technical Information Bulletin, issued June 1985 by National Technical Information Services, Springfield, VA. 75pp. Lyon, J. P. 1984. La minuese serpentine americaine: Liriomyza trifolii (Burgess). Revue Horticole. 245:13-15, 17. Marlowe, Jr., G. A., Overman, A. J., and Schuster, D. J. 1983. Growth and development studies of the tomato. Proc. Fla. State Hort. Soc. 96:103-107.

PAGE 124

116 Maynard, D. N. 1986. •Commercial vegetable varieties for Florida 1986. Circular 530-B. Florida Cooperative Extension Services, IFAS, University of Florida, Gainesville, FL. Minkenberg, 0. P. M. J., and Van Lenteren J. C. 1986. The leafminers Liriomyza bryoniae and L. trif olii (Diptera: Agromyzidae) their parasites and host plants: A review. Agric. Univ. Press. Wageningen, The Netherlands. (86)2:50. Monteith, L. G. 1956. Influence of host movement on selection by Drino bohemica Mesn. as determined by an olfactometer. Can. Entomol. 88:583-86. Morrison, G., and Strong, D. R., Jr. 1981. Spatial variation in egg density and the intensity of parasitism in a neotropical chrysomelid ( Cephaloleia consanguinea ) Ecol. Entomol. 6:55-61. Musgrave, C. A., Poe, S. L., and Weems, H. V. Jr. 1975. The vegetable leafminer, Liriomyza sativae in Florida. Fla. Dep. Agric. Consum. Serv, Div. Plant Indus. Entomol. Circ. 162. 4pp. Musgrave, C. A., Sraerage, G. H., Eshleman, W. D. and Poe, S. L. 1980. Chapter II: Simulations of the leaf miner-parasite complex in celery, iii Smerage, G. H., Musgrave, C. A., Poe, S. L., and Eshleman, W. D. Systems analysis of insect population dynamics. IPM-3. IFAS, Univ. Florida. Gainesville, Fl. Nedstam, B. 1981. Mineraflugor (Fam. Agromyzidae) i vaxthusLiriomyza bryoniae (Kaltenbach) L_. trif olii (Burgess) och Phytomyza syngenesiae (Hardy). Vaxtskyddnotiser 44(6) :135-37. Nyrop, J. P., and Simmons, G. A. 1986. Temporal and spatial activity patterns of an adult parasitoid, Glypta f umif eranae (Hymenoptera: Ichneumonidae) and their influence on parasitism. J. Environ. Entomol. 15(3) :481-87. Parrella, M. P., and Keil, C. B. 1984. Insect Pest Management: The lesson of Liriomyza Bull. Entomol. Soc. Amer. 30(2):22-25. Parrella, M. P., Christie, G. D., and Robb, K. L. 1983a. Compatibility of insect growth regulators and Chrysocharis parksi (Hymenoptera : Eulophidae) for the control of Liriomyza trifolii (Diptera: Agromyzidae) J. Econ. Entomol. 76(4):949-51.

PAGE 125

Parrella, M. P., Jones, V. P., and Christie, G. D. 1985. The interactions of parasites and leafminers on commercially grown chrysanthemum. Pages 13-18 ill An Informal Conference on Liriomyza leafminers. USDA ARS Technical Information Bulletin, issued June 1985 by National Technical Information Services, Springfield, VA. 75pp. Parrella, M. P., Robb, K. L., and Bethke, J. 1983b. The influence of selected host plants on the biology of Liriomyza trifolii Ann. Entomol. Soc. Amer. 76:112-15. Patel, K. J. 1981. Influence of temperature on the fecundity of Liriomyza sativae Blanchard (Diptera: Agromyzidae) and rate of development of L_. sativae and Diglyphus intermedius (Girault) (Hymenoptera : Eulophidae) M. S. Thesis. University of Florida. 38pp. > Patel, K. J. and Schuster, D. J. 1983. Influence of temperature on the rate of development of I), intermedius (Girault) (Hymenoptera: Eulophidae), a parasite of Liriomyza spp. leafminers (Diptera: Agromyzidae). Environ. Entomol. 12(3):885-887. Pena, J. E. 1983. Tomato pinworm, Keif eria lycopersicella ( Walsingham) : Population dynamics and assessment of plant in southern Florida. Ph.D Dissert. University of Florida. 265pp. Peterson, A. 1979. Larvae of insects. Part II. Edward Brothers Inc, Ann Arbor, MI. Petitt, F. L. 1984. Oviposition behavior of Opius dissitus and its use for management of Liriomyza sativae on greenhouse tomatoes. Pages 81-86 iji S. L. Poe, ed. Proc. 4th. Ann Indus. Conf. on the leafminer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria, VA. 191pp. Poe, S. L. 1985. Liriomyza leafminers: potential for management a summary. Pages 73-75 tn an informal conference on Liriomyza leafminers. USDA ARS Technical Information Bulletin, issued June 1985 by National Technical Information Services, Springfield, VA. 75pp. 75pp. Poe, S. L., Everett, P. H., Schuster, D. J., and Musgrave, C. A. 1978. Insecticidal effects on Liriomyza sativae larvae and their parasites on tomato. J. Georgia Entomol. Soc. 13(4):322-27.

PAGE 126

Poe, S. L., and Montz, J. K. 1981a. Preliminary results of a leafminer species survey. Pages 24-34 iji D.J. Schuster ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Poe, S. L., and Montz, J. K. 1981b. Leafminer parasite interactions. Pages 82-98 In D.J. Schuster, ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Poe, S. L., and Montz, J. J. 1982. Collection and identification of Liriomyza from cultivated areas. Pages 79-89 in S. L. Poe, ed. Proc. 3rd. Ann. Indus. Conf. on the Leafminer. SAF, The Center for commercial Floriculture, Growers Division, Alexandria, VA. 216pp. Pohronezny, K. L. 1985. Avoiding pest control entropy: A review of sound integrated pest management principles for Florida tomatoes. Pages 52-59 in_ Florida Tomato Institute. Univ. Fla. Veg. Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2. Pohronezny, K., Simone, G. W. and Waddill, V. 1984. A tomato scouting manual with color-illustrated dichotomous keys for insects and disease's in Florida tomato fields. IFAS. Univ. Fla. Coop. Ext. Serv. Gainesville, Spec. Public. SP-22. Pohronezny, K. L. and Waddill, V. 1978. Integrated pest management-development of an alternative approach to control of tomato pests in Florida. IFAS. Univ. Fla. Ext. Plant Path. Report. No. 22. 7pp. Powell, D. F. 1981. The eradication campaign against American serpentine leaf miner Liriomyza trif olii at Efford Experimental horticulture Station. Plant Pathol. 30:195-204. Price, J. F. 1981. Ecologia, biologia y control de Liriomyza trif olii (Burgess), minador de hojas de crisantemo en America. Mem. VIII Congr. Soc. Entomol. Colombia. pl3-18. Rembold, H. Sharma, G. K., Czoppelt, Ch., and Schmutterer, H. 1982. Azadirachtin : A potent insect growth regulator of plant origin. Z. Ang. Entomol. 93:12-17. Robb, K. L., and Parrella, M. P. 1984. Efficacy of selected new insecticides against Liriomyza trifolii (Burgess). Pages 157-162 in S. L. Poe, ed. Proc. 4th. Ann. Indus. Conf. on the leafminer. SAF. The center for commercial Floriculture, Growers Division, Alexandria, VA. 191pp.

PAGE 127

Ruesink, W. G. 1975. Analysis and modeling in pest management. Pages 353-76. _in_ R. L. Metcalf, and W. H. Luckman eds., Introduction to Insect Pest Management. John Wiley and Sons New York. Salt, G. 1940. Experimental studies in insect parasitism. VII The effects of different hosts on the parasite Trichogramma evanescens Westwood. Proc. R. Entomol. Soc. Lond. A. 15:81-95. SAS 1982. SAS Institute Inc. SAS user's guide: statistics, 1982 edition. Cary, NC. 584pp. Saraoucha, Y., and Cohen, Y. 1984. Differential sensitivity to mancozeb of metalaxyl-sensitive and metalaxylresistant isolates of Pseudoperonospora cubensis Phytopath. 74:1437-39. Schuster, D. J. 1981. Preface, in^ D.J. Schuster, ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Schuster, D. J. 1985. Seasonal abundance of Liriomyza leafminers and their parasitoids in fresh market tomatoes grown on the west coast of Florida. Pages 19-21 jji an informal conference on Liriomyza leafminers. USDA ARS Technical Information Bulletin, issued June 1985 by National Technical Information Services, Springfield, VA. 75pp. Schuster, D. J., and Beck, H. W. 1983. Visual rating systems for assessing Liriomyza spp. (Diptera: Agromyzidae) leafmining on tomato. J. Econ. Entomol. 76:1465-66. Schuster, D. J., and Everett, P. H. 1983. Response of Liriomyza trifolii (Diptera: Agromyzidae) to insecticides on tomato. J. Econ. Entomol. 76:1170-74. Schuster D. J., Montgomery, R. T., Gibbs, D. L., Marlowe, G. A., Jones, J. P., and Overman, A. J. 1980. The tomato pest management program in Manatee and Hillsborough Counties, 1978-80. Proc. Fla. State Hort. Soc. 93:235-239. Schuster, D. J., Musgrave, C. A., and Jones, J. P. 1979. Vegetable leafminer and parasite emergence from tomato foliage sprayed with oxamyl. J. Econ. Entomol. 72(2):208-210. Schuster, D. J., and Patel, K. J. 1985. Devlopment of Liriomyza trifolii (Burgess) larvae on tomato at constant temperatures. Fla. Entomol. 68( 1 ): 158-61

PAGE 128

120 Schuster, D. J., and Price, J. F. 1985. Impact of insecticides on lepidopterous larval control and leafminer parasite emergence on tomato. Fla. State Hort. Soc. (In Press). Scott, J. W. 1985. Recent findings in the IFAS tomato breeding program. Pages 38-40 In Florida Tomato Institute. Univ. Fla. Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2. Shoemaker, C. A. 1980. The role of systems analysis in integrated pest management. Pages 25-49 iji C. B. Huffaker, ed. New Technology of Pest Control. Wiley-Interscience New York. 500pp. Shorey, H. H., and Hall, I. M. 1963. Toxicity of chemical and microbial insecticides to pest and beneficial insects on poled tomatoes. J. Econ. Entomol. 56( 5) : 813-17 Smerage, G. H. 1980. Population modeling for insects. _in S. L. Poe and J. W. Standberg, eds. Opportunities for integrated pest management in celery production. IPM-2. IFAS, Univ. Fla. Gainesville, FL. 104pp. Smerage, G. H., Musgrave, C. A., Poe, S. L., and Eshleman, W. D. 1980. Systems analysis of insect population dynamics. IPM-3. IFAS, Univ. Fla. Gainesville, FL. 92pp. Smilowitz, Z., and Iwantsch, G. F. 1975. Relationship between the parasitoid Hyposeter exigua and the cabbage looper Trichoplusia ni Can. Entomol. 108:61-68. Smith, R. F.,and Broadman, J. M. 1986 Rates of feeding, oviposition, and survival of Liriomyza trif ollii (Burgess) (Diptera: Agromyzidae) on several weeds. Can. Entomol. 118(8) :753-59. Spencer, K. A. 1963. A synopsis of neotropical Agromyzidae. Trans. R. Entomol. Soc. Lond. 115:352-72. Spencer, K. A. 1964. The species-host relationship in the Agromyzidae (Diptera) as an aid to taxonomy. Proc. XII. Int. Congr. Entomol. London, p 101-02. Spencer, K. A. 1965. A clarification of the status of Liriomyza trif olii (Burgess) and some related species. Proc. Entomol. Soc. Wash. 67(l):32-40. Spencer, K. A. 1973. Agromyzidae (Diptera) of economic importance. Dr. W. Junk, The Hague.

PAGE 129

121 Spencer, K. A. 1981a. Morphological characteristics and brief history of Liriomyza Pages 12-23 In D.J. Schuster, ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Spencer, K. A. 1981b. A revisionary study of the leafmining flies (Agromyzidae) of California. Univ. Calif. Div. of Agric. Sci. Special Public. 3273. Spencer, K. A. 1985. East African Agromyzidae (Diptera): further descriptions, revisionary notes and new records. J. Nat. History. 19:969-1027. Spencer, K. A., and Stegmaier, C. E., Jr. 1973. Agromyzidae of Florida with a supplement of species from the Carribean. Arthropods of Florida and neighboring areas, vol. 7. Fla. Dep. Agric. Consum. Serv. Div. Plant Indus. Gainesville, FL. Stegmaier, C. E., Jr. 1968. A review of recent literature on the host plant range of of the genus Liriomyza Mik (Diptera: Agromyzidae) Fla. Entomol. 49(2):75-80. Stegmaier, C. E., Jr. 1981. The host plant ranges of Liriomyza sativae and Liriomyza trif olii and notes on their parasites. Pages 56-60 iii D.J. Schuster, ed. Proc. IFAS-Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Stimac, J. L. 1982. History and relevance of behavioral ecology in models of insect population dynamics. Fla. Entomol. 65(1):9-16. Sugimoto, T., and Ishii, M. 1979. Mortality of larvae of a ranunculus leaf mining fly, Phytomyza ranunculi (Diptera: Agromyzidae), due to parasitization and host feeding by its Eulophid parasite Chrysocharis pentheus (Hymenoptera:Eulophidae) Appl. Entomol. Zool. 14(4):410-418. Trumble, J. T. 1985. Integrated pest management of Liriomyza trif olii : Influence of avermectin, cyromazine, and methomyl on leafminer ecology in celery. Agriculture, Ecosysytems, and Environment. 12:181-88 Trumble. J. T., and Nakakihara, H. 1983. Occurence, parasitization, and sampling of Liriomyza species infesting celery in California. Environ. Entomol. 12:810-814.

PAGE 130

122 Tuovinen, T., and Aapro, H. 1981. Liriomyza trifolii (Diptera: Agromyzidae) introduced on chrysanthemum in Finland. Notulae Entomol. 61:173-74. Van Sickle, J. J., and Belibasis, E. 1985. Update on Florida West Mexico competition in the fresh market tomato industry. Pages 79-97 Jji Florida Tomato Institute. Univ. Fla. Veg. Crops Dept. (D. N. Maynard, Prog. Coord.) Veg. Crops Ext. Report VEC 85-2. Vercambre, B. 1980. Etudes realises a la Reunion sur la mouche maraichere: Liriomyza trifolii (Burgess). Rev. Agric. Sure. lie Maurice. 59:147-57. Vinson, S. B. 1975. Biochemical coevolution between parasitoids and their hosts. Pages 14-18 i_n P. W. Price, ed. Evolutionary strategies of parasitic insects and mites. McgrawHill, New York. 224pp. Vittum, P. J. 1982. Leafrainers in Massachusetts greenhouse chrysanthemums. Pages 168-169 in_ S. L. Poe, ed. Proc. 3rd. Ann. Indus. Conf. on the Leafminer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria, Va. 216pp. Waage, J. K. 1982. Sib-mating and sex ratio strategies in scelionid wasps. Ecol. Entomol. 7:103-12. Waage, J. K. 1983. Aggregation in field parasitoid populations: foraging time allocation by a population of Diadegma (Hymenoptera: Ichneumonidae) Ecol. Entomol. 9. Waddill, V. H. 1978. Contact toxicity of four synthetic pyrethroids and methomyl to some adult insect parasites. Fla. Entomol. 61:27-30. Waddill, V. H. 1981. Effects of insecticides on non-target organisms. Pages 186-89 in D.J. Schuster, ed. Proc. IFAS Ind. Conf. Biol, and Cont. of Liriomyza leafminers. IFAS, Univ. Fla. Gainesville, FL. 205pp. Webb, R. E., Larew, H. G., Wieber, A. M., Ford, P. W. and Warthen, Jr., J. D. 1984. Systemic activity of neem seed extract and purified azadirachtin against Liriomyza leafrainers. Pages 118-27 in S. L. Poe, ed. Proc. 4th. Ann. Indus. Conf. on the leafminer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria, VA. 191pp. Webster, F. M., and Parks, T. H. 1913. The serpentine leafminer. J. Agric. Sci. 1:59-87.

PAGE 131

123 Wene, G. P. 1955. Effects of some organic insecticides on the population levels of the serpentine leafrainer and its parasites. J. Econ. Entomol. 48:596-97. Zehnder, G. W. and Trumble, J. T. 1982. Monitoring leafminer activity in pole tomatoes. Pages 153-161 iji S. L. Poe, ed. Proc. 3rd. Ann. Indus. Conf. on the Leafminer. SAF, The Center for Commercial Floriculture, Growers Division, Alexandria, VA. 216pp. Zehnder, G. W., Trumble, J. T, and White, W. R. 1983. Discrimination of Liriomyza species (Diptera: Agromyzidae) using electrophoresis and scanning electron microscopy. Proc. Entomol. Soc. Wash. 85. Zehnder, G. W. and Trumble, J. T. 1984a. Host selection of Liriomyza spp. (Diptera:Agrorayzidae) J. Econ. Entomol. 13:492-96. Zehnder, G. W., and Trumble, J. T. 1984b. Spatial and diel activity of Liriomyza sp. ( Diptera : Agromyzidae ) in fresh market tomatoes. Environ. Entomol. 13(5) : 1411-1416. Zehnder, G. W. and Trumble, J. T. 1985. Impact of currently registered insecticides on the Liriomyza / parasite complex in celery, 1984. Pages 21-27 iii an informal conference on Liriomyza leafminers. USDA ARS Technical Information Bulletin, June 1985 by National Technical Information Services, Springfield, VA. 75pp. Zoebisch, T. G., 1984. Oviposition and development of Liriomyza trif olii (Burgess) (Diptera:Agromyzidae) on foliage of tomato and selected weeds. M.S. thesis. Univ. Fla. 48pp. Zoebisch, T. G., and Schuster, D. J. 1984. Liriomyza trif olii : Oviposition and development in foliage of tomato tomato and common weed hosts. Fla. Entomol. 67(2) : 250-54.

PAGE 132

BIOGRAPHICAL SKETCH Kirtikumar Jashbhai Patel was born in Mbale, Uganda, in 1951 and attended Mbale Senior Secondary School during the years 1965 to 1970. In 1972, he obtained a scholarship to study at Balliol College, Oxford University, England. He graduated in 1974 from Oxford with a B.A. Honors in agriculture and forestry sciences. From 1974 to 1978 he pursued various activities including a year spent in India working on a farm run by a charity to help the local population. He enrolled in the M.S. program in the Entomology and Nematology Department of the University of Florida in 1978 under the supervision of Dr. Schuster. After completing the M.S. program, he pursued the Ph.D. degree under the continued supervision of Dr. Schuster. It is his ambition to follow a career in pest management at a research institution. 124

PAGE 133

I certify that I have read this study and that in ray 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. David J. SchMister, Chairman Professor of Entomology and Nematology I certify that I have read this study and that in ray 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^~XXgA X rx^ ^4 .y^ v> nrv Stratton H. Kerr Professor of Entomology and Nematology 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. Glen H. Smerage / Professor of Agricultural Engineering This dissertation was presented to the Graduate Faculty of the College of Agriculture and to the Graduate School, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy May 1987 Dean, Graduate School


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EWSTHN0VG_2W5B00 INGEST_TIME 2015-01-21T18:37:24Z PACKAGE AA00026636_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES