Citation
Light color attraction and dietary sugar composition for several mosquito (Diptera: Culicidae) species found in north central Florida

Material Information

Title:
Light color attraction and dietary sugar composition for several mosquito (Diptera: Culicidae) species found in north central Florida
Creator:
Burkett, Douglas Arthur, 1964-
Publication Date:
Language:
English
Physical Description:
xvi, 141 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Amino acids ( jstor )
Crops ( jstor )
Female animals ( jstor )
Honeydew ( jstor )
Insects ( jstor )
Nectar ( jstor )
Species ( jstor )
Sugar feeding ( jstor )
Sugars ( jstor )
Wavelengths ( jstor )
Dissertations, Academic -- Entomology and Nematology -- UF ( lcsh )
Entomology and Nematology thesis, Ph.D ( lcsh )
Mosquitoes -- Behavior -- Florida ( lcsh )
Mosquitoes -- Feeding and feeds ( lcsh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1998.
Bibliography:
Includes bibliographical references (leaves 125-139).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Douglas Arthur Burkett.

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:
029532218 ( ALEPH )
40058826 ( OCLC )

Downloads

This item has the following downloads:


Full Text










LIGHT COLOR ATTRACTION AND DIETARY SUGAR
COMPOSITION FOR SEVERAL MOSQUITO (DIPTERA: CULICIDAE) SPECIES
FOUND IN NORTH CENTRAL FLORIDA










By

DOUGLAS ARTHUR BURKETT















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 1998


































Dedicated to
Laura, Alexis, Sara, Betty and Bob














ACKNOWLEDGMENTS


Many and most sincere thanks to the Air Force Institute of Technology whose Civilian Institution's program sent me back to school for what I consider the best educational experience ever. Next, I would like to thank the group of scientists and staff at the U.S. Department of Agriculture, Animal Research Service and the Department of Entomology and Nematology for supplying me with an office, lab space, vehicles, and administrative support. They treated me as one of their own.

I wish to express my sincere gratitude to the members of my committee. To Dr. Dan Kline who always welcomed me into his office, and was never too busy to help out, thanks for generously sharing your resources, space and equipment. These same sentiments are for Dr. Jerry Butler as well, who served as a great mentor and friend. This dissertation would not have been possible without the support and labor provided by him and his medical entomology staff. A special thanks to Dr. Gene Gerberg from whose extraordinarily long entomological career, experiences, and library I have greatly benefited. Thanks to Dr. Sandy Allan for finding a way to be on my committee. Her expertise in insect vision and excellent editorial skills have been very helpful.

A number of other scientists and staff members deserve mention. Dr. Dave Carlson, I learned a hell of lot from you. Thank you for your support, equipment, enthusiasm, exceptional writing skills and your friendship. I also wish to thank the other iii








USDA chemists, Dr. Bob VanderMeer and Dr. Uli Bernier, for sharing their equipment and expertise. Thank you Dr. John Reinert for your help with mosquito identifications and relevant reprints and to Dr. Pete Perkins for his sand fly identifications. Thanks Dr. Don Barnard for allowing me to raid your library and giving me free reign in the CMAVE. Charles Schmidt, of Environmental Engineering, I greatly appreciate your warm welcome into your lab and admire your love of analytical chemistry. You and your staff treated me as one of your own and donated your equipment and supplies. Henry McKeithen, thank you for your support, computer help, and braving the swamps with me. Thanks to Genie White for sharing SAS and wild Culex with me. Eric Daniels, you have been a tremendous associate. Your computer help, support and friendship will never be forgotten. Thank you Dr. Dan Haile, who was initially one of my committee members but transferred to Panama before my degree completion. Our aerial spray adventures were great. Thanks to Drs. Don Hall and Grover Smart for helping me get through my qualifying exams. I also wish to acknowledge Erin Rountree and Margaret Martin for their administrative support and Johnny Jackson for the large supply of mosquitoes from the insectary. From the Department of Entomology and Nematolgy, I wish to thank Dr. Butler's crew including Diana Simon, Tim Robson, Haze Brown, and Tamara Landau. Thanks to Drs. Frank Slansky and Harvey Cromroy, for your excellent reviews of my manuscripts. Debbie Hall, you were amazing in the front office thanks. As my sole military associates while attending graduate school, I acknowledge the DoD/USDA Liaison officers, Lt. Col. Terry Carpenter and Captain Herb Bolton for their encouragement along the way.

iv








I would not have gone on to complete this degree if not for the urging of Dr. Terry Biery, my supervisor for 5 years at the Aerial Spray mission. Thanks Terry. I also wish to acknowledge Dr. Alberto Broce, my Master's program advisor at Kansas State University, who further enhanced my interest in entomology and pursuing additional graduate education.

Lastly, a huge thanks to my parents who provided the initial education funding, guidance and support. I owe the utmost gratitude to my wife, Laura, whose love, enthusiasm and understanding have been instrumental in completing all these years of graduate school and Air Force directed moves.






























v














TABLE OF CONTENTS

ACKNOWLEDGMENTS ........................................... 111

LIST OF TABLES ............................................ ix

LIST OF FIGURES .......................................... x

ABSTRACT ..................................................... xv

1. MOSQUITO SUGAR FEEDING AND VISUAL ATTRACTANTS REVIEW ..... 1
Introduction To Sugar Feeding ................................... 1
Carbohydrate and Amino Acid Sources ........................... 3
The Need for Dietary Sugars .................................. 5
Floral Nectaries ............................. ........... 7
Extrafloral Nectaries ................... ............. 11
Honeydew ........................................... 13
Other Sugar Sources ...................................... 16
Diel Sugar Feeding Activities ................................. 17
Morphology and Physiology of Mosquito Alimentary Tract ......... 18 Crop Enzymatic Activity ............................. 21
Introduction To Mosquito Visual Attractants Review .................... 22
Reflected Light ............................................. 23
Transmitted Light ...................................... 24

2. ANALYSIS OF SUGAR MEAL COMPOSITION OF WILD MOSQUITOES BY
GAS CHROMATOGRAPHY .......................................26
Introduction ................................................. 26
Methods and Materials ........................................... 28
Specimen Preparation ..................................... 28
Sample Preparation for GC Analysis ............................ 29
Crop Sucrose and Melezitose Hydrolysis ........................ 30
Results and Discussion ............................................ 31
Sucrose and Melezitose Fed Mosquitoes ........................ 32

3. SUGAR MEAL COMPOSITION OF FIVE NORTH CENTRAL FLORIDA
MOSQUITO SPECIES AS DETERMINED BY GAS CHROMATOGRAPHY 41 Introduction .....................................................41

vi








Methods and Materials ......................................... 43
Specimen Preparation ................................... 43
Sample Preparation for GC Analysis ............................ 44
Standards ...........................................45
Results and Discussion ...................................... 45
Ecological Significance of Honeydew in Mosquito Diet ............. 46
Sugar Feeding Field Observations ............................48

4. EVALUATION OF COLORED LIGHT EMITTING DIODES AS ATTRACTANTS
FOR WOODLAND MOSQUITOES AND OTHER DIPTERA IN NORTH
CENTRAL FLORIDA ............................................ 56
Introduction .....................................................56
Methods and Materials ................... ..................... 57
Six Colors With and Without Carbon Dioxide (Trials 1 and 2) ....... 57 Eight Colors With Carbon Dioxide (Trial 3) ...................... 59
Results ........................................................ 59
Six Colors With and Without Carbon Dioxide (Trials 1 and 2) ....... 59 Eight Colors With Carbon Dioxide (Trial 3) ...................... 60
Discussion .............. ............................... 61

5. EVALUATION OF BLUE AND GREEN LIGHT EMITTING DIODES AS
ATTRACTANTS FOR WOODLAND MOSQUITOES AND OTHER
DIPTERA IN NORTH CENTRAL FLORIDA ........................76
Introduction ............. ................................. 76
Methods and Materials ............ ................................ 77
Results .................. ..................................... 79
D iscussion ................... ...................................81

6. LABORATORY EVALUATION OF COLORED LIGHT AS AN ATTRACTANT
FOR FEMALE AEDES AEGYPTI, AEDES ALBOPICTUS, ANOPHELES
QUADRIMACULATUS AND CULEX NIGRIPALPUS ................. 87
Introduction ...................................................87
Methods and Materials ........................................... 87
Visualometer and Data Collection .............................. 89
Artificial Host .............................. ....... .... 90
Light Source and Filters .................................... 91
M osquito Species .........................................91
Statistical Analysis ..........................................92
R esults ........... ... ...................................... 92
Aedes aegypti ............................................. 93
Aedes albopictus .................... ..... ............... 93
Anopheles quadrimaculatus ................................... 94
Culex nigripalpus ...................... .. ............... .... 94

vii








Discussion ................................................ 95

7. MOSQUITO HOST/RESOURCE FINDING: THE IMPORTANCE OF SUGAR
FEEDING AND ATTRACTION TO ARTIFICIAL LIGHT .............. 110
Introduction ................ .................... ........... 110
Sugar Feeding .................. ............................. 113
Attraction to Artificial Light ................................... 119
Summary ...................................... .........123

REFERENCES CITED ................................................ 125

BIOGRAPHICAL SKETCH .......................................... 140







































viii













LIST OF TABLES



Table Pagge

3-1. Summary of percent wild mosquitoes containing identifiable sugars using
gas chromatography. ................................................... 50

4-1. CDC Trap counts using colored LEDs or incandescent light with 200
ml/min CO2 (Means + SEM). Means within each row having the same
letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple
Range Test). n = 6 nights ......................................... 65

4-2. CDC Trap counts using colored LEDs or incandescent light only (Means
SEM). Means within each row having the same letter are not significantly
different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). n = 6 nights. ... 66

4-3. CDC Trap counts using colored LEDs or incandescent light with 200
ml/min CO2 (Means SEM). Means within each row having the same
letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple
Range Test). n = 8 nights ......................................... 67

5-1. CDC Trap counts using colored LEDs or incandescent light in different
orientations with 200 ml/min CO2 (Means SEM). Means within each
row having the same letter not significantly different
(Ryan-Einot-Gabriel-Welsh Multiple Range Test). n = 6 nights ............ 83













ix














LIST OF FIGURES


Figure Page

2-1. Chromatogram of combined standards (ca. 0.1%) for common sugars
associated with plants. Multiple peaks of the same sugar are anomeric
form s............. ....................................... 35

2-2. Representative chromatograms of crude crop extracts from two wild
individuals of Culiseta melanura. Multiple peaks of the same sugar are
anomeric forms........................................... ..36

2-3. Representative chromatograms of crude crop extracts from two wild
individuals of Anopheles quadrimaculatus. Multiple peaks of the same
sugar are anomeric forms. ......................................... 37

2-4. Timed feeding trials for male and female Aedes albopictus showing the
hydrolysis of melezitose and sucrose occurring in the crop (n = 2 to 5) ...... 38

2-5. Representative chromatogram showing hydrolysis of sucrose into fructose
and glucose 2 hours following ingestion in Aedes albopictus. Multiple
peaks of the same sugar are anomeric forms ........................... 39

2-6. Representative chromatogram showing hydrolysis of melezitose 2 hours
following ingestion in Aedes albopictus ........................... .40

3-1. Representation of crop sugar distributions and proportions of wild
Anopheles quadrimaculatus (n = 33). The presence of the trisaccharides,
erlose and melezitose, is evidence of honeydew feeding. Maltose and
turanose have identical retention times ........................... .51

3-2. Representation of crop sugar distributions and proportions of wild
Coquilletidia perturbans (n = 20). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and
turanose have identical retention times ................................ 52



x








Figure Page

3-3. Representation of crop sugar distributions and proportions of wild
Culiseta melanura (n = 99). The presence of the trisaccharides, erlose and
melezitose, is evidence of honeydew feeding. Maltose and turanose have
identical retention times .. ........ ............................. 53

3-4. Representation of crop sugar distributions and proportions of wild d)
Culex nigripalpus (n = 26). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and turanose have
identical retention times ....... ............................... 54

3-5. Representation of crop sugar distributions and proportions of wild
Psorophoraferox (n = 14). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and turanose have
identical retention times. .................................................. 55

4-1. Modified CDC-type trap equipped with LED (shown with arrow). .......... 69

4-2. Relative percent composition of mosquito species captured in C02 baited
CDC-type traps using colored LEDs or incandescent light only. Means
within each species group having the same letter are not significantly
different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). a = 0.05, n =
6 nights. ...................... ............................. 70

4-3. Relative percent composition of mosquito species captured in C02 baited
CDC-type traps using colored LEDs or incandescent light only. Means
within each species group having the same letter are not significantly
different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). alpha = 0.05,
n = 6 nights. ................................................... 71

4-4. Relative percent composition of mosquitoes captured in CDC-type traps
using colored LEDs or incandescent light only. Means within each
species group having the same letter are not significantly different (RyanEinot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 6 nights. ...... 72

4-5. Relative percent composition of mosquitoes captured in CDC-type traps
using colored LEDs or incandescent light only. Means within each
species group having the same letter are not significantly different (RyanEinot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 6 nights. ...... 73




xi








Figure Page

4-6. Relative percent composition of mosquitoes captured in C02 baited CDCtype traps using colored LEDs or incandescent light only. Means within
each species group having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 8
nights........ ........................................... 74

4-7. Relative percent composition of mosquitoes captured in C02 baited CDCtype traps using colored LEDs or incandescent light only. Means within
each species group having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 8
nights..........................................................75

5-1. Wiring assemble representation for LED orientations: a) Top view of LED
assembly oriented 360 out and b) Top view of LED assemblly oriented "up." For green LEDs: 3 V with separate battery pack or if wired to trap
electronics use x ohm resistor. For blue LEDs: 4.5 V with separate battery
pack or if wired to trap electronics use y ohm resistor. .................... 84

5-2. Relative percent composition of common mosquito species captured in
C02 baited CDC-type traps using different orientations of green or blue LEDs or incandescent light. Means within each species group having the
same letter are not significantly different (Ryan-Einot-Gabriel-Welsh
Multiple Range Test). alpha = 0.05, n = 6 nights ........................ 85

5-3. Relative percent composition of common mosquito species captured in
C02 baited CDC-type traps using different orientations of green or blue LEDs or incandescent light. Means within each species group having the
same letter are not significantly different (Ryan-Einot-Gabriel-Welsh
Multiple Range Test). alpha = 0.05, n = 6 nights ........................ 86

6-1. Visualometer (a) Top view (b) Side view. (A) 430 x 5 cm dia. transparent
plexiglass lid (B) Artificial host, (35 mm film canister lid) (C) 30 cm high
plastic side piece with holes for plastic tubing (D) Aluminum base with holes for fiber optic cables (E) Support leg (F) Plexiglass filter support
(G) 10 cm diameter Tygon tubing for incoming air/C02 (H) Tubing for exhaust (I) Assembly screw (J and K) Probe inserted into top of artificial
host and bottom sensor fitted under artificial host (feeding mosquitoes complete circuit logged by computer), one sensor per artificial host (L) Interference (bandbass) filter (M) Neutral density filter (N) Fiber optic
cable (attached to light source). Drawing not to scale. ................... 101


xii








Figure Page

6-2. Duration of feeding (seconds) during a 4 hour exposure (means SEM)
for Aedes aegypti on artificial hosts illuminated with different wavelengths
of light. Means within each species group with the same letter are not
significantly different (alpha=0. 10, Duncan's Multiple Range Test). ....... 102

6-3. Duration of feeding (seconds) during a 4 hour exposure (means + SEM)
for Aedes albopictus on artificial hosts illuminated with different
wavelengths of light. Means within each species group with the same
letter are not significantly different (alpha=0.10, Duncan's Multiple Range
Test). ................... ............................... 103

6-4. Duration of feeding (seconds) during a 4 hour exposure (means + SEM)
for Anopheles quadrimaculatus on artificial hosts illuminated with
different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple
Range Test) ................. ................................. 104

6-5. Duration of feeding (seconds) during a 4 hour exposure (means SEM)
for Culex nigripalpus on artificial hosts illuminated with different
wavelengths of light. Means within each species group with the same
letter are not significantly different (alpha=0. 10, Duncan's Multiple Range
Test). .......................... ....................... 105

6-6. Numbers of fecal specks (means + SEM) after an 16 hour exposure in a 5
cm circle around each artificial host/wavelength combination for Aedes
aegypti on artificial hosts illuminated with different wavelengths of light.
Means within each species group with the same letter are not significantly
different (alpha=0.10, Duncan's Multiple Range Test). .................. 106

6-7. Numbers of fecal specks (means + SEM) after an 16 hour exposure in a 5
cm circle around each artificial host/wavelength combination for Aedes albopictus on artificial hosts illuminated with different wavelengths of
light. Means within each species group with the same letter are not
significantly different (alpha=0.10, Duncan's Multiple Range Test). ....... 107

6-8. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5
cm circle around each artificial host/wavelength combination for
Anopheles quadrimaculatus on artificial hosts illuminated with different
wavelengths of light. Means within each species group with the same
letter are not significantly different (alpha=0.10, Duncan's Multiple Range
Test) ........................... .............. 108

Xiii












Figure Page

6-9. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5
cm circle around each artificial host/wavelength combination for Culex nigripalpus on artificial hosts illuminated with different wavelengths of
light. Means within each species group with the same letter are not
significantly different (alpha=0.10, Duncan's Multiple Range Test). ....... 109






































xiv














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


LIGHT COLOR ATTRACTION AND DIETARY SUGAR
COMPOSITION FOR SEVERAL MOSQUITO (DIPTERA: CULICIDAE) SPECIES FOUND IN NORTH CENTRAL FLORIDA By

Douglas Arthur Burkett

August 1998


Chair: Daniel L. Kline
Cochair: Jerry F. Butler
Major Department: Entomology and Nematology

The behavioral response of mosquitoes to different wavelengths of light and an evaluation of sugar meals was studied for several species of mosquito found in north central Florida. Gas chromatography (GC) was used to analyze dietary sugars of Anopheles quadrimaculatus, Coquillettidia perturbans, Culex nigripalpus, Culiseta melanura and Psorophoraferox. GC was also used to determined whether carbohydrases are present in the mosquito diverticula. A wide range of sugars was found in wild mosquitoes including fructose, glucose, sucrose, maltose, turanose, melezitose, raffinose, erlose, arabinose, rhamnose, and several unknowns. Laboratory time course studies with Aedes albopictus showed rapid hydrolysis of sucrose within 2 hours of ingestion.


xv








Conversely, melezitose remained relatively unchanged after 8 hours. The frequency of wild mosquitoes containing sugars ranged from 10-11% in An. quadrimaculatus and Ps. ferox to 48% in Cq. perturbans. All species tested contained honeydew sugars including An. quadrimaculatus (55%), Cs. melanura (33%), Cx. nigripalpus (15%), Cq. perturbans (10%) and Ps. ferox (7%). Light trap capture numbers for woodland mosquitoes were evaluated using light emitting diodes (LEDs) of different colors. Analysis of data by species showed significant differences in color attraction (green and blue best) for Ae. atlanticus, Ae. dupreei, Ae. infirmatus, An. crucians, Cs. melanura, Cx. nigripalpus, Ps. columbiae and Uranotaenia sapphirina. Attraction was also evaluated using different orientations (light reflected of off lid, or 3600 radius) of green/blue LEDs and found significant capture number differences for only An. crucians, Cs. melanura, Cx. (Melanoconion spp.) and Ur. sapphirina. In the lab, a ten port visualometer was constructed to evaluate the feeding response of 4 mosquito species over various wavelengths of light. Color preferences were based on feeding durations obtained electronically over 4 hour intervals and numbers of fecal specks after a 16 hour exposure. Feeding duration and fecal speck data were collected for female Ae. aegypti, Ae. albopictus, An. quadrimaculatus and Cx. nigripalpus in the visualometer on an artificial host illuminated from below with light of equal intensities in 50 nm increments (700-350 nm). Significant color preferences based on feeding durations were detected for Ae. albopictus, An. quadrimaculatus and Cx. nigripalpus. Fecal speck numbers significantly correlated with feeding durations for An. quadrimaculatus and Cx. nigripalpus.



xvi














CHAPTER 1
MOSQUITO SUGAR FEEDING AND VISUAL ATTRACTANTS REVIEW Introduction To Sugar Feeding


An increased understanding of mosquito biology, improved surveillance and effective control practices are the main driving forces that has made Florida and other coastal states habitable. Blood loss, annoyance and their ability to transmit a variety of serious diseases have given mosquitoes dubious notoriety. Research on the sugar sources of mosquitoes can have great potential returns to biologists and mosquito control personnel. Determination of crop carbohydrate composition would allow researchers to make inferences about the identification of plants and plant-derived products being used to provide metabolic demands of wild mosquitoes. These inferences, however, are only possible if common plant-derived oligosaccharides remain unhydrolyzed and identifiable in the crop for a significant amount of time. A better understanding of these facets of mosquito ecology can be applied directly to the development of new attractants and improving control methods.

It is currently unknown as to whether natural sugars are an unlimited and

universally available resource, or if suitable sugars are scarce and their lack can adversely affect mosquito survival and reproduction. In general, mosquitoes obtain sugars from a wide range of sources (Gadawski and Smith 1992, Grimstad and DeFoliart 1974,

1








2

Magnarelli 1977, 1979, 1983). Furthermore, not much is known about species-specific preferences for natural sugar hosts. Are mosquitoes generalists and opportunistic, obtaining sugar from sources that are readily available or do they use a narrow set of discrimination criteria to choose their natural sugar sources?

Plant taxa differ in the proportion and to a lesser extent, the composition of sugars present in nectar. If sugar meals remain unchanged upon ingestion and storage in the crop, it may be possible to identify specific plants or at least groups of plants with similar properties based on the presence or absence and relative concentrations of sugars present in the mosquito crop. It is currently unknown how many mosquito species contain salivary or crop enzymes which hydrolyze sucrose and other common plant oligosaccharides into simpler molecules.

As a precursor to determining sugar sources utilized by various wild mosquito

species, one must determine if, how much, and how fast the plant-derived mono- di- and trisaccharides are broken down in the crop. If these ingested sugars remain unchanged in the crop for any length of time, inferences could be made as to their sugar source utilization of wild mosquitoes. Ultimately, we could determine if certain mosquito species preferentially utilize various plant families with certain nectar characteristics, or if they utilize sugary exudates from Homoptera (e.g. honeydew) or some other sugar source. For a given mosquito species, what proportion of the population feeds on nectars or other plant juices and what proportion feeds on honeydew?








3

Carbohydrate and Amino Acid Sources


Much of the evidence for mosquito nectar feeding comes from direct field

observations rather than through indirect chemical qualitative analysis of crop contents. West and Jenkins (1951), Sandholm and Price (1962), Gadawski and Smith (1992), Breeland and Pickard (1961, 1967), and Magnarelli (1977) have observed some mosquito species feeding during the day. Most direct observations, however, have been made at night, dawn or dusk (Grimstad and DeFoliart 1974, Magnarelli 1983, Andersson and Jaenson 1987, Bowen 1992, Vargo and Foster 1984, Yee et al. 1992). Interestingly, not all common mosquito species have been observed feeding on sugar in the field. Bidlingmayer and Hem (1973) found evidence of fructose in the crops of Culiseta melanura (36%), Culex nigripalpus (17%), Psorophora ferox (20%) and Anopheles quadrimaculatus (15%). Likewise, Magnarelli (1978) obtained similar results for Cq. perturbans (57%). There have been no published accounts of field sugar feeding for Aedes albopictus, An. quadrimaculatus, or Cs. melanura and only a few for Cx. nigripalpus (see Nayar 1982). Regardless of species, most direct field observations of sugar feeding have occurred at ground level nectar sources on flowering herbs and low growing shrubs. These observations do not incorporate sugar feeding in trees or at the potentially numerous honeydew, extrafloral nectary or other plant exudates scattered up in the canopy and throughout the environment.

As indirect evidence for "nectar" feeding, dozens of mosquito species have been chemically analyzed for the presence or absence of fructose. Fructose is a ubiquitous








4

monosaccharide present in almost all plant sugars. Many authors (Andersson and Jaenson 1987, Edman et al. 1992, Van Handel et al. 1972, 1994, Bidlingmayer and Hem 1973, Magnarelli 1978, 1979, 1980, 1983, Reisen et al. 1986, Nasci and Edman 1984, Smith and Kurtz 1994) have used the cold anthrone test (Van Handel 1972) to qualitatively establish the presence or absence of fructose (or oligosaccharides containing fructose or other reducing sugars). According to Van Handel (1967), the cold anthrone reagent reacts with fructose, inulin, sucrose, melezitose, and raffinose. These sugars all contain the fructose moiety and give a positive or negative response based on a resulting color reaction. Unfortunately, the cold anthrone test does not provide information about the origin of crop sugars or the exact carbohydrate composition. Many other common mono-, di- and trisaccharides are present in potential natural sugar meals, but few attempts have been made to categorize the individual sugars and relate the meals back to the original source.

Although usually used as evidence for "nectar feeding," the cold anthrone test will also react positively for insects having fed on honeydew or other non-floral sugar sources. Few attempts have been made to determine the exact crop sugar concentration and composition for various Dipteran crops using modem chromatographic techniques. Thin Layer Chromatography (TLC) has been used by Magnarelli and Anderson (1977), Magnarelli (1980), and Van Handel (1984). High performance liquid chromatography (HPLC) was successfully used by MacVicker et al.(1990), to examine the contents of crops of five wild Italian sand fly species. Hoppe (1983), used HPTLC for determining the sugars found in horse flies. In spite of its common use in other systems, Moore et al.








5

(1987), Chang et al. (1977), and Alexander (1988), were among the first (and few) to use modern and sensitive gas liquid chromatography (GC o r GLC) for investigating dipteran diets.


The Need for Dietary Sugars


With few exceptions, mosquitoes generally require dietary sugars for survival, longevity, host finding and reproductive success. In terms of survival, female Ae. aegypti, Cx. quinquefasciatus and Cx. tritaeniorhynchus lived much longer (Briegel and Kaiser 1973, Harada et al. 1976) and laid more eggs (Nayar and Sauerman 1975a) in the laboratory when provided with both sugar and blood meals than when given blood alone. When fed individual sugars, Galun and Fraenkel (1957) reported that Ae. aegypti survived the longest (at least 25 days) on glucose, fructose, sucrose, maltose, trehalose, melibiose, raffinose or melezitose. With the exception of melibiose, similar results were found for Cs. inornata (Salama 1967). Other sugars reduce the longevity or have no effect on survival or fecundity (Nayar and Sauerman 1971a). The sugars enhancing survival are commonly found directly or indirectly with plants. Nayar and Sauerman (1975a) reported duration of mosquito survival to be directly related to the rate of disappearance of glucose which followed an exponential law of decline. Furthermore, the disappearance rate differed between mosquito species. Ae. aegypti and An. quadrimaculatus, for example, used glucose faster than Ae. sollicitans and Ae. taeniorhynchus under similar conditions.








6

Some species or populations of some species may require minimal or no sugar to survive and reproduce. An Ae. aegypti population from Thailand was found to rarely, if ever take sugar meals, apparently obtaining all their sugar requirements from multiple blood meals (Day et al. 1994, Edman et al. 1992). Interestingly, this Ae. aegypti population produced as many eggs as those that took sugar meals. In another similar case, a Puerto Rican Ae. aegypti population contained only 5% of the mosquitoes positive for fructose (Van Handel et al. 1994). Whereas, in a Florida population, 50-75% of the sampled mosquitoes were positive for fructose. Some authors find sugar vital for stimulating ovarian development. For example, Klowden (1986), found that laboratory reared Ae. aegypti did not develop a batch of eggs without obtaining a sugar meal before a blood meal. Much remains to be learned about sugar feeding. A lack of sugar feeding by some species or populations of some species may be more common than portrayed in the literature. Indeed, many species have only been rarely observed sugar feeding in the wild and of those species examined, some contain only a very low percentage of sugars in their crops. Foster (1995) summarizes that some species have only 1-4% of the population containing fructose. In most mosquito species, however, greater than 40% contain fructose and a few species have more than 80%.

In terms of host-finding, female sugar feeding mosquitoes may enhance their

vectorial competence and improve their success of obtaining a blood meal. Walker and Edman (1985), reported that female Ae. triseriatus and Ae. aegypti that have had access to sugar sources were more persistent (thus more successful) in their attempts at blood feeding than were sugar-deprived females. However, in terms of frequency blood feeding








7

by An. quadrimaculatus was not affected by the availability of sugar (Foster and Eischen 1987).


Floral Nectaries


The most documented source of carbohydrate for mosquitoes is floral nectaries. Haeger (1955) ranked floral nectar as most important followed by honeydew and extrafloral nectars for natural sugar sources. In a recent review article, Foster (1995) agreed that floral/extrafloral nectars and honeydew are probably mosquitoes' main source of sugar for mosquitoes, but stated that their importance was difficult to determine. What exactly is nectar? Kevan and Baker (1983) state that nectar is a phloem-sap derivative produced in a series of complex physiological processes in special glands of plants. It is composed mostly of sugars, but also may contain free amino acids, proteins, lipids, antioxidants, alkaloids, vitamins, organic acids, allantoin and allantoic acid, dextrins and inorganic materials such as minerals. Crop sugar contents of mosquitoes were 20-50% w/v as reported by Hocking (1953, 1968). The majority of evidence indicates that mosquitoes are selective about the source of sugar meals and may not feed on whatever is present.

All authors agree that the most common nectar sugars consist of sucrose, glucose, and fructose (Wykes 1951, Van Handel 1972, Baker and Baker 1983a, 1983b, Kevan and Baker 1983, Percival 1961). There tends to be disagreement, however, as to the exact composition and relative abundance of some of the less common sugars. Wykes (1951), Percival (1961), Baker and Baker (1983b) state that the monosaccharides galactose and








8

arabinose, the disaccharides maltose and melibiose, and the trisaccharides melezitose and raffinose are present in some plant species. It is rare to find a nectar with only one detectable sugar and none contain only fructose, or sucrose and fructose in the absence of glucose (Baker and Baker 1983b). Some plant families may be characterized by the relative proportions of the three major sugars. For example, certain plant families such as Lamiaceae (mints) and Ranuculaceae (buttercups) are characterized by sucrose-rich nectars (Baker and Baker 1983a). Other families such as Brassicaceace (mustards) and Asteraceae (composites) have hexose-rich nectars. Wykes (1951), Percival (1961), and Van Handel (1972) examined many plant species in several families and concluded the constituent sugars and their relative proportions in nectar tend to remain constant from any one species, while their occurrence appears to be characteristic for certain families. Percival (1961) tabulates the major plant families likely to contain the less common nectar sugars such as raffinose, maltose, and melibiose. If salivary or crop enzymatic activity is not found in the mosquito species being tested, it may be possible to provide generalizations on the relative abundance of each of the dominant sugars present in wild sugar sources.

A highly relevant, but almost completely ignored aspect of mosquito sugar feeding, is that nectar also contains detectable quantities of amino acids. If indirect determination of sugar feeding origin is not possible by examining sugar composition alone, it may be possible that examining both the amino acid and composition sugar will provide valuable clues about where mosquitoes obtain their sugar meals. As listed in descending order of commonness, Baker and Baker (1983a), report floral nectary amino








9

acids as follows (percentage in ()): alanine (96), arginine (90), serine (89) proline (87), glycine (84), isoleucine (73), threonine (67), valine (66), leucine (66), glutamic acid (62), cysteine, etc. (55), phenylalanine (55), tyrosine (52), tryptophan (48) lysine (41), glutamine (41), aspartic acid (32), asparagine (27) methionine (20), histidine (19), and nonprotein amino acids (36). Baker and Baker (1973) report the vast majority of the plant species tested to have nectar amino acids in detectable quantities. Furthermore, they state that the amino acid compliments, as well as the concentration of these acids in the floral nectar, will help to determine the nectar's "taste" to the flower visitor in addition to whatever nutritional significance it may have. Similar to sugar concentration/composition, amino acids vary depending on the nutritional requirements of the pollinator. In general, the concentrations of nectar amino acids seem to be greater if nectar is the only or the predominate source of protein for the flower visitor (Baker and Baker 1983a). For example, ants can act as selective agents, and favor plants with particular amino acids in their nectar (Lanza and Krauss 1984). Amino acids also appear to extend life expectancy of insects. Eischen and Foster (1983) reported that nectar amino acids extend survival, but are of insufficient quantity to trigger ovarian development in mosquitoes. It remains unclear as to whether mosquitoes choose their plant sugar sources based on the amino acid contents.

Many species of mosquitoes have been observed feeding on a large variety of flower species. Flowers or some plant sugars may serve as attractants. Indeed, sugarcontaining flower or honey extracts are attractants for An. quadrimaculatus, and Cx. nigripalpus (Kline et al. 1990, Hancock and Foster 1993, see Foster and Hancock 1994),








10

Breeland and Pickard (1961, 1967), observed several flood water species, Ae. vexans, Ae. trivattatus, Ps. cyansens, Ps. ferox, Ps. ciliata feeding on composites such as goldenrod (Solidago) and ageratum (Eupatorium), dock (Rumex) and ironweed (Vernonia). The composites, ox-eye daisy (Chrysanthemum), yarrow (Achillea), golden rod (Solidago), and the common milkweed (Asclepias) were the dominant feeding sources of An. earlei, An. walkeri, Cq. perturbans, Ae. canadensis, Ae. communis, Ae. sticticus, Ae. stimulans, Ae. vexans, Cs. silvestris and Cx. restuans (Grimstad and DeFoliart 1974). With the exception of milkweed, the dominant nectar sources all contained high levels of glucose (considered hexose-rich). The univoltine snow pool mosquito, Ae. provocans, preferred feeding on trees and shrubs (particularly Rosaceae) over herbaceous plants (Gadawski and Smith 1992). Magnarelli (1983) observed Ae. canadensis and Ae. stimulans commonly probing white baneberry (Actaea pachypoda), bird's rape (Brassica rapa), maple leaved viburnum (Viburnum acerifolium), and less commonly on yarrow (Achellea millefolium), evening lychnis (Lychinis alba), wild lily-of-the valley (Maianthemum canadense) and false Solomon's-seal (Smilacina racemosa). Direct field observations made by Sandholm and Price (1962), found 10 species of Minnesota mosquitoes feeding on a wide variety of flowering species from a local arboretum. Ae. vexans was present feeding on 39 species of plant, the most common being common milkweed (Asclepias syriaca), wild cucumber (Echinocystis lobata), fall phlox (Phlox paniculata), goldenrod (Solidago latifolia), and meadowsweet (Spiraea latifolia). Ae. vexans and Cx. resturans were the common species of mosquito observed by Vargo and Foster (1984), on Canada goldenrod (Solidago) and white snakeroot (Eupatorium rugosum). The most common










floral nectaries visited by Ae. vexans, and Ae. trivittatus included Queen Anne's lace, (Daucus carota), common milkweed (Asclepias), dogbane (Apocynum medium), and oxeye daisy (Chrysanthemum)Yee et al. (1992). Interestingly, although blooming yarrow was also common, mosquitoes were not observed feeding on it as in other aforementioned studies. Based on the large percentage of mosquitoes that contained crop sugars and the relatively few direct observations of flower sugar feeding, Magnarelli (1980) concluded that Psorophoraferox obtains most of its sugars from sources other than floral nectaries.

Both male and female mosquitoes seek a sugar meal shortly after emergence and continue to take nectars throughout adulthood (Yuval 1992). Males probably feed nearly every day (McCrae et al.1976, Magnarelli 1979, 1983, Reisen et al. 1986). Field observations and chemical analyses (cold anthrone test) indicate that females of mosquito species obtain sugar meals throughout the gonotrophic cycle (Magnarelli 1978, 1979, 1983, Andersson 1990, Nasci and Edman 1984, Vargo and Foster 1984, Reisen et al. 1986, Andersson and Jaenson 1987, Haramis and Foster 1990). According to Pappas and Larsen (1978), mosquitoes may remain feeding on a floral nectary for up to 18 minutes. Similarly, lab work by Jepson and Healy (1988), found that Ae. aegypti feed on flowers for an average of 21 minutes.


Extrafloral Nectaries


Plant nectaries located outside the flowers are termed extrafloral nectaries. Most extrafloral nectaries do not involve pollination, and their function is not as uniform as that of floral nectaries (Koptur 1992). Extrafloral nectaries may be present on virtually every








12

vegetative and reproductive structure. They occur on the petiole, rachis, upper and lower surfaces of the blade, leaf margin, stipules and on most external parts of the flower. Although inconspicuous and often widely scattered, extrafloral nectars are readily available on assorted ferns and about one quarter of all angiosperms (Koptur 1992). They are unknown in gymnosperms (Keeler and Kaul 1984). Koptur (1992) provides an extensive list of plant species and the locations of their extrafloral nectaries.

The sugar and amino acid contents of extrafloral nectars contain slightly different constituents compared with their floral counterparts. Extrafloral nectar has a wider range of sugars, usually being poor in sucrose (Koptur 1992). That sugar concentrations of extrafloral nectars vary over a much wider range than floral nectars of the same species. Similar to the sugar composition, the amino acid composition within a species usually also differs slightly from that of the floral nectaries (Baker et al. 1978). Most notably, the cysteine group (lysine, asparagine, and tyrosine) are more common in extrafloral nectar. Differences in amino acid composition presumably relate to attraction of different insects such as ant or wasp "guards" which may have different nutritional requirements than those of pollinators. Baker et al. (1978) reported that nonprotein amino acids are more frequent in extrafloral nectars.

There have been only a few direct observations of mosquitoes feeding on

extrafloral nectaries. Of the >20 nectar sources reported by Gadawski and Smith (1992), for Aedes provacans, only one source was extrafloral (Crataegus, hawthorn). Haeger (1955) observed Aedes taeniorhynchus feeding on the tender branch tips of buttonwood (Conocarpus erectus) on Sanibel Island Florida. More recently, Taylor and Foster (1996)








13

observed Cx. nigripalpus feeding on an extrafloral nectary on caster beans as well as several mosquito species feeding on extrafloral nectars on cashew trees.


Honeydew


Perhaps the single most underrated and common sources of sugar in dipteran

biology are honeydews. The sugary exudate, "honeydew," refers to the liquid excretions from the alimentary tract, as released through the anus by aphids, coccids, and many other plant sucking insects. Auclair (1963) reports that honeydew (a.k.a. mannas) are usually complex mixtures of a large variety of chemical compounds, including several sugars, amino acids, amides, organic acids, alcohols, auxins, and salts. Honeydew typically contains about 11% dry matter, of which 88 % are carbohydrates and 7 % nitrogenous compounds. Honeydew is probably extremely common and may be used by mosquitoes and other flies far more often than what the literature reports. Auclair (1963) also summarized that although honeydew production varies widely between Homoptera species, in general, aphids produced 1.7-20 drops per aphid every 10 hours with an average volume and weight of 0.5 mm3 and 0.9 to 8.6 mg, respectively.

Although, undoubtedly varying by species, it is currently unknown what

proportion of the medically important mosquitoes of Florida utilize honeydew regularly as a sugar source. It is possible that honeydew feeding may have at least partially influenced mosquito evolution. Downes and Dahlem (1987), convincingly argue that Diptera (Cyclorrhapha in particular) evolved ingesting honeydew. They conclude that honeydew-producing Homoptera (present in the Permian era) occurred prior to the first








14

undisputed fossil evidence of Diptera, (occurring later in the Triassic era). Honeydew feeding and/or perhaps the piercing and sucking of sugar-containing plant juices must have occurred. Flowering plants, with their associated nectaries, did not occur until much later in the Cretaceous era.

Only a few floral or extrafloral nectars contain sugars, amino acids, or other chemicals unique to specific plant families or genera. This can make generalizations about specific mosquito and plant relationships very difficult if these are based solely on comparative chemical analysis. Honeydew, however, contains oligosaccharides uncommon in plant nectars, and which are largely unique to honeydew and their presence may be indicative of feeding on honeydew. A review of the literature on the sugars commonly found in honeydews produced by Homoptera is summarized by Auclair (1963). Fructose, glucose and sucrose were present in all species tested, glucosucrose (a.k.a. fructomaltose, erlose, 4-alpha-glucosylsucrose, alpha-maltosylfructoside) and melezitose were present in most species, and maltose and maltotriosucrose were present in a few species. A melezitose hydrolysis product, turanose, is also indicative and uniquely associated with homopterous honeydew (Hudson 1946). Using several modern chromatographic techniques, Lombard et al. (1987), found fructose, glucose, sucrose, maltose, turanose, trehalose, melibiose, raffinose, melezitose and some other unknown oligosaccharides in the honeydews of several homopterous species. More recently, Bates et al. (1990) used N.M.R. and Byrne and Miller (1990) used HPLC to characterize trahalulose (a disaccharide) as the dominant honeydew sugar produced by white flies. Further analysis by Yee et al. (1996) ranked white fly honeydew sugar components as








15

follows: trehalulose > melezitose > sucrose > fructose > glucose. Some sugar alcohols are also commonly found in honeydew. Stachyose has been reported from honeydew but not in the host plant (Byrne and Miller 1990). Ribitol, dulcitol, and mannitol have also been reported from honeydew (Ewart and Metcalf 1956). These alcohols were not considered unique products of honeydew, however, but were thought to pass unchanged from the plant through the insects alimentary tract. Melezitose, glucosucrose, turanose, stachyose, and trehalulose are not, or only rarely found in plant nectar. By association, the presence of any of these sugars in the crops of mosquitoes is indicative of honeydew feeding.

Amino acids are also a common component of honeydew, comprising up to 315% of the total composition (Auclair 1963, Ewart and Metcalf 1956). The most common amino acids present in honeydew are alanine, asparagine, aspartic acid, glutamic acid, glutamine, leucine, phenylalanine, proline, serine, threonine and valine (summarized by Auclair 1963). Unfortunately, in most cases, the amino acids present in honeydew were also present in the host plant. There do not seem to be any amino acids uniquely associated with Homoptera excretion.

Mosquitoes can, and probably do, routinely ingest honeydew sugars. Honeydew probably ranges from liquid to a semi-liquid syrup to dry solid residues on stems and leaves. The latter may not pose a problem for ingestion by some species. Mosquitoes have been shown to ingest solid sugars by first liquefying with saliva. For example, Eliason (1963) found Cx. tarsalis, Cx. pipiens, Cs. incidens and Cs. inornata fed readily








16

on solid, crystalline sugar; however, An. freeborni, Ae. sierrensis, Ae. aegypti and Ae taeniorhynchus did not readily feed on solid sugars.

Just as mosquitoes appear to have preferences for certain species of plant nectars, there is no reason to believe they do not also discriminate between different insect/plant honeydews. There have been a few direct observational accounts or chemical analyze proving that mosquitoes ingesting honeydew. Nielsen and Greve (1950), observed Ae. cantans feeding on honeydew and concluded it was probably its main sugar source. They also observed Ae. taeniorhynchus, Ae sollicitans, Cx. nigripalpus, and An. atropos feeding on green aphid honeydew on Bidens spp.. Honeydew feeding by Cx nigripalpus was also observed by Haeger (1955) who found honeydew second in importance compared to nectar. Killick-Kendrick (1987) conducted laboratory experiments demomstrating that the sand fly, Phlebotomus ariasi, ingests honeydew from aphids on plants, and presented strong circumstantial evidence indicating that wild flies feed on honeydew. Using wild sand flies, MacVicker et al.(1990) categorized sugars in the crop using HPLC and found significant levels of melezitose in five sand flies species from different habitats.


Other Sugar Sources


The piercing-sucking mouthparts of mosquitoes are utilized for feeding on

animals, and could possiblly be used for feeding on plant tissues. Patterson et al. (1969) showed plant tissue feeding by mosquitoes, but was insufficient to sustain the life of mosquitoes and thought to be a means of obtaining water. Joseph (1970) observed








17

mosquitoes feeding on fruit. Other researchers (McCrae et al. 1976), Mogi and Miyagi 1989) observed mosquitoes imbibing exudates from leaves damaged by feeding insects. By products of fermenting sugars may be attractive to some mosquitoes. Worth (1975) noted that a wide variety of mosquito species were attracted to a sugar-beer-rum bait painted onto the sides of trees. One interesting theory of plant juice ingestion was proposed by Schlein and Muller (1995), who found evidence of plant tissue feeding by sand flies and mosquitoes and concluded that although low in sugar content and questionable as an energy source, the plant juices may be a means of helping to control internal parasites.


Diel Sugar Feeding Activities


Sugar and blood feeding may not be mutually exclusive events. Field data from Grimstad and Defoliart (1975), Yee et al. (1992), and Yee and Foster (1992) reported that nectar feeding and blood feeding activities occur at or about the same time of day with respect to photoperiod. Although not directly observed, both Reisen et al. (1986), and Yuval et al. (1994), found the percentage of mosquitoes containing fructose changed depending on what time of night the mosquitoes were sampled. The highest percentage occurred around sunrise for Cx. tarsalis and male An. freebornii. Conversely, however, in laboratory studies by Jepson and Healy (1988), Ae. aegypti fed on flowers most frequently at dusk or dawn, slightly different from its normal blood-seeking times. Habitat and trap type may also bias sugar feeding results. A significantly higher percentage of Cs. melanura were positive for fructose at the perimeter of its swamp








18

breeding habitat than closer to the center (Nasci and Edman 1984). For Cx. nigripalpus, differences in fructose levels depended on the habitat collected, and the type of trap used in sampling (Bidlingmayer and Hem 1973). Conversely, MacVicker et al. (1990) reported that sand flies captured from different sites and habitats did not differ significantly in the crop sugar composition between any of the species examined. Morphology and Physiology of Mosquito Alimentary Tract


Several factors appear to govern the destination of sugar meals. Nearly all of the mosquitoes investigated to date, store sugar solutions obtained from plant juices in the crop diverticula and not in the midgut (Clements 1992). The alimentary tract of the mosquito is typical of most Diptera. Near the posterior end of the foregut are two small dorsal and one large ventral diverticula or crop. Valves separate the three diverticula and the posterior foregut and midgut (see Clements 1992). In general, sugar meals, depending on their composition and concentration, initially go to the crop and are gradually released into the midgut (Friend 1981). The crop is considered a storage organ with no known secretory capabilities (Christophers 1960). Sugar meals may be partially converted into glycogen or fat, but unlike other insects and birds, mosquitoes do not use fat as fuel for flight. Mosquitoes can only use glycogen and simple sugars for flight fuel (Nayar and Van Handel 1971). Likewise, Yuval et al. (1994) showed that only simple sugars and glycogen are used during swarming. Friend et al. (1988, 1989) and Schmidt and Friend (1991), studied the effects of sugars on ingestion and diet destination in Cs. inornata. They found two response systems regulating the crop, one controlling the








19

amount ingested, the other controlling diet destination. Apparently, the destination depends on the molecular configuration of the meal. At a concentration of 0.5 M for example, common plant sugars like sucrose, maltose, fructose, alpha-glucose, and isomaltose induce crop valve opening and closing of the midgut valve. The valve closing results in the meal being shunted almost exclusively into the crop in the majority of the insects tested. Conversely, sugars not commonly found in nectars such as cellobiose, 0glucose, lactose and gentiobiose induce much less crop opening and almost no midgut closing. As for controlling the amount consumed, Friend et al. (1989), found ingesting large quantities of sucrose directed the meal to the crop, smaller sucrose meals were directed to both the crop and midgut. When sucrose concentrations were 0.4 M or more, most insects ingested large amounts and deposited it into the crop. At concentrations below 0.4 M the mosquitoes did not consume as much, and the meal went to both the crop and midgut. Contrary to popular belief, crop contents are not always strictly sugar meals. Trembly (1952) found crops of several species to contain remnants of blood meals.

In order to accurrately determine the natural sugar sources of the mosquitoes, samples must be taken as soon as possible after feeding. The crop volume varies depending on species. Aedes communis and Ae. punctor have volumes of about 0.91 and

3.39 microliters, respectively (Hocking 1953). Sugar content for a typical crop as in Cs. melanura contains about one microliter of liquid (Friend et al. 1988) and according to Schaefer and Miura (1972), typically contains about 0.087 to 0.45 mg of sugar per crop. Shortly after sugar feeding, the sugars may be wholly or partially broken down








20

enzymatically (see later section) and begin to slowly empty as needed energetically by the mosquito. In general, sugars appear to be digested in a species-specific manner and the half-life may depend on the ambient temperature and the individual energy requirement for the mosquito. As an example, Andersson and Jaenson (1987) found fructose in 90% of wild mosquitoes immediately after being caught, 30% in mosquitoes held and tested 10 hours later, and undetectable 20 hours after sampling. Likewise, Smith and Kurtz (1994) found Aedes triseriatus could empty its crop in as little as 12 hours when fed 10% glucose. If, as stated by Foster (1995), temperate and subtropical mosquitoes only ingest sugar every 2-5 days, collecting large enough samples to determine the relative proportion of the mosquito population feeding on preferential sugars sources (honeydew for example) or amino acids may prove difficult indeed.

Clements (1992) states that nearly all pentoses, hexoses, and di- and trisaccharides stimulated feeding; however, the acceptance thresholds varied between sugars or mosquito species. Conversely, trioses, tetroses and heptoses failed to stimulate feeding. Clements (1992), also states that the common nectar sugars, fructose and sucrose had low median acceptance thresholds of 0.020 and 0.023 M for Ae. aegypti, but for glucose was considerably higher at 0.11 M. The most potent phagostimulant for Cs. inornata was sucrose followed by a 1:1 mixture of glucose and fructose, maltose, fructose, and alphaglucose (Friend et al. 1988). The trisaccharides, raffinose and melezitose also stimulated gorging (Clements 1992).








21

Crop Enzymatic Activity


It is often assumed that the crop is simply a storage organ with little digestion of sugars or amino acids occurs there. Enzymatic activity in the saliva and/or crop may greatly and quickly alter the sugar and amino acid composition of a nectar or honeydew meal making it impossible to determine the exact sugar meal source based on composition and concentrations. To complicate matters further, crop and salivary enzymatic activity may vary by species. Several hematophagous Diptera contain carbohydrases in their saliva (Gooding 1975) which are presumably shunted to the crop upon ingestion of sugar meals. Marinotti and James (1990) and Marinotti et al. (1990) show Ae. aegypti to possess an alpha-glucosidase in the saliva which breaks down some sugars in the crop. Of the sugars tested, Marinotti and James (1990) indicated that sucrose was broken down the fastest, followed by maltotriose, maltopentaose and maltose. Melezitose, trehalose, raffinose, starch and others were only minimally affected. Schaefer and Miura (1972) also found carbohydrate enzymes in the saliva of Cx. tarsalis as well as invertase, maltase, melezitase, amylase and lactase. Indirect evidence for salivary and/or crop enzymes also exists. Crops of sucrose fed sand flies were analyzed with a GC and found to contain the hydrolysis products of fructose and glucose (Moore et al. 1987, Alexander 1988). However, there was no evidence of hydrolysis products for melezitose-fed flies (Alexander 1988). Interestingly, the problem of crop microbes digesting and altering a sugar meal may be reduced. Mosquito saliva contains a








22

bacteriolytic factor in the glands of both sexes and this substance is present in the crop after a sugar meal (Pimentel and Rossignol 1990, Rossignol and Lueders 1986).


Introduction To Mosquito Visual Attractants Review


Insects have long been known to have an attraction to certain wavelengths of light (Herms and Ellsworth 1934, Gui et al. 1942, Pfrimmer 1957) The attraction of many haematophageous Diptera to light is thought to be a host-seeking response. Indeed, the majority of insects haematophageous insects attracted to light traps are host-seeking females (see Service 1995). Host-seeking responses of female mosquitoes or other blood feeding Nematocera is largely due to the physiological requirement for a blood meal required to stimulate and complete ovarian development. The initiation of mosquito hostseeking is a least partially due to visual cues which are used to detect illumination levels and perceive objects (Laarman 1955, 1958). Mating, dispersal, appetitive flight, location of sugars, hosts, resting, oviposition and overwintering sites are all governed to some degree by vision (Allan et al. 1987). Many authors' have examined the important visual components of host finding and divided them into shape, color (reflected and transmitted), size, contrast, light intensity, texture and movement. All of these factors alone or in combination appear to play an important role in the ability of a blood-feeding fly to discern preferences and successfully find a suitable host or other resource.








23

Reflected Light


Although this paper is concerned mostly with host-seeking preference as it relates to various wavelengths of filtered transmitted light, much of the work done with mosquitoes and other dipteran color vision has focused on host-seeking behavior toward various wavelengths of reflected light. In general, research on reflected light research has found flies to be most attractive to darker, less reflective colors. Brown (1951, 1954) reported attractiveness to vary inversely with reflectivity or brightness between 475 and 625 nm wavelengths and found darker colors (those with lower reflectivity) more attractive than lighter colors for a variety of wild Canadian mosquito species. Browne and Bennett (1981 ) reported host-seeking female Aedes spp. and Mansonia perturbans were most attracted to black followed by red, blue, white and yellow. Similar results for An. maculipennis and Ae. aegypti respectively (Brett 1938). O'Gower (1963) reported black moist surfaces more attractive than gray moist surfaces for Ae. aegypti during hostseeking. Gilbert and Gouck (1957) also found darker shades to attract the most Ae. aegypti, and lighter shades more attractive to Ae. taeniorhynchus. Bracken et al. (1962), Granger (1970), Bradbury and Bennett (1974), Browne and Bennett (1980), and Allan and Stoffolano (1986) all found similar results for tabanids or black flies where low reflective colors like blue, black and red were more attractive than white, yellow and ultraviolet. For nocturnal mosquito species, Barr et al. (1963) reported light trap color to have some effect on catches, but concluded intensity was the most important factor.








24

Transmitted Light


Most dipteran visual research, has focused on diurnal species host-seeking preferences as a function of reflected light color. Significantly less has been written about day or night blood feeding responses to various wavelengths of transmitted light. Ultraviolet lamps have long been known to increase the numbers of host or resource seeking mosquitoes captured at light traps (Headlee 1937, Weiss 1943, Williams et al. 1955, and Breyev 1963). Electroretinograph (ERG) studies consistently show most flies possess a bimodal spectral response. ERG's conducted by Muir et al. (1992) found Ae. aegypti to have spectral sensitivities ranging from ultraviolet (323 nm) to orange-red (621 nm) with sensitivity peaks in the ultraviolet (323-345 nm) and green (523) wavelengths. Similarly, Smith (1986) reported that several tabanid species to have a peak activity of 400-600 nm and a smaller peak between 330-400 nm. No ERG references could be found for nocturnal mosquitoes. Using a different approach, Browne and Bennett (1981) tested filtered light of known wavelengths to equate host preference with landing rates for Mansonia perturbans. Shorter wavelengths (400-600 nm or blue-green) attracted significantly more mosquitoes than longer wavelengths, particularly those above 600 nm (infrared). In fact, for all wavelengths tested, the greatest numbers of mosquitoes landed on colored filters with the infrared wavelengths removed. In Georgia, Bargren and Nibley (1956) found several local species to have varying levels of attractiveness to New Jersey light traps using different color bulbs of similar intensities. Aedes vexans and Cx. salinarius were most attracted to blue (447 nm) lamps followed by red (670 nm), yellow








25

(570 nm) and white. Culexpipiensfatigans, however, was attracted to red, followed by yellow, white and blue. Culex nigripalpus showed no preference for any of the four colors tested. In field tests, Vavra et al. (1974a,b) tested several types and colors of light and with no significant difference in the numbers of mosquitoes attracted to each of the colors. Similarly, in a lab test using Culex tarsalis, Culexpipiensfatigans and Anopheles sierrensis, Gjullin et al. (1973) tested New Jersey light traps with ceramic dipped bulbs colored red, green, blue, orange and white incandescent bulbs and an ultraviolet light. No convincing difference in attraction between any of the colored lights tested were observed. Conversely, Wilton and Fay (1972) tested Anopheles stephensi, a nocturnally active mosquito, against a clear incandescent bulb and monochromatic light of various wavelengths. This species was highly attracted to 290 and 365 nm in the ultraviolet region and 690 nm in the infrared. Blue, green and yellow (490, 540, and 590 nms) were found to be repellent compared to the clear bulb.














CHAPTER 2
ANALYSIS OF SUGAR MEAL COMPOSITION OF WILD
MOSQUITOES BY GAS CHROMATOGRAPHY


Introduction


Adequate dietary sugar is a critically important aspect of mosquito and other

dipteran survival that is often overlooked. Some species have been commonly observed feeding on natural sugar sources including floral nectars (Haeger 1955, Breeland and Pickard 1961,1967, Sandholm and Price 1962, Grimstad and DeFoliart 1974, Magnarelli 1980, 1983, Vargo and Foster 1984. Yee et al. 1992, Yuval 1992, Foster 1995), extrafloral nectars (Haeger 1955, McCrae et al. 1976, Gadawski and Smith 1992, Taylor and Foster 1996), honeydew (Nielsen and Greve 1950, Haeger 1955), fruit (Joseph 1970), and other sources (Worth 1975, Patterson et al. 1969, Mogi and Miyagi 1989). Several aspects of mosquito sugar feeding have been reviewed by Yuval (1992) and Foster (1995).

Few attempts have been made to determine the exact sugar composition within dipteran crops using modem chromatography. Thin layer chromatography (TLC) has been used for pooled samples of tabanids by Magnarelli and Anderson (1977) and of mosquitoes by Magnarelli (1980). Hoppe (1983) and Burgin and Hunter (1997) used high performance thin layer chromatography (HPTLC) for qualitative determination of


26








27

sugars found in tabanids and black flies, respectively. High performance liquid chromatography (HPLC) was successfully used by MacVicker et al. (1990) to examine the crops of five wild Italian sand fly species. Despite its common use in other systems, only 4 studies have used gas chromatography (GC) for investigating dipteran diets, including Schaefer and Miura (1972) for Culex tarsalis; Moore et al. (1987) and Alexander (1988) for phlebotomine sand flies and Chang et al. (1977) in tephritid fruit flies. All of these studies analyzed pooled crop samples.

As indirect evidence for "nectar" feeding, thousands of mosquitoes and other

Diptera have been tested using the cold anthrone method that determines the presence or absence of reducing sugars (e.g. fructose) found in almost all plant nectars. Many authors (Van Handel et al. 1972, Bidlingmayer and Hem 1973, Magnarelli 1978, 1979, 1980, Reisen et al. 1986, Andersson and Jaenson 1987, Nasci and Edman 1984, Smith and Kurtz 1994 and Edman et al. 1992) have used the cold anthrone test developed by Van Handel (1972) to qualitatively establish the presence or absence of crop "fructose." According to Van Handel (1967), the cold anthrone reagent reacts with fructose, inulin, sucrose, melezitose, raffinose and others. These sugars all contain the fructose moiety and give positive or negative responses based on a resulting color reaction. Unfortunately, the cold anthrone test does not provide any information about the origin or exact composition of sugars found in the dipteran crop and provides no quantitative information. Indeed, providing there is no enzymatic breakdown of sugars in the crop, it would be possible to determine if flies have preferences for certain families of plants or types of nectar (sucrose-rich for example). Furthermore, we suspect that many of the








28

reports for "nectar" feeding due to the presence of "fructose" in the crop, may actually have been a positive anthrone reaction to melezitose (a common trisaccharide found in honeydew) or other reducing sugar not commonly found in floral nectaries.

GC is an excellent and powerful tool for qualifying and quantifying the exact dietary sugar preferences, occurrence and composition of wild flies. Processing crop contents for GC is only slightly more labor intensive than the methods used for the cold anthrone test. A quantity of less than a microliter is all that is required. The technique is relatively rapid, employing silylation (using a TSMI derivitizing agent) of crude crop contents and subsequent GC analysis. We describe here a method using gas chromatography allowing researchers to determine the individual sugar meal components for mosquitoes and other flies, or the apparent presence or absence of carbohydrases in the crops of mosquitoes and/or other flies by monitoring the decline in parent sugar concentrations, and the increase in the metabolites over time.



Methods and Materials

Specimen Preparation



Representative samples of wild mosquitoes were vacuumed from resting sites shortly after sunrise in a ca. I hectare cypress swamp surrounded by pine flat woods located north of Gainesville, FL. Mosquitoes were kept alive, chilled, and processed within four hours of capture. Specimens were sacrificed by laterally inserting a # 0 or smaller insect pin just above the mesothoracic spiracle. Legs and wings were removed








29

using fine forceps. The crop (ventral diverticulum) was exposed by grasping third or fourth abdominal segment and pulling back slowly so that the crop would emerge between the abdominal segments. For crops containing liquid contents, 1 Cl or less of material was sampled using a fine-tipped 10 pl capillary tube made from heated and pulled glass tubing. The capillary tube contents were transferred to a 200 pl glass insert tube inserted into a 3 ml glass GC sampling vial (National Scientific Company, Lawrenceville, GA) secured with a teflon lined cap. A separate vial was used for each specimen.


Sample Preparation for GC Analysis


Sugars are highly polar compounds and their analysis by gas chromatography

requires silylation to derivitize the polar carboxyl and hydroxyl groups. The derivitizing agent Tri-Sil Z," (Pierce Chemical Company, Rockford IL) composed of TMSI (N(TMS) imidazole), in dry pyridine was used to process crop contents. Tri-Sil Z" (100 pl) was added to each vial containing one crop extract and each vial was vortexed, heated at 60-700C for 15 minutes and frozen until analysis. Using a 1.0/100 ptl aliquot sample, GC was performed using an Hewlett Packard 6890 instrument with an on-column auto injector, flame ionization detector, and equipped with a DB-5 fused silica capillary column (30 m X 0.25 pm, J & W Co., Folsum, CA). The column was heated from 60 to 300'C at a ramp of 200C/min for 20 minutes. Pyridine and acetonitrile were used as solvents to clean the syringe between samples. The following sugars (Aldrich Chemicals, Milwaukee, WI) were made up as 0.01% standards in distilled water: D(-)fructose, D-








30

glucose, sucrose, maltose, D(+) melezitose, L-arabinose, L-rhamnose. D(+)melibiose, D(+)raffinose, turanose (a hydrolysis product of melezitose), and trehalose. Trehalose is a sugar present in insect hemolymph (Friedman 1985); the others have all been found associated directly or indirectly with plants (Percival 1961, Van Handel et al. 1972). Melezitose and turanose are known to be associated with honeydew (Auclair, 1963). The data and resulting chromatograms and integrations were recorded and processed using a PE Nelson 900 Series (970A) interface with Turbochrome" software (Ver 4.1, 1995 [Perkin-Elmer Corp., Cupertino, CA]).


Crop Sucrose and Melezitose Hydrolysis


Thirty to fifty laboratory reared Aedes albopictus (July 1995, USDA, Gainesville FL) pupae were placed in clean 35 mm film canister lids and allowed to emerge in new 200 ml urine cups with fitted fine mesh screen lids. A total of 25 cups with pupae were assembled. The adult mosquitoes were deprived of sugar and water for two days and kept in a rearing chamber at 28oC, 80% RH, and 14:10 L:D photoperiod. The mosquitoes then were provided with two hour unlimited access to 10% standard solutions of sucrose and melezitose. Blue food coloring was added to the sugar solutions to aid detection in the crop. Mosquitoes were fed by placing 10 drops of the standard sugar solutions atop the screens of each cup. Mosquitoes from each cup were sacrificed until several samples of males and females were found to have contents in their crops. Individual mosquito samples were taken at 2, 4, 8, and 20 hours after ingesting one of the sugar solutions. A








31

sample of the dyed standard sugar solutions was used as a control (time 0). Mosquitoes were otherwise processed and analyzed as discussed above.



Results and Discussion


Our preliminary analyses show a wide variation in the types of sugars present in wild mosquitoes. Our chromatographic analysis of combined standard sugars indicate that these common plant mono-, di-, and trisaccharides have unique retention times (Fig. 2-1). Some of the sugars, such as fructose, glucose and others, are anomeric molecules and thus display one peak for each form. Qualitative analysis of crude crop contents of wild Culiseta melanura (Coquillett) (Figs. 2-2) and Anopheles quadrimaculatus s.l. Say (Figs 2-3), indicate that most of the peaks are identifiable by GC. Figures 2-2 and 2-3 show that these female mosquitoes had recently fed on sugar sources containing fructose, glucose, sucrose, turnanose, melibiose, melezitose, raffinose, and a few unknowns. Later GC runs identified the unknown trisaccharide as erlose (glucosucrose). Honeydew has been found to contain oligosaccharides (e.g. melezitose, erlose, turanose, trehalulose) largely unique to homopteran exudates and are uncommon in plant nectars (Auclair 1963, Hudson 1946, Lombard et al. 1987, Bates et al. 1990, Byrne and Miller 1990, Yee et al. 1996). The presence of a large sucrose peak in Cs. melanura (Fig. 2-2), suggests that either salivary sucrase is not present in this species, or more likely, was not secreted at the time of feeding. Another possibility is that the sugar source contains carbohydrase-








32

inhibiting compounds that prevents the immediate hydrolysis of sucrose into glucose and fructose.

Much of the evidence for mosquito nectar feeding has come from direct field

observations rather than through specific chemical qualitative crop content analysis. West and Jenkins (1951), Sandholm and Price (1962), Gadawski and Smith (1992), Breeland and Pickard (1961, 1967), and Magnarelli (1977) have observed some mosquito species feeding at sugar sources during the day. Others have observed them feeding on sugar sources at night, dawn or dusk (Grimstad and DeFoliart 1974, Magnarelli 1983, Andersson and Jaenson 1987, Bowen 1992, Vargo and Foster 1984, Yee et al. 1992). Many mosquito species have been found to contain "fructose" in their crops (Bidlingmayer and Hem 1973), yet some of these species have not been observed sugar feeding.


Sucrose and Melezitose Fed Mosquitoes


Extensive work has been done characterizing protein acquisition and the enzymatic processes of protein (blood meal) digestion in the mosquito midgut. Relatively little, however, has been done to assess the relative importance of sugar acquisition and its subsequent digestion. GC analysis provides a rapid method of indirectly determining the presence or absence of carbohydrases in the crop (from saliva shunted to the crop).

A total of 45 samples of sucrose and melezitose-fed Ae. albopictus were processed. Figure 2-4 shows the time course of crop sucrose and melezitose,








33

respectively, for males and females following ingestion. Almost complete hydrolysis (>90%) of sucrose occurred within 2 hours of ingestion, while melezitose remained relatively unchanged even 8 hours after ingestion. There was no significant difference between the response of the males and females fed sucrose or melezitose (p = 0.61 and

0.65 respectively). In general, the crops were largely empty 16 hours after ingestion and the number of unidentified peaks increased after 4 hours in the crop. Figure 2-5 shows a representative chromatogram of crude crop sucrose extract 2 hours after ingestion. This chromatogram shows sucrose broken down into the anomers of fructose and glucose. Conversely, melezitose-fed Ae. albopictus remained relatively unaltered 2 hours after ingestion (Figure 2-6) and was found to remain largely unhydrolyzed even 20 hours after ingestion.

Crop and salivary enzymatic activity undoubtably varies by species. Depending on the species and type of sugar, GC analysis can reveal important primary sugar meal preferences. Only a few hematophagous Diptera have been reported to contain salivary carbohydrases (Gooding 1975), which are presumably shunted to the crop upon ingestion of sugar meals. Marinotti and James (1990) and Marinotti et al. (1990) demonstrated that Ae. aegypti possessed an alpha-glucosidase in the saliva which works to break down some sugars in the crop. Of the sugars tested, they found that sucrose was broken down the fastest, followed by maltotriose, maltopentaose and maltose. Melezitose, trehalose, raffinose, starch and others were only minimally affected. Indirect evidence for salivary and/or crop enzymes also exists. Schaefer and Miura (1972) found carbohydrases in the saliva of Culex tarsalis including invertase, maltase, melezitase, amylase and lactase.








34

Crops of sucrose-fed sand flies were analyzed with GC and found to contain the hydrolysis products fructose and glucose (Moore et al. 1987, Alexander 1988). Likewise, in preliminary GC runs, there was no evidence of hydrolysis products for melezitose-fed flies (Alexander 1988). At least for some species, it is possible to use GC to determine the proportions and sugars present in the crops of wild mosquitoes and the relative importance that melezitose or other unhydrolyzed sugars play in the diets of medically important Diptera.

In most cases, sugar-fed female mosquitoes should be able to enhance their survival, host finding ability, and ultimately improve their chances of vectoring pathogens. A few studies have shown that female sugar-fed mosquitoes of 3 species have lower host avidity than starved or water-fed females (Foster and Eischen 1987, Xue and Barnard 1997). Other authors, however, have found that females with access to sugar sources are more persistent in their attempts at blood feeding than sugar and water starved females (Walker and Edman 1985). Likewise, Kelly and Edman (1996) found higher parasite transmission rates in sugar-fed Ae. aegypti.

This paper proposes a technique allowing an accurate representation of sugars present in individual mosquito meals. Unlike other studies which typically use pooled samples, our GC technique, when applied to sufficient sample sizes, can provide valuable insight into the occurrence, composition, and importance of certain life-sustaining sugar sources for wild mosquito populations. From this information, attractants or other methods could be developed to help control or improve sampling of mosquito populations.








Disaccharides Sugar Standards

-oo Monosaccharides g
800a
S.00 Trisaccharides
700 %.oo 2

o 400
* 300 200 2
100
0
8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5


Time (min.)

Figure 2-1. Chromatogram of combined standards (ca. 0. 1%) for common sugars associated with plants. Multiple peaks of
the same sugar are anomeric forms.










aLL Cs. melanura- 1

900
800 700




400

300

200










9 oo Cs. melanura 2

800 2 700


400

S00



200 22 100 ME

0
8.0 8.5 0.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5


Time (min.)


Figure 2-2. Representative chromatograms of crude crop extracts from two wild individuals of Cs. melanura. Multiple
peaks of the same sugar are anomeric forms.








0
C


0oo- An. quadrimaculatus 1
800
U
700

60o 0

- 500 0

S400 2 0
Ou






300
200 2 2




0 D




















Time (m n.)
Figure 2-3. Representative comatorams of crude crop extracts from two wild individuals oAn. qdrimcltaculatus s2
Multiple eaks of the same suar are anomeric forms
700 2





0 400 2 W CD 300 VE

200

100


8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5


Time (min.)

Figure 2-3. Representative chromatograms of crude crop extracts from two wild individuals of An. quadrimaculatus.
Multiple peaks of the same sugar are anomeric forms.








38


100 11 S80



S-0- Sucrose Fed Females
40 ----------------- Sucrose Fed Males
20


0 I I I
0 5 10 15 20 Hours Post Ingestion









Figure 2-4. Timed feeding trials for male and female Ae. albopictus showing the
hydrolysis of melezitose and sucrose occurring in the crop (n = 2 to 5).












0

L2 Hr. Sucrose
300 0
U

E o

O
0

o
100 .




0
8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5

Time (min.)



Figure 2-5. Representative chromatogram showing hydrolysis of sucrose into fructose and glucose 2 hours following
ingestion in Ae. albopictus. Multiple peaks of the same sugar are anomeric forms.

















soo0 2 Hr. Melezitose

500

400
0
CL 300

200

1 0 0



8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5

Time (min.)






Figure 2-6. Representative chromatogram showing hydrolysis of melezitose 2 hours following ingestion in Ae. albopictus.














CHAPTER 3
SUGAR MEAL COMPOSITION OF FIVE NORTH CENTRAL FLORIDA
MOSQUITO SPECIES AS DETERMINED BY GAS CHROMATOGRAPHY


Introduction


That male and female mosquitoes and other medically important Diptera feed on a variety of sugar sources of plant origin has long been established. For many species, however, their sugar sources remain speculative. Direct evidence of sugar feeding is documented by observing mosquitoes feeding on a variety of natural sources. These sugar sources include floral nectars (Haeger 1955, Breeland and Pickard 1961,1967, Sandholm and Price 1962, Grimstad and DeFoliart 1974, Magnarelli 1980, 1983, Vargo and Foster 1984, Yee et al. 1992), extrafloral nectars (Haeger 1955, McCrae et al. 1976, Gadawski and Smith 1992, Taylor and Foster 1996), honeydew (Nielsen and Greve 1950, Haeger 1955), fruit (Joseph 1970), and other sources (Worth 1975, Patterson et al. 1969, Mogi and Miyagi 1989). As indirect evidence of sugar feeding, most researchers have used the cold anthrone test (Van Handel 1972), to determine the occurrence of crop fructose (or other reducing sugars) in wild populations. According to Van Handel (1967), the cold anthrone test reacts with sugars containing the fructose moiety (sucrose, melezitose, raffinose, etc.) giving positive responses based on a color reaction. Many authors (Magnarelli 1978, 1979, 1980, Reisen et al. 1986, Andersson and Jaenson 1987, 41








42

Nasci and Edman 1984, Smith and Kurtz 1994, Edman et al. 1992, Holliday-Hanson et al. 1997) have used the cold anthrone test to qualitatively establish the presence of crop "fructose." However, the cold anthrone test does not provide any ecological information about the origin, distribution or exact composition of sugars found in the dipteran crop and provides little quantitative information. Recent reviews by Yuval (1992) and Foster (1995) covered much of what is known about mosquito sugar-feeding ecology.

Few attempts have been made to use modem chromatography to determine the

composition of sugars found in dipteran crops. Thin layer chromatography has been used for pooled samples of tabanids (Magnarelli and Anderson 1977, Hoppe 1983), mosquitoes (Magnarelli 1980) and individual black flies (Burgin and Hunter 1997). High performance liquid chromatography was used by MacVicker et al. (1990) to examine Italian sand fly crops. Despite gas chromatography's (GC) common use in other systems, only 4 studies have used GC for investigating dipteran diets, including Schaefer and Miura (1972) for Culex tarsalis; Moore et al. (1987) and Alexander (1988) for phlebotomine sand flies and Chang et al. (1977) in tephritid fruit flies. All of these GC studies analyzed pooled crop samples.

Burkett et al. (1998a) presented a method and included representative

chromatograms for analyzing sugar meal composition and hydrolysis in individual mosquitoes and found that GC can be used to show honeydew feeding based on crop sugar contents (e.g. melezitose, erlose, turanose). However, they did not assess the importance of honeydew in the diet in different mosquito species. Honeydew is speculated to be a common sugar source for mosquitoes, but its relative importance








43

among different taxa has not been established. This study identifies many of the mono-, di-, and trisaccharides found in field collected individuals of 5 mosquito species and shows that honeydew feeding is an important dietary component of wild mosquito populations.


Methods and Materials


Specimen Preparation


Wild adult mosquitoes (Anopheles quadrimaculatus s.l. (Say), Coquillettidia perturbans (Walker), Culiseta melanura (Coquillett), Culex nigripalpus Theobald, and Psorophoraferox (von Humboldt)) were collected from several locations in and around Gainesville, FL over the course of 2 seasons (April-October). Ps. ferox were aspirated while attempting to take a blood meal. All other species were vacuumed from typical resting sites (i.e., under bridges, in tree holes, bases of trees, etc.). All collections were made from 0700-0900 and mosquitoes were kept alive, chilled, identified, and processed within four hours of capture. Individuals were sacrificed by laterally inserting a #0 insect pin just above the mesothoracic spiracle. Legs and wings were removed using fine forceps. The crop (ventral diverticulum) was exposed by grasping the third or fourth abdominal segment and pulling back slowly so that it would emerge between the segments. For crops containing liquid contents, I pl or less of material was sampled using a fine-tipped 10 ptl capillary tube made from heated and pulled glass tubing. Capillary tube contents were transferred to a 200 1l glass insert tube inserted into a 3 ml








44

glass GC sampling vial (National Scientific Company, Lawrenceville, GA) secured with a teflon lined cap. A separate vial was used for each specimen.



Sample Preparation for GC Analysis


Sugars are highly polar compounds. Their analysis by GC requires silylation to derivitize the carboxyl and hydroxyl groups. The derivitizing agent, Tri-Sil Z (100 pl) (Pierce Chemical Company, Rockford IL), was added to each vial containing one crop extract. Each vial was vortexed, heated at 60-70'C for 15 minutes and frozen until analysis. Using a 1.0/100 pC aliquot sample, GC was performed on one of the following instruments. Some samples were run using an Hewlett Packard 6890 instrument with an on-column auto injector, flame ionization detector (FID), and equipped with a DB-5 fused silica capillary column (30 m X 0.25 mm, 0.25 tm J & W Co., Folsum, CA). The column was heated from 60 to 3000C at a ramp of 20C/min. Pyridine and acetonitrile were used as solvents to clean the syringe between samples. Depending on availability, other sample runs were made using either an HP 5890 GC, split-splitless injector (250oC), FID, J&W DB-5 (30m X 0.32 mm, 0.25 lim), heated from 125 to 300C at 50C/min., 15 min. hold or a Varian 3700 GC, split-splitless injector (250oC), J&W DB-1 (30 m X 0.32 mm, 0.32 pLm) heated from 125-285oC at 5oC/min., 18 min hold. The resulting chromatograms and integrations were recorded and processed using a PE Nelson 900 Series (970A) interfaced with Turbochrome* software (Ver 4.1, 1995 [Perkin-Elmer Corp., Cupertino, CA]).








45

Standards


The following sugars (Aldrich Chemical, Milwaukee, WI) were made up as

0.01% standards derivatized as above: D(-)fructose, D(-)glucose, sucrose, maltose, D(+) melezitose, L-arabinose, L-rhamnose, D(+)melibiose, erlose (glucosucrose), D(+)raffinose, turanose (a hydrolysis product of melezitose (Hudson 1946)), and trehalose. Trehalose is a sugar present in insect hemolymph (Friedman, 1985), but also found in some honeydews (Hendrix and Wei, 1994). Turanose, melezitose and erlose are known to be associated with honeydew (Hudson 1946, Auclair 1963, Wolf and Ewart 1955). The other sugars have been found associated directly or indirectly with plants (Percival 1961, Van Handel et al. 1972).


Results and Discussion


The sugars reported here were unambiguously identified by comparison of

retention times on 2 GC columns of slightly different polarities. Crop samples contained fructose, glucose, sucrose, maltose, turanose, melezitose, raffinose, erlose, and traces of arabinose, melibiose and rhamnose (Figures 3-1, 3-2, 3-3, 3-4 and 3-5). Fructose and glucose were found in all samples that contained sugars. All species tested contained trehalose, a sugar not found in nectar, that could have originated from hemolymph (Friedman 1985) or honeydew (Hendrix and Wei 1994). Maltose and turanose had similar retention times and could not be distinguished with confidence. A total of thirty unknown peaks were observed across all samples. Most of these unknown compounds








46

eluted during times atypical for mono-, di- or trisaccharides and were probably either incompletely silyated sugars or other miscellaneous crop components. Rarely were the unknown peaks consistent between species, and few constituted a significant percentage of any of the samples.

The percentage of mosquitoes containing detectable crop sugars ranged from 10-11% in An. quadrimaculatus and Ps. ferox to 47.7% in Cq. perturbans (Table 3-1). Crop sugar occurrence was significantly greater in females than males in An. quadrimaculatus (chi-square, p=0.0006) and Cs. melanura (p=0.00004). These findings agree with those for other Florida mosquitoes. Bidlingmayer and Hem (1973), for example, found similar sugar positivity results using the cold anthrone test for An. quadrimaculatus (? =15%); Cs. melanura (? =36%, e=29%); Ps. ferox (? =20%); and Cx. nigripalpus (? and e=17%). Moreover, Magnarelli (1978) obtained similarly high values for Cq. perturbans (? =57%). Our results differed only in that more female An. quadrimaculatus contained sugars than males. The diverse sugar composition, proportion, and occurrence found in the crops of all species tested show that these mosquitoes tend to be opportunistic and that a variety of sugar sources are used. Ecological Significance of Honeydew in Mosquito Diet


The widespread and often invisible nature of honeydew complicates direct feeding observations (Foster 1995). Chromatography has been used to identify honeydew sugars in mosquitoes (Schaefer and Miura 1971, Burkett et al. 1998b) and other Diptera (Hoppe 1983, MacVicker et al. 1990, Moore et al. 1987, Burgin and Hunter 1997. However, until








47

now the proportion of mosquito populations feeding on honeydew and its ecological significance has only been speculative. All species tested (Figs. 3-1, 3-2, 3-3, 3-4 and 35) contained melezitose and/or erlose (An. quadrimaculatus (55%), Cs. melanura (33%), Cx. nigripalpus (15%), Cq. perturbans (10%), and Ps. ferox (7%)). These data show honeydew to be an extremely important resource for both Cs. melanura and An. quadrimaculatus comprising 1/3 and 1/2 of the sugar meals, respectively. These observations make sense, because unlike Cq. perturbans (Grimstad and DeFoliart 1974, 1975), Ps. ferox (Magnarelli 1980) and Cx. nigripalpus (see Nayar 1982), which have been observed feeding on a variety of natural sugar sources, An. quadrimaculatus and Cs. melanura have not been reported to feed on floral/extrafloral nectars. Of the mosquitoes investigated in this study, only Cx. nigripalpus had been previously observed feeding on honeydew (Haeger 1955).

Chromatography on a relatively small number of Homoptera has found honeydew to be a diverse complex of 20 or more sugars (Hendrix and Wei 1994), many of which are not found in plant tissues (Byme and Miller 1990, Tarczynski et al. 1992). A comparison of honeydews among different Homoptera shows that unique sugars are found among various families (Gray and Fraenkel 1953, Wolf and Ewart 1955, Ewart and Metcalf 1956, Lombard et al. 1987, Bates et al. 1990, Hendrix and Wei 1994, Wei et al. 1996,1997, Yee et al. 1996). It is reasonable to expect that different Homoptera produce different sugars and ratios of sugar components that would be identifiable using GC or other modem chromatographic techniques such as HPLC. Although yet to be determined, mosquitoes may prefer honeydew from certain groups of Homoptera.








48

Percival (1961), and Van Handel et al. (1972) characterized nectar sugars and

ratios for many plant species. We initially hoped that GC could be used for determining which plant(s) or even family of plants that mosquitoes use for sugar meals based on direct comparisons with published reports. Unfortunately, unlike melezitose and other trisaccharides, sucrose is rapidly hydrolyzed into fructose and glucose in insect crops (Schaefer and Miura 1972, Burkett et al. 1998b) negating any possibility of determining from which plant a sugar meal came. One interesting observation, however, is that some crop samples from An. quadrimaculatus, Cs. melanura and Cx. nigripalpus contained relatively large amounts of sucrose in their crops, showing that either some sugar sources contain carbohydrase (sucrase) inhibiting compounds, or that salivary sucrase is not always produced and shunted to the crop with sugar meals. Sugar Feeding Field Observations


From 1900-1100 hrs in July and August 1997, two species of plants were

commonly observed as hosts for sugar-feeding mosquitoes. Ps. ferox, Ae. albopictus, Cx. nigripalpus, Cx. (melanoconion) spp., and Uranotaenia sapphirina were observed feeding on the extrafloral nectaries of partridge pea (Cassiafasciculata Michx.). More interesting from a public health perspective was the observation of many insects including mosquitoes such as Cx. nigripalpus and Ae. albopictus feeding on the sugary exudates from the developing fruits of bahiagrass (Paspalum notatum Fluegge). Analyses of sugars from samples of bahiagrass exudates matched those from the crops of Cx. nigripalpus sampled while feeding on bahiagrass, showing raffinose to be one of the








49

primary sugars. Bahiagrass is a commonly planted turf and pasture grass whose characteristic "V" shaped seed heads are a familiar sight in residential areas in north central Florida. The presence of bahiagrass around human domiciles, and the absence of other suitable sugar sources may influence epidemiological factors for some of Florida's medically important mosquitoes.

GC has proved to be an excellent and powerful tool for identifying and

quantifying the occurrence and composition of sugar feeding in Diptera. In some cases, the sugar source can be determine (i.e., honeydew). Extracting/analyzing crop contents is not difficult, requiring only small quantities (0.1-1 Pl) of crude crop content. Future efforts could address/utilize larger sample sizes, test for seasonal differences, run a variety of different honeydews, and incorporate mass spectrometry to determine unknowns. From this information, attractants or other methods could be developed to help control or improve sampling of mosquito populations.















Table 3-1. Summary of percent wild mosquitoes containing identifiable sugars using gas chromatography.

Males Females % Total With % Males With % Females With Species n
n n Sugars Sugars Sugars
An. quadrimaculatus 370 119 251 11 3.4 15.9 Cq. perturbans 44 17 27 47.7 47.1 48.1 Cs. melanura 471 129 342 26.7 10.1 35.5 Cx. nigripalpus 184 41 143 27.5 26.8 27.8 Ps. ferox 134 8 126 10.4 12.5 11.1















O
100
SAn. quadrimaculatus 4
0
80

a)
CL 60.. V
(U
T=40



0 0
0 O)



U lo-,.,,



Figure 3-1. Representation of crop sugar distributions and proportions of wild An. quadrimaculatus (n = 33). The presence
of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. 'Maltose and turanose have
identical retention times.













O
100-
o Cq. perturbans
0 80-Q. 60-o) 40-0
a. 20--

S 0
0 0 E 2 LE w L


Figure 3-2. Representation of crop sugar distributions and proportions of wild Cq. perturbans (n = 20). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. 'Maltose and turanose have identical
retention times.














O

Cr 100
DCs. melanura
0

80 V0

0
o 60 --


0)
40 O
0
CL OI
020-- O 0 9

S0 0 v


E 2 -- 0



Figure 3-3. Representation of crop sugar distributions and proportions of wild Cs. melanura (n = 99). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. 'Maltose and turanose have identical
retention times.














O
.- 100
Cx. nigripalpus a

0o 80
M A v a 60
0..

o) 40 -0- 20 -0 a



,' E 2 N W I-- -Figure 3-4. Representation of crop sugar distributions and proportions of wild Cx. nigripalpus (n = 26). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. 'Maltose and turanose have identical
retention times.
















.0 100-Ps. ferox CT
o 80 -S60 A I
40-


S20




J: c I-Figure 3-5. Representation of crop sugar distributions and proportions of wild Ps. ferox (n = 14). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. 'Maltose and turanose have identical
retention times.














CHAPTER 4
FIELD EVALUATION OF COLORED LIGHT EMITTING DIODES AS
ATTRACTANTS FOR WOODLAND MOSQUITOES AND OTHER DIPTERA IN NORTH CENTRAL FLORIDA


Introduction


Light emitting diodes (LED) were evaluated as an alternative light source for use as an adult mosquito attractant. Much of the research on Dipteran color preference is based on reflected light (Brett 1938, Bracken et al. 1962, O'Gower 1963, Granger 1970, Bradbury and Bennett 1974, Browne and Bennett 1980, 1981, Allan and Stoffolano 1986). Many authors have shown that mosquitoes are attracted to transmitted light (Headlee 1937, Weiss 1943, Williams et al. 1955, Breyev 1963, Bargren and Nibley 1956, Gjullin et al. 1973, Wilton and Fay 1972, Browne and Bennett 1981). Few reports detail the response of individual species. Several colors (100 nm width) of highly efficient, low cost, "super bright" LEDs have recently been developed. These colored LEDs when used in CDC traps have a greater intensity and require significantly lower amounts of energy (ca. 0.125 ma/hr vs. 150 ma/hr for standard CM-47 bulb). We evaluated the relationship between transmitted light color and its attractiveness to woodland mosquito and other dipteran species.





56








57

Methods and Materials


Three field trials were conducted using standard CDC-type traps (John W. Hock Company, Model 512, Gainesville FL) modified by replacing the standard bulbs with the LEDs. The LED was secured into a piece of 2 X 2 cm plexiglass and fastened to the "screen" atop the lid assembly 3 cm below the aluminum trap lid (Fig. 4-1). Trial 1 used 9 kg compressed-gas (carbon dioxide) cylinders equipped with double-stage (Victor Equipment Company, Model VTS453B-320, Denton TX) and microregulators (Nupro Inc., Series M, Wiloughby OH) to maintain a constant gas flow of 200 ml per minute. Carbon dioxide was delivered to the trap through a 3 m piece of 8 mm clear plastic tubing secured with a rubber band so the top of the tubing was even with the top of the trap opening. Gas flow was checked each morning and evening using an in-line flowmeter (Gilmont Instruments, Great Neck NY, no. 12). Mosquitoes attracted near the trap intake were drawn in by the trap fan, blown through a screen funnel and into a quart polypropylene jar containing a 3 X 6 cm piece of dichlorvos impregnated vinyl strip used as a killing agent. Batteries consisted of 6V, 10 ampHrs rechargeable gel cell batteries (Powersonic Corp., San Diego CA) which were used to run the fan motor and standard incandescent light.


Six Colors With and Without Carbon Dioxide (Trials 1 and 2)


From July 15-20 (Trial 1) and July 22-27 (Trial 2), 1996, different colored lights were used as attractants in standard CDC-type surveillance traps using a 6 X 6 Latin








58

square design. Light, day, and position effects were evaluated using a three-way ANOVA (SAS Institute, 1995) for the total number and most common species represented in the traps. Multiple comparisons were made using the Ryan-Einot-Gabriel-Welsh multiple range test (alpha = 0.05). Trial 1 used carbon dioxide as an additional attractant and Trial

2 used the same randomization, but did not use carbon dioxide. Four different colored "super bright" LEDs (Toshiba Tosbright," Martech Optoelectronics, Latham, NY) were compared to no light and a standard incandescent bulb (John W. Hock Company, CM-47, 6.3 V, 520 millicandela [mcd], Gainesville, FL) used as controls. The diodes tested were red (613 50 nm, 1600 mcd, 220 ); orange (605 50 nm, 2000 mcd, 220); yellow (587 50 nm, 2300 mcd, 22') and green (567 50 nm, 2400 mcd, 80). Each LED was powered by 2 alkaline D cell batteries at 2.8 0.2 volts and 18 2 ma. A 10 ohm resistor was placed in series to prevent over driving the LED.

Trials 1 and 2 were conducted at the University of Florida's Austin Cary Memorial Forest, a research area located north of Gainesville, FL. The habitat consisted of a cypress swamp surrounded by pine flatwoods. Traps were hung 165 cm above ground level and placed every 30 meters near the banks of a seasonal forest stream that runs through the middle of the swamp. Due to the thick vegetation, none of the traps were visible to each other. The trapping period followed a 24.5 and 6.83 cm rainfall on July 5 and 9, respectively.

These trials ran from 1800 to 0800 hr for six days in a row using a 6 X 6 Latin

square design. The trap and motor assembly remained stationary, but the lights or diodes were changed nightly so each light would occupy each position during the six-day period.








59

After each trap night, the captured mosquito adults were identified and counted. All Culex (Melanoconion) Theobald, Anopheles quadrimaculatus Say, An. crucians Wiedemann and Aedes atlanticus Dyar and Knab/Ae. tormenter Dyar and Knab species were pooled, as these taxa could not be distinguished with confidence. Eight Colors With Carbon Dioxide (Trial 3)


Trial 3 was conducted from August 12-21 (Trial 3), 1996. Due to the fluctuating water levels and mosquito populations at the Austin Cary Forest site, a similar, but more permanent cypress swamp habitat was chosen north of Gainesville. In addition to the four LED colors and two controls previously discussed, two additional LED wavelengths, infrared (940 +50 nm, 22 [Martech Optoelectronics, Latham NY, model MTE1080]) and blue (450 50 nm, 800 mcd, 22' [Panasonic," Digikey Corp, Thief River Falls, MN]]) were evaluated also. Using an 8 X 8 Latin square design, traps were placed around the perimeter of the swamp. Each trap was baited with 200 ml/min carbon dioxide as in trial 1 and otherwise treated as before.

Results


Six Colors With and Without Carbon Dioxide (Trials I and 2)


During the six trap-nights of trials I and 2, 32,059 and 1,916 specimens of

mosquitoes were collected respectively. The mosquito species composition attracted to the incandescent light trap agree with those found by Mann (1993). Response of the most numerous mosquito species are shown in Figures 4-2, 4-3, 4-4 and 4-5. Means, standard








60

errors, p-values, and significant differences for species represented in large enough numbers are shown in Tables 4-1 and 4-2. As noted in the tables, there were significant trap-position and day effects for some species.

No significant differences were observed for the total number of mosquitoes

captured at different colors in either the C02 baited or unbaited trials (p = 0.08, p = 0.24, respectively). Differences were observed for individual species however. In trial 1, Aedes dupreei (Coquillett), Ae. infirmatus (Dyar and Knab), Anopheles crucians s.l., Culiseta melanura ((Coquillett), and Uranotaenia sapphirina (Osten Sacken) showed significant color preferences. In trial 2, only Ae. atlanticus, An. crucians s.l., and Ur. sapphirina showed significant preferences. Aedes dupreei was the predominant species and was the only species preferring the carbon dioxide baited trap using no light. This species was also abundant in the no light control. Three female Lutzomyia shannoni (Dyar) and 1 Lutzomyia vexator (Coquillett) were collected during trial 1. None were collected in trial

2.


Eight Colors With Carbon Dioxide (Trial 3)


During the eight trap-nights, 4,668 specimens of mosquitoes, 1,189 tabanids (Diachlorusferrugatus Osten Sacken), 3,667 chaoborids (Corethrella spp.), and 3 phlebotomine sand fly specimens were collected. Response of the most numerous mosquito species is shown in Figures 4-6 and 4-7. There was a highly significant difference in the total numbers of mosquitoes captured for the different colors (p =

0.0001). Means, standard errors, p-values, and significant differences for species








61

represented in large enough numbers are shown in Table 4-3. As seen in trials 1 and 2, there were significant trap-position and day effects for some species. Overall capture of mosquitoes was significantly greatest with the standard white broad spectrum incandescent, followed by blue, green, orange, yellow, red, no light control, and infrared respectively. When collections were classified by mosquito species, clear preferences were seen between species. Anopheles crucians s.l., Cs. melanura, Cx. nigripalpus Theobald, Ps. columbiae (Dyar and Knab) and Ur. sapphirina showed significant color preferences. White light captured the most An. crucians s.1.. The greatest numbers of Cs. melanura were captured in traps with white, green, and orange. The most Ps. columbiae were collected in traps with blue and significantly more Ur. sapphirina were captured in traps with standard white or blue. No colors were found to be repellent to mosquitoes when compared to the no light controls. No significant difference (p = 0.26) for color attraction were obtained for the tabanid, Diachlorusferrugatus. Conversely, the chaoborids, Corethrella spp., showed significant color attraction (p = 0.002), preferring white and blue over the other colors.


Discussion


Many common Florida woodland mosquitoes are medically important. Although one of the primary means of evaluating the presence/absence and relative abundance of certain mosquitoes is through the use of light traps, few studies have evaluated mosquito response to different wavelengths of transmitted monochromatic light. Even fewer studies have detailed the response of individual species. Browne and Bennett (1981) tested








62

filtered light of known wavelengths to equate host preference with landing rates for Coquillettidia perturbans (Walker). They found shorter wavelengths (400-600 nm or blue-green) attracted significantly more mosquitoes than longer wavelengths. Their results correspond well to ours in Trial 3. In Georgia, Bargren and Nibley (1956) found Ae. vexans, Cx. salinarius, and Cx. quinquefasciatus to have varying levels of attractiveness to New Jersey traps using different color bulbs of similar intensities. Other species, such as Cx. nigripalpus, showed no preference for any of the four colors (447, 570, 659, 670 nm) tested. This finding agrees with ours in Trials 1 and 2, but differs from those of Trial 3 that found significant color preferences (blue, green, white) for Cx. nigripalpus. Vavra et al. (1974a) tested several types and colors of light and found no significant differences in the total numbers of mosquitoes attracted to each of the colors. Attraction of individual species of mosquitoes was not examined. In a laboratory test using Cx. tarsalis, Cx. quinquefasciatus and An. sierrenis, Gjullin et al. (1973) tested New Jersey light traps equipped with either ultraviolet light or ceramic dipped bulbs colored red, green, blue, orange, and white. They found no significant differences in attraction between any of the colors tested. Wilton and Fay (1972) tested An. stephensi against a clear incandescent bulb and monochromatic light of various wavelengths. They found this mosquito highly attracted to 290 and 365 nm in the ultraviolet region and 690 nm, but that blues, greens and yellows (490, 540, and 590 nms) were found not as attractive as the clear bulb.

Allan et al. (1987) stated that crepuscular and nocturnal biting flies are unlikely to have well-developed color vision, but their abilities to detect differences in intensity








63

contrast are likely to be well-developed. Spectral sensitivity (i.e. relative sensitivity of the retina to light of different wavelengths) studies consistently showed most flies possess a bimodal spectral response with peaks around 340 and 525 nm (White 1985). Considering the variation in mosquito species attractiveness to light traps (Huffaker and Back 1943, Bidlingmayer 1967), it is not unreasonable to expect differences in color preference based on variations in their spectral sensitivities.

Alternatively, attractiveness may not be due to color per se. Different wavelengths may be physiologically perceived as more intense and subsequently more attractive. Barr et al. (1963) concluded that more intense light (up to a point) is more attractive than less intense light. Electroretinographs (ERG) conducted by Muir et al. (1992) found Ae. aegypti to have spectral sensitivities ranging from ultraviolet (323 nm) to orange-red (621 nm) with sensitivity peaks in both the ultraviolet (345 nm) and green (523 nm) wavelengths. By studying the relationship between illumination and suitability of an oviposition site for Ae. aegypti, Snow (1971) found a similar bimodal response. Unfortunately, all mosquito spectral sensitivity studies have focused on Ae. aegypti, none have been conducted on nocturnal or other species actually commonly attracted to artificial light.

Unlike the standard bulbs which radiate light in a 3600 pattern, the diodes emit a narrow beam (8 or 220), which in these tests was oriented upwards and reflected off the shiny surface of the aluminum pan covering the CDC trap. If the mosquitoes are responding to differences in perceived intensity, an assortment of several of the most promising LED colors (e.g., blues and greens) could be arranged to produce a 3600 pattern








64

and perhaps greatly increase the efficiency of the trap. Several LEDs in series would be many times brighter and still use significantly less battery power than a single incandescent bulb. Several of the ERG studies previously mentioned have shown peak dipteran spectral activity from 450 to 550 nm. Based on the numbers of certain mosquito species that were attracted to the blue and green wavelengths, an LED producing wavelengths between 450 and 550 nm may produce excellent results. Currently technology limits production of LEDs producing these wavelengths. Super-bright blues (450 nm) have only recently become available, and perhaps future technology will produce a blue-green diode peaking at about 500 nm.

LEDs run on significantly lower amounts of energy (ca. I ma/8 hrs) than incandescent bulbs resulting in substantial savings in battery life and expense. Hours of use (means SEM) with the no light control (697.5), blue (639) and green (635.7) were found to last significantly more hours (n= 4, p=0.02) than the standard white (360) bulb. For convenient use, LEDs can be soldered in series directly into the existing trap circuitry. Best results for all colored LEDs except blue (100 ohm) were obtained using 200 ohm resistors connected to the light motor assembly. Future studies should focus on combinations of colors oriented in different directions. The use of "superbright" LEDs warrants serious consideration as a mosquito replacement for standard incandescent bulbs used in light traps. These results have potential for use in population dynamics studies or for enhancing the attractivity of certain species.












Table 4-1. CDC Trap counts using colored LEDs or incandescent light with 200 ml/min C02 (Means SEM). Means within each row having the
same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). n = 6 nights.

Species Red Orange Yellow Green No Light Incandescent P-Value

Ae. dupreei 3 549.71108.9ab 619.7163.lab 411.798.5b 502.5117.1b 825.8205.9a 499.8i112.5b 0.02
Ae. fulvus pallens 0.30.3a 0.20.2a 0.50.3a 0.00. Oa 0.50.3a 0.70.2a 0. 57
Ae. atlanticus',2,3 306.264.0a 232.031.4a 227.557.8a 198.740.2a 251.251.2a 269.764.9a 0.34
Ae. canadensis 1.2l. Oa 0.20.2a 0.20.2a 0.50.5a 0.00.Oa 0.00.03 0.37
Ae. infirmatus2,3 60.215.la 33.85.2ab 27.34.8b 31.85.4b 34.7t9.Oab 35.27.8ab 0.01
An. crucians s.l. 6.21.2ab 5.72.1ab 9.22.4ab 5.31.4ab 0.80.4b 13.84.8a 0.03
An. quadrimaculatus s.l. 0.oo.o 0.00.0 0.00.0 0.010.0 0.00.0 0.30.3
Cx. nigripalpus2'3 5.72.Oa 8.74.4a 4.71.0a 6.01.6a 3.01.3a 7.02.0a 0.19
Cx. (Melanoconion) spp. 1.50.Oa 3.31.8a 2.81.la 4.31.2a 2.31.la 3.21.3a 0.65
Cx. quinquifaciatus 0.00.0 0.20.2 0.30.3 0.30.3 0.00.0 0.50.3
Ps. ciliata2 1.2+0.7a 1.7+0.8a 2.81.5a 1.20.7a 2.71.1a 2.31.2a 0.54
Ps. columbiae 0.70.5 0.00.0 0.010.0 0.00.0 0.20.2 0.00.0
Ps.ferox2'3 26. 04.9a 26.05.8a 18.86.5a 17.33.6a 28.012.Oa 27.54.9a 0.47
Ps. howardii 0.30.3a 0.20.2a 0.30.2a 0.20.2a 0.00.Oa 0.20.2a 0.88
Cq. perturbans 1.2l.Oa 1.31.1a 1.50.5a 1.00.5a 4.32.8a 1.20.7a 0.51
Cs. melanura 9.23.8ab 18.75.7ab 16.75.6ab 24.35.2a 4.72.2b 20.86.3ab 0.03
Ur. sapphirina 1.20.6ab 3.02.6ab 1.00.4b 1.50.8ab 0.00.Ob 8.73.9a 0.04
Tx. rutilus 0.00.0 0.50.3 0.20.2 0.20.2 0.20.2 0.00.0
Total No. Mosquitoes2 940.2104.Oa 952.0163. la 725.0113.6a 793.7119.1a 1157.3190.8a 759.7122.1a 0.08

1Adults could not be distinguished from Ae. tormenter.
2Significant day effect (p < 0.05).
3Significant position effect (p < 0.05).














Table 4-2. CDC Trap counts using colored LEDs or incandescent light only (Means SEM). Means within each row having the same letter are
not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). n = 6 nights.


Species Red Orange Yellow Green No Light Incandescent P-Value
Ae. dupreei 2,3 34.510.8a 33.313.7a 26.57.2a 48.224.2a 33.28.6a 32.7i 6a 0.51
Ae. fulvus pallens 0.20.2 0. 00.0 0.00.0 0. 00.0 0. 00. 03 00.0
Ae. atlanticus2',3 8.31.3abc 15.3+6.9ab 7.22.2bc 9.83.6abc 4.81.2c 17.55.1 a 0.009
Ae. canadensis 0.20.2 0.210.2 0. 00.0 0.00.0 0.00.0 0.00.0
Ae. infirmatus2 1.70.7a 4.812.9a 2. 71.Oa 2.710.8a 0.30.2a 4.31.3a 0.1
An. crucians s.1. 2 0. 80.5b 0. 70.3b 0.70.4b 1.311. b 0.00.Ob 4.51. 4a 0. 004
An. quadrimaculatuss.l. 00.0 0.00. 0 0.00. 0 000.0 0. 00.0 0.20.2
Cs. melanura o. 8+0.5a 2.0+1.3a 1.30.9a 2.5. 0a 00. Oa 3.50.8a 0.13
Cx. nigripalpus 0.30.2a 0.70.7a 0. 20.2a 0. 30. 2a 0. 20.2a 0. 80.4a 0.68
Cx. (Melanoconion)spp. 0.20.2a 0.20.2a 0.70.3a 0.20.2a 0.00.a 0.70.3a 0.25
Ps. ciliata 0. 00.0 0.0 .. 010. 0 0. 20.2 0. 00.0 0. 00.0
Ps. ferox3 0. 20.2a 0.20.2a 0.00.0a 0.20.2a 0.210.2a 0.00.Oa 0.72
Ur. sapphirina 0.70.3b 0.80.3b 0.710.3b 0. 80.3b 0.00.Ob 4.5I. Oa 0.001
Total No. MosquitoesZ3 47.811.8a 58.223.4a 39.89.2a 66.2230.5a 38.719.0a 68.716.1a 0.24
1Adults could not be distinguished from Ae. tormenter.

2Significant day effect (p < 0.05).
3Significant position effect (p < 0.05).












Table 4-3. CDC Trap counts using colored LEDs or incandescent light with 200 ml/min C02 (Means SEM). Means within
each row having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). n = 8 nights.

Species IR Red Orange Yellow Green Blue

Diachorusferrugatus2.3 19.64. 5a 16.93.2a 18.14.1a 15.52.7a 21.55.0a 25.84.4a
Corethrellaspp.3 2.60. 8c 21.114.2c 45.09.9c 43.57.3c 54.112.3c 119.328.9b
Ae. dupreei 23 1.00.4a 0.610.3a 0.60.5a 0.60.5a 0.60.3a 0.30.2a
Ae.fulvuspallens 0. 40.2a 0. 40.2a 0. 60.3a 0.30.2a 0. 30.2a 0.50.4a
Ae. atlanticus'2,3 13.13.la 14.13.4a 15.33.9a 12.64.1a 17.55.5a 11.63.4a
Ae. canadensis 1.10.4a 0.80.3a 0.8i0.4a 0.00.Oa 0.40.2a 0.40.4a
Ae.infirmatus2 2.90.4a 5.30.7a 5.0fl1. 8a 3.91.la 5.51.7a 3.50.7a
An. crucians s.l. 2.3 2.00.4d 6.1+1.7cd 7.81.9bc 8.11.7bc 5.90.9cd 12.13.3b
An. quadrimaculatus s. o.10.1 0.0o.0 0.10.1 0.10.1 0. o0. 0 o. 00. o
Cq. perturbans 1.5i0.6a 1.50.4a 0. 90.3a 1.3+0.3a 1.80.5a 2.30.8a
Cs. melanura 2.4i0.8d 9.42.3cd 25.14.2ab 20.62.7cb 25.92.7ab 18.52.9bc
Cx. nigripalpus 2.80.25b 4.81.lab 6.5il.lab 6.0l.lab 8.01.2a 7.8i1.7ab
Cx. (Melanoconion) spp.2 8.51. 3a 11.52.0a 10.83. Oa 11.02.8a 11.42.6a 17.42.9a
Cx. salinanrus 0. 30.2 0.30.2 0.40.3 0.40.2 0.5+0.3 0.50.3
Ma. dyari 0. 00. o o0. o00.0 0. 00. 0 0.10.1 0.10.1 0. 00. 0
Ps. ciliata 0. 00. o 0. 00.0 0.10.1 0. 0. 0 0. 00. 0 0.10.1
Ps. columbiae2 0.40.3ab 0. 00. Ob 1.00.8ab 0.50.4ab 0. 90.5ab 1.91.la
Ps. ferox2 4.91.3a 4.61.3a 4.31.6a 4.61.8a 5.62. la 4.00.8a
Ur. sapphirina2.3 2.81.3b 13.43.3b 25.56. lb 22.34.1b 25.96.2b 53. 913.4a
Total No. Mosquitoes2'3 44.05.1e 72.69.5cde 104.8l1.7bc 92.311.4bcd 110.1110.9bc 134.620.3b

'Adults could not be distinguished from Ae. tormenter.
ZSignificant day effet (p < 0.05).
aSignificant position effect ( < 0.05).












Table 4-3 --continued.


Species No Light Incandescent P-Value

Diachlorusferrugatus 23 20.95.3a 15.55.2a 0.26 Corethrella spp. 6.912.Oc 211.930.la 0.002 Ae. dupreei23 0. 30.2a 0.40.2a 0.52 Ae. fulvus pallens 0. 60. 3a 0. 4 0.2 0.89 Ae. atlanticus U3 15.85.7a 14.95. Oa 0.43 Ae. canadensis 0.6io0.4a I. 010.4a 0.38 Ae. infirmatus2 3.441.la 7.0+3.5a 0.42 An. crucians s. 2,3 2.0i0.5d 20.1I2. 9a 0.0001 An. quadrimaculatus s.l. 0. o0. 0 30.2 Cq. perturbans 2.4+0.7a 1.60.7a 0.58 Cs. melanura 4.3i.4d 33.614.7a 0. 0001 Cx. nigripalpus 3.60.7ab 7.1+1.6ab 0.02 Cx. (Melanoconion) spp.2 11.44.Oa 20.07.6a 0.07 Cx. salinarius 0. 1 0. 1 0. 510 2 Ma. dyari 0. 0o0.0 0.o00. Ps. ciliata 0. otO.0 0. 0o0. Ps. columbiae2 0. 50.4ab 0.40.4ab 0.08 Ps.ferox2 3.51.4a 6.93.1a 0.47 Ur. sapphirina2.3 2.4i1.1b 71.516.7a 0.0001 Total No. MosquitoesZ3 50.8i0.4ed 185.626.8a 0.001

'Adults could not be distinguished from Ae. tormenter.
-2Significant day effect (p < 0.05).
'Significant position effect (p < 0.05).








69








































Figure 4-1. Modified CDC-type trap equipped with LED.











30
a M Red E Orange a
a2? Yellow
25 ol Green 5 No Light
] White ab
" a Sa a ab 20
ab a a aa


0 15
a)ab aaa
C 1 ab0 aa 1) i~a






5
a








0
Ae. dupreei Ae. atlanticus Ae. infirmatus Ps. ferox Cs. melanura



Figure 4-2. Relative percent composition of mosquito species captured in C02 baited CDC-type traps using colored LEDs or
incandescent light only. Means within each species group having the same letter are not significantly different
(Ryan-Einot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 6 nights.










70


60
a


S50

=3 a
4.
c40
0
a


Sa a a
Saaba a
aa
ab
20 -a a a a 20


10 --
b b


An. crucians Cx. nigripalpus Cx.(Melanoconion) Ps. ciliata Cq. perturbans Ur. sapphirina

Figure 4-3. Relative percent composition of mosquito species captured in C02 baited CDC-type traps using colored LEDs or
incandescent light only. Means within each species group having the same letter are not significantly different
(Ryan-Einot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 6 nights.











40

a
M Red
M Orange
0 Yellow
a 0 Green 30 N No Light "a [ White Q) a
a a a ab
a'


n 20



Cl)

10 -





0
Ae. dupreei Ae. atlanticus Ae. infirmatus Ps. ferox Cs. melanura
Figure 4-4. Relative percent composition of mosquitoes captured in CDC-type traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (Ryan-Einot-GabrielWelsh Multiple Range Test). alpha = 0.05, n = 6 nights.










70

60

50

m 40
43 0 a

5 30 0)
o 20
a a

10

0
An. crucians Cx. nigripalpus Cx. (Melanoconion) Ur. sapphirina

Figure 4-5. Relative percent composition of mosquitoes captured in CDC-type traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (Ryan-Einot-GabrielWelsh Multiple Range Test). alpha = 0.05, n = 6 nights.











35
a
a El IR 30 M Red
IM Orange
D Yellow
D Green
-o 25 a a M Blue SU- No Light El White

a 20
Sab ab b
0 a

S 15
a)a b b a
-bc
10 c b a
ad b
a

-d"







Ae. atlanticus Cs. melanura An. crucians Ur. sapphirina Cx. (Melanoconion)


Figure 4-6. Relative percent composition of mosquitoes captured in C02 baited CDC-type traps using colored LEDs or
incandescent light only. Means within each species group having the same letter are not significantly different
(Ryan-Einot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 8 nights.














30


-0 25

CL a S20
2 ab a
a a aa
15
aa
"a at
Ua a a a
a -a sab




5

10

Ps. columbiae Ps. ferox Cq. perturbans Ae. infirmatus Cx. nigripalpus

Figure 4-7. Relative percent composition of mosquitoes captured in C02 baited CDC-type traps using colored LEDs or
incandescent light only. Means within each species group having the same letter are not significantly different
(Ryan-Einot-Gabriel-Welsh Multiple Range Test). alpha = 0.05, n = 8 nights.













CHAPTER 5
EVALUATION OF BLUE AND GREEN LIGHT EMITTING DIODES AS
ATTRACTANTS FOR WOODLAND MOSQUITOES AND OTHER DIPTERA IN NORTH CENTRAL FLORIDA


Introduction


Field trials by Burkett et al. (1998a) evaluated capture numbers and species

composition for light traps equipped with different colored light emitting diodes (LEDs) and compared them to those using standard incandescent bulb typically used in CDC-type traps. Results of these trials showed blue (450 + 50 nm) and green (567 50 nm) "superbright" LEDs attracted nearly as many mosquitoes (both in numbers and species composition) as the standard bulb. The LEDs used in those trials were oriented "up" to reflect the LED "light" off the aluminum trap lid. The number of mosquitoes captured in light-baited traps is directly proportional to the intensity of the light (Barr et al. 1963). Our objective was to determine if trap efficiency could be improved by orienting the LED's away from the trap in a 3600 pattern rather than reflecting the light off the trap lid. Orienting the LEDs away from the trap could increase the visible radius and thus allow the traps to be detected from a greater distance.

Aedes aegypti (Muir et al. 1992, Snow 1971), and other Diptera (Agee and Patterson 1983, Allan and Stoffolano 1986, Smith 1986) have bimodal spectral sensitivities with slight variations in X max occurring in the green and blue areas of the 76








77

electromagnetic spectrum. Not all mosquitoes are attracted to light. Likewise, not all species are attracted to the same wavelengths. Much of the research on dipteran visual attractant research is based on the response of diurnal species to reflected light (Brett 1938, Bracken et al. 1962, O'Gower 1963, Granger 1970, Bradbury and Bennett 1974, Browne and Bennett 1980, 1981, Allan and Stoffolano 1986). Visual attractant research based on transmitted light is less common (Headlee 1937, Williams et al. 1955, Breyev 1963, Bargren and Nibley 1956, Gjullin et al. 1973, Wilton and Fay 1972, Browne and Bennett 1981). Few reports detail the response of individual species and often focus on species easily reared in the laboratory, but not commonly captured in light-baited traps. Several colors (100 nm width) of highly efficient, low cost, "super bright" LEDs have recently been developed. LEDs are an improvement over standard incandescent bulbs because they produce greater light intensity and require significantly lower amounts of energy (ca. 0.125 ma/hr vs. 150 ma/hr). In this study, we evaluated the relationship between orientation of the LED on a light trap and capture numbers and species composition for woodland mosquitoes and other dipteran species.


Methods and Materials


Field trials conducted 1800 to 0800 hrs from August 9-14, 1997 in a 6 X 6 Latin square design evaluated light-trap captures of common woodland haematophageous Diptera including mosquitoes, Corethrella spp. (Diptera: Chaoboridae), Diachlorus ferrugatus Osten Sacke n (Diptera: Tabanidae), and phlebotomine sand flies (Diptera: Psychodidae). Trials were conducted in a ca. 1 hectare cypress swamp surrounded by








78

pine flat woods located north of Gainesville, FL. Light, day, and position effects were evaluated using a three-way ANOVA (SAS Institute, 1995). Multiple comparisons were made using the Ryan-Einot-Gabriel-Welsh multiple range test (alpha = 0.05). The trap and motor assembly remained stationary, but the lights were changed nightly so each would occupy every position during the six-day period. After each trap night, trap collections were separated, weighed, identified and counted.

Standard CDC-type traps (John W. Hock Company, model 512, Gainesville FL) were modified by replacing the standard bulbs with the LEDs. LEDs oriented "out" in a 3600 pattern were compared to those oriented "up" and reflecting off the aluminum trap lid. LEDs were soldered in series to insulated 22 gauge wire shaped to form a 40 cm diameter circle (Fig. 5) and were secured with rubber bands to the 1/4" mesh screen atop the trap lid assemblly. The green or blue LEDs oriented up or out were compared to controls consisting of no light (fan only) and a standard incandescent bulb (John W. Hock Company, CM-47, 6.3 V, 520 millicandela [mcd], Gainesville, FL). The LEDs used were green (567 50 nm, 2400 mcd, 8 [Toshiba Tosbright, Martech Optoelectronics, Latham NY ]) and blue (450 50 nm, 800 mcd, 22' [Panasonic,' Digikey Corp, Thief River Falls, MN]). Power for the green and blue LEDs were supplied by 2 and 3 alkaline D cell batteries at (2.8 0.2 volts, 18 2 ma) and (4.3 0.2, 20 + 3 ma) respectively. A 10 ohm resistor was soldered in series to prevent over-driving the LED.

Trials used 9 kg compressed-gas (carbon dioxide) cylinders equipped with double-stage (Victor Equipment Company, model VTS453B-320, Denton TX) and microregulators (Nupro, Series M, Wiloughby OH) to maintain a constant gas flow of








79

200 ml per minute. Carbon dioxide was delivered to the trap through a 3 m piece of 8 mm transparent plastic tubing secured with a rubber band so the top of the tubing was even with the top of the trap opening. Gas flow was checked each morning and evening using an in-line flowmeter (Gilmont Instruments, no. 12, Great Neck NY). Mosquitoes attracted near the trap intake were drawn in by the trap fan, blown through a screen funnel and into a quart polypropylene jar containing a 6 X 6 cm piece of dichlorvos impregnated vinyl strip used as a killing agent. Six volt, 10 ampHrs rechargeable gel cell batteries (Powersonic Corp, San Diego CA) were used to run the fan motor and standard incandescent light. Traps were hung 165 cm above ground level and placed every 30 meters around the outer perimeter of the swamp. Due to the thick vegetation, none of the traps were visible to each other. The trapping period followed a 1.7, 2.3, 1.2, 2.4, and 1.9 cm rainfall on July 28, 29 and August 1, 2 and 7 respectively. The following species groups, Cx. (Melanoconion) spp. Theobald, Anopheles quadrimaculatus Say, An. crucians Wiedemann and Aedes atlanticus/tormenter Dyar and Knab were pooled, as these taxa could not be distinguished with confidence.


Results


During the six trap-nights, 102,917 mosquitoes, 94 tabanids (Diachlorus ferrugatus Osten Sacken), 2,841 chaoborids (Corethrella spp.), and 26 Lutzomyia shannoni (Dyar) were collected respectively. Proportions of trap collections captured at the different light color/orientation combinations for the most common species are shown in Figures 5-2 and 5-3. Means, standard errors, p-values, and significant differences for








80

the common species are shown in Table 5-1. As noted in the tables, there were significant trap-position and day effects for some species. No significant differences were observed for the total number of mosquitoes captured at the light color/orientation combinations (p = 0.21). However, there were significant differences in capture numbers for individual species including An. crucians s.l. (p=0.0001), Culiseta melanura (Coquillett) (p=0.02), Cx. (Melanoconion) spp. (p=0.03), and Uranotaenia sapphirina (Osten Sacken) (p=0.001). Anopheles crucians and Ur. sapphirina were captured in the largest numbers at the incandescent bulb and the blue LEDs oriented out. Conversely, Cs. melanura was captured in greatest numbers in the traps baited with the incandescent bulb, and blue and green LEDs oriented up. Blue LEDs "out" were most attractive for Cx. (melanoconion) spp. Although female Lutzomyia shannoni (Dyar), were collected, no significant color differences were observed given the small sample size. The tabanid, Diachlorusferrugatus Osten Sacken, also showed no significant color preference. Chaoborids (Corethrella spp.), however showed significant color preferences (p=0.0009) for the incandescent, and both blue orientations. No colors were found to be repellent to mosquitoes when compared to the no-light controls. The mosquito species composition attracted to the incandescent light trap agree with those found by Mann (1993) and Burkett et al. (1998a).

The weights of non-target insects collected in the different CDC light trap

combinations showed the incandescent bulb to have significantly more (p=0.001) than any of the LED color/orientations.








81

Discussion


Few field studies have scientifically evaluated mosquito trap collections based on using different colors of transmitted light of known wavelength and intensity. This is surprising, considering the primary means of evaluating mosquitoes' relative abundance, presence or absence, and effectiveness of control measures is through the use of lightbaited traps. Most studies use pooled species; few report results of individual species. Significant color preferences for some species were demonstrated from trap collections using LED baited traps (Burkett et al. 1998a). In general, traps with LEDs emitting primarily green and blues wavelengths captured the most mosquitoes. In this study, the attraction to greens and blues were further evaluated by changing the orientation of the LEDs in the traps.

Collections of most of the common representative species such as Ae. dupreei, Ae. atlanticus/tormentor, Ae. infirmatus, Ps. ferox, Cx. nigripalpus, Ps. ciliata, and Cq. perturbans did not differ significantly between any of the color/orientation combinations tested. Only Cs. melanura, An. crucians, Ur. sapphirina and Cx. (Melanoconion) spp. were collected in significantly greater numbers to the different trap LED color/orientations when compared to the "no light"/CO2 only controls. Allan et al. (1987) stated that while crepuscular and nocturnal biting flies are unlikely to have welldeveloped color vision, their abilities to detect differences in intensity contrast are likely to be well-developed. Even though there were relatively large differences in intensity between the incandescent (520 mcd), blue (700 mcd) and green (2400 mcd) LEDs used in








82

the traps, there were significant differences in trap collection for only 4 species. This may be due to the heavily vegetated and confined habitat of the cypress swamp. Greater differences in trap collections due to differences in intensity may result in more open habitats.

We recommend additional work aimed at species from different habitats and in different geographical locations. Furthermore, the attraction of phlebotomine sand flies deserves additional investigation. Light-emitting diodes require significantly lower amounts of energy (ca. 1 ma/8 hrs) than incandescent bulbs (See Burkett et al.1998a) saving both battery life and expense during use. For convenient use, LEDs could be soldered directly into the existing trap circuitry (in series) using a 100 and 200 ohm resistors for blue and green LEDs respectively. Future trap designs should incorporate a mixture of green and blue diodes oriented both "out and up." The use of "super-bright" LEDs warrants serious consideration as a replacement for standard incandescent bulbs used in light-baited traps. These results have potential for use in population dynamics studies or for enhancing the attractivity of certain species.









Table 5-1. CDC Trap counts using colored LEDs or incandescent light in different orientations with 200 ml/min CO2 (Means SEM). Means
within each row having the same letter not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test). n = 6 nights.

Species Blue Out Blue Up Green Out Green Up No Light Incandescent P-Value Sample Wt (mg)2 878.5t200.2a 797.592.1a 767.2113.1a 856.0140.3a 856.3278.5a 1268.0398.5a 0.35
Trash Wt (mg) 48.87.5b 28.05.9bc 17.82.9c 13.01.0c 6.52.3c 97.213.6a 0.001
Ae. dupreei2 74.521.7a 85.210.4a 62.1 1i2.5a 116.838.1a 109.516.8a 170.885.la 0.22
Ae.fulvuspallens 0.70.3 0.50.3 0.80.5 0.50.2 0.80.5 0.30.2
Ae. atlanicus'.2 626.0127.7a 649.5103.0a 651.5104.3a 637.7i86.9a 786.7211.5a 973.8283.0a 0.39
Ae. triseriatus 0.00.0 0.00.0 0.00.0 0.30.3 0.70.4 0.00.0
Ae. canadensis 0.30.3 0.00.0 0.00.0 0.20.2 0.00.0 0.00.0
Ae. infirmatus2 117.841.8a 117.721.la 100.0f18.3a 138.334.4a 136.0141.4a 177.864.8a 0.57
Ae. mitchellae 0.20.2 0.00.0 0.20.2 0.00.0 0.70.5 0.00.0
An. crucians sI.2 149.322.9ab 97.711.7bcd 107.723.2bc 88.7-3.3cd 43.36.6d 187.321.Oa 0.0001
An. quadrimaculatus s.l. 7.71.9a 4.02.Oa 5.32.8a 4.22.5a 1.00.4a 8.33.la 0.26
An. puntipennis 0.00.0 0.20.2 0.00.0 0.30.3 0.00.0 0.00.0
Cx. nigripalpus 2 78.511.8a 76.323.4a 66.716.2a 75.2110.9a 49.311.2a 67.113.4a 0.35
Cx. (felanoconion) spp.3 51.315.7a 42.56.6ab 40.26.lab 36.82.9ab 19.53.3b 25.85.4ab 0.03
Cx. quinquifaciatus 7.34.1 4.32.2 1.50.4 4.01.8 1.51.0 1.70.9
Cx. territans 2.81.0 2.71.6 2.51.0 1.30.7 0.80.5 1.30.9
Ps. ciliata2'3 17.25.0a 16.26.5a 23.76.3a 22.77.4a 29.08.la 13.55.3a 0.11
Ps. columbiae 1.31.3 0.80.8 0. 00.0 0.50.3 0.00.0 2.32.3
Ps.ferox2 140.039.0a 138.736.5a 140.3t36.6a 181.241.7a 181.061.5a 266.21105.9a 0.43
Ps. howardii 2.52.0 2.51.3 1.00.4 2.30.6 1.20.7 1.80.5
Cq. perturbans2 164.6a 10.52.7a 14.05.2a 5.32.5a 4.72.la 9.03.6a 0.1
Cs. melanura3 54.517.Oab 74.528.0a 49.512.2ab 83.023.8a 9.84.3b 77.714.9a 0.02
Ma. dyari 0.00.0 0.00.0 0.00.0 0.20.2 0.30.2 0.20.2
Ur. sapphirina2 100 023.0ab 72.58.6bc 26.24.6dc 25.06.8dc 2.5+1.5d 141.821.4a 0.001
Tx. rutilus 0.30.3 0.090.0 0.50.3 0.50.5 0.00.0 0.2020
Total No. Mosquitoes2 1448.3285.8a 1396.2188.1a 1293.7194.7a 1425.0153.4a 1378.3331.2a 2127.7568.1a 0.21
Diachlorusferrugatus2 4. 01.6a 2.3. 5a 2.30.7a 1.70.6a 3.31.la 4.21.7a 0.21
Corethrellaspp. 103.816.3abc 123.327.lab 19.54.2bc 33.87.7bc 3.31.1c 189.762.2a 0.0009
Lutzomyia shannoni 1.00.5a 0.70.3a 0.30.2a 1.30.3a 0.30.2a 0.70.5a 0.32
Adults could not be distinguished fromrnAe. tormenter.
2Sigificant day effect (p < 0.05).
3Sig"ificant position effect (p < 0.05).








84


a. b.





Battery/trap
Battery rap electronics
elecdroics


Green or Blue LED '







Fig. 5-1. Wiring assemble representation for LED orientations: a) Top view of LED
assembly oriented 360 out and b) Top view of LED assemblely oriented "up." For green LEDs: 3 V with separate battery pack or if wired to trap
electronics use 200 ohm resistor. For blue LEDs: 4.5 V with separate
battery pack or if wired to trap electronics use 100 ohm resistor.




Full Text

PAGE 1

LIGHT COLOR ATTRACTION AND DIETARY SUGAR COMPOSITION FOR SEVERAL MOSQUITO (DIPTERA: CULICIDAE) SPECIES FOUND IN NORTH CENTRAL FLORIDA By DOUGLAS ARTHUR BURKETT 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 1998 i

PAGE 2

Dedicated to Laura, Alexis, Sara, Betty and Bob

PAGE 3

ACKNOWLEDGMENTS Many and most sincere thanks to the Air Force Institute of Technology whose Civilian Institution's program sent me back to school for what I consider the best educational experience ever. Next, I would like to thank the group of scientists and staff at the U.S. Department of Agriculture, Animal Research Service and the Department of Entomology and Nematology for supplying me with an office, lab space, vehicles, and administrative support. They treated me as one of their own. I wish to express my sincere gratitude to the members of my committee. To Dr. Dan Kline who always welcomed me into his office, and was never too busy to help out, thanks for generously sharing your resources, space and equipment. These same sentiments are for Dr. Jerry Butler as well, who served as a great mentor and friend. This dissertation would not have been possible without the support and labor provided by him and his medical entomology staff. A special thanks to Dr. Gene Gerberg from whose extraordinarily long entomological career, experiences, and library I have greatly benefited. Thanks to Dr. Sandy Allan for finding a way to be on my committee. Her expertise in insect vision and excellent editorial skills have been very helpful. A number of other scientists and staff members deserve mention. Dr. Dave Carlson, I learned a hell of lot from you. Thank you for your support, equipment, enthusiasm, exceptional writing skills and your friendship. I also wish to thank the other iii

PAGE 4

USDA chemists, Dr. Bob VanderMeer and Dr. Uli Bernier, for sharing their equipment and expertise. Thank you Dr. John Reinert for your help with mosquito identifications and relevant reprints and to Dr. Pete Perkins for his sand fly identifications. Thanks Dr. Don Barnard for allowing me to raid your library and giving me free reign in the CMAVE. Charles Schmidt, of Environmental Engineering, I greatly appreciate your warm welcome into your lab and admire your love of analytical chemistry. You and your staff treated me as one of your own and donated your equipment and supplies. Henry McKeithen, thank you for your support, computer help, and braving the swamps with me. Thanks to Genie White for sharing SAS and wild Culex with me. Eric Daniels, you have been a tremendous associate. Your computer help, support and friendship will never be forgotten. Thank you Dr. Dan Haile, who was initially one of my committee members but transferred to Panama before my degree completion. Our aerial spray adventures were great. Thanks to Drs. Don Hall and Grover Smart for helping me get through my qualifying exams. I also wish to acknowledge Erin Rountree and Margaret Martin for their administrative support and Johnny Jackson for the large supply of mosquitoes from the insectary. From the Department of Entomology and Nematolgy, I wish to thank Dr. Butler's crew including Diana Simon, Tim Robson, Haze Brown, and Tamara Landau. Thanks to Drs. Frank Slansky and Harvey Cromroy, for your excellent reviews of my manuscripts. Debbie Hall, you were amazing in the front office thanks. As my sole military associates while attending graduate school, I acknowledge the DoD/USDA Liaison officers, Lt. Col. Terry Carpenter and Captain Herb Bolton for their encouragement along the way. iv

PAGE 5

I would not have gone on to complete this degree if not for the urging of Dr. Terry Biery, my supervisor for 5 years at the Aerial Spray mission. Thanks Terry. I also wish to acknowledge Dr. Alberto Broce, my Master's program advisor at Kansas State University, who further enhanced my interest in entomology and pursuing additional graduate education. Lastly, a huge thanks to my parents who provided the initial education funding, guidance and support. I owe the utmost gratitude to my wife, Laura, whose love, enthusiasm and understanding have been instrumental in completing all these years of graduate school and Air Force directed moves. v

PAGE 6

TABLE OF CONTENTS ACKNOWLEDGMENTS "i LIST OF TABLES ix LIST OF FIGURES x ABSTRACT xv 1. MOSQUITO SUGAR FEEDING AND VISUAL ATTRACTANTS REVIEW 1 Introduction To Sugar Feeding 1 Carbohydrate and Amino Acid Sources 3 The Need for Dietary Sugars 5 Floral Nectaries 7 Extrafloral Nectaries 11 Honeydew 13 Other Sugar Sources 16 Diel Sugar Feeding Activities 17 Morphology and Physiology of Mosquito Alimentary Tract 18 Crop Enzymatic Activity 21 Introduction To Mosquito Visual Attractants Review 22 Reflected Light 23 Transmitted Light 24 2. ANALYSIS OF SUGAR MEAL COMPOSITION OF WILD MOSQUITOES BY GAS CHROMATOGRAPHY 26 Introduction 26 Methods and Materials 28 Specimen Preparation 28 Sample Preparation for GC Analysis 29 Crop Sucrose and Melezitose Hydrolysis 30 Results and Discussion 31 Sucrose and Melezitose Fed Mosquitoes 32 3. SUGAR MEAL COMPOSITION OF FIVE NORTH CENTRAL FLORIDA MOSQUITO SPECIES AS DETERMINED BY GAS CHROMATOGRAPHY 41 Introduction 41 vi

PAGE 7

Methods and Materials 43 Specimen Preparation 43 Sample Preparation for GC Analysis 44 Standards 45 Results and Discussion 45 Ecological Significance of Honeydew in Mosquito Diet 46 Sugar Feeding Field Observations 48 4. EVALUATION OF COLORED LIGHT EMITTING DIODES AS ATTRACTANTS FOR WOODLAND MOSQUITOES AND OTHER DIPTERA IN NORTH CENTRAL FLORIDA 56 Introduction 56 Methods and Materials 57 Six Colors With and Without Carbon Dioxide (Trials 1 and 2) 57 Eight Colors With Carbon Dioxide (Trial 3) 59 Results 59 Six Colors With and Without Carbon Dioxide (Trials 1 and 2) 59 Eight Colors With Carbon Dioxide (Trial 3) 60 Discussion 61 5. EVALUATION OF BLUE AND GREEN LIGHT EMITTING DIODES AS ATTRACTANTS FOR WOODLAND MOSQUITOES AND OTHER DIPTERA IN NORTH CENTRAL FLORIDA 76 Introduction 76 Methods and Materials 77 Results 79 Discussion 81 6. LABORATORY EVALUATION OF COLORED LIGHT AS AN ATTRACTANT FOR FEMALE AEDES AEGYPTI, AEDES ALBOPICTUS, ANOPHELES QUADRIMACULATUS AND CULEX NIGRIPALPUS 87 Introduction 87 Methods and Materials 87 Visualometer and Data Collection 89 Artificial Host 90 Light Source and Filters 91 Mosquito Species 91 Statistical Analysis 92 Results 92 Aedes aegypti 93 Aedes albopictus 93 Anopheles quadrimaculatus 94 Culex nigripalpus 94 vii

PAGE 8

Discussion 95 7. MOSQUITO HOST/RESOURCE FINDING: THE IMPORTANCE OF SUGAR FEEDING AND ATTRACTION TO ARTIFICIAL LIGHT 110 Introduction 110 Sugar Feeding 113 Attraction to Artificial Light 119 Summary 123 REFERENCES CITED 125 BIOGRAPHICAL SKETCH 140 viii

PAGE 9

LIST OF TABLES Table Page 31 Summary of percent wild mosquitoes containing identifiable sugars using gas chromatography 50 41 CDC Trap counts using colored LEDs or incandescent light with 200 ml/min C02 (Means SEM). Means within each row having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), n = 6 nights 65 4-2. CDC Trap counts using colored LEDs or incandescent light only (Means SEM). Means within each row having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), n = 6 nights. ... 66 43. CDC Trap counts using colored LEDs or incandescent light with 200 ml/min C02 (Means SEM). Means within each row having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), n = 8 nights 67 51 CDC Trap counts using colored LEDs or incandescent light in different orientations with 200 ml/min C02 (Means SEM). Means within each row having the same letter not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), n = 6 nights 83 ix

PAGE 10

LIST OF FIGURES Figure Page 21 Chromatogram of combined standards (ca. 0. 1 %) for common sugars associated with plants. Multiple peaks of the same sugar are anomeric forms 35 2-2. Representative chromatograms of crude crop extracts from two wild individuals of Culiseta melanura. Multiple peaks of the same sugar are anomeric forms 36 2-3. Representative chromatograms of crude crop extracts from two wild individuals of Anopheles quadhmaculatus. Multiple peaks of the same sugar are anomeric forms 37 2-4. Timed feeding trials for male and female Aedes albopictus showing the hydrolysis of melezitose and sucrose occurring in the crop (n = 2 to 5) 38 2-5. Representative chromatogram showing hydrolysis of sucrose into fructose and glucose 2 hours following ingestion in Aedes albopictus. Multiple peaks of the same sugar are anomeric forms 39 26. Representative chromatogram showing hydrolysis of melezitose 2 hours following ingestion in Aedes albopictus 40 31 Representation of crop sugar distributions and proportions of wild Anopheles quadhmaculatus (n = 33). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and turanose have identical retention times 51 3-2. Representation of crop sugar distributions and proportions of wild Coquilletidia perturbans (n = 20). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and turanose have identical retention times 52 X

PAGE 11

Figure Page 3-3. Representation of crop sugar distributions and proportions of wild Culiseta melanura (n = 99). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and turanose have identical retention times 53 3-4. Representation of crop sugar distributions and proportions of wild d) Culex nigripalpus (n = 26). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and turanose have identical retention times 54 35. Representation of crop sugar distributions and proportions of wild Psorophora ferox (n = 14). The presence of the trisaccharides, erlose and melezitose, is evidence of honeydew feeding. Maltose and turanose have identical retention times 55 41 Modified CDC-type trap equipped with LED (shown with arrow) 69 4-2. Relative percent composition of mosquito species captured in C02 baited CDC-type traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (Ryan-Einot-GabrielWelsh Multiple Range Test), a = 0.05, n = 6 nights 70 4-3. Relative percent composition of mosquito species captured in C02 baited CDC-type traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), alpha = 0.05, n = 6 nights 71 4-4. Relative percent composition of mosquitoes captured in CDC-type traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (RyanEinot-Gabriel-Welsh Multiple Range Test), alpha = 0.05, n = 6 nights 72 4-5. Relative percent composition of mosquitoes captured in CDC-type traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (RyanEinot-Gabriel-Welsh Multiple Range Test), alpha = 0.05, n = 6 nights 73 xi

PAGE 12

Figure Page 4-6. Relative percent composition of mosquitoes captured in C02 baited CDCtype traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), alpha = 0.05, n = 8 nights 74 47. Relative percent composition of mosquitoes captured in C02 baited CDCtype traps using colored LEDs or incandescent light only. Means within each species group having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), alpha = 0.05, n = 8 nights 75 51 Wiring assemble representation for LED orientations: a) Top view of LED assembly oriented 360 out and b) Top view of LED assemblly oriented "up." For green LEDs: 3 V with separate battery pack or if wired to trap electronics use x ohm resistor. For blue LEDs: 4.5 V with separate battery pack or if wired to trap electronics use y ohm resistor 84 5-2. Relative percent composition of common mosquito species captured in C02 baited CDC-type traps using different orientations of green or blue LEDs or incandescent light. Means within each species group having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), alpha = 0.05, n = 6 nights 85 53. Relative percent composition of common mosquito species captured in C02 baited CDC-type traps using different orientations of green or blue LEDs or incandescent light. Means within each species group having the same letter are not significantly different (Ryan-Einot-Gabriel-Welsh Multiple Range Test), alpha = 0.05, n = 6 nights 86 61. Visualometer (a) Top view (b) Side view. (A) 430 x 5 cm dia. transparent plexiglass lid (B) Artificial host, (35 mm film canister lid) (C) 30 cm high plastic side piece with holes for plastic tubing (D) Aluminum base with holes for fiber optic cables (E) Support leg (F) Plexiglass filter support (G) 10 cm diameter Tygon tubing for incoming air/C02 (H) Tubing for exhaust (I) Assembly screw (J and K) Probe inserted into top of artificial host and bottom sensor fitted under artificial host (feeding mosquitoes complete circuit logged by computer), one sensor per artificial host (L) Interference (bandbass) filter (M) Neutral density filter (N) Fiber optic cable (attached to light source). Drawing not to scale 101 xii

PAGE 13

Figure Page 6-2. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Aedes aegypti on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0. 1 0, Duncan's Multiple Range Test) 1 02 6-3. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Aedes albopictus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple Range Test) 103 6-4. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Anopheles quadrimaculatus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple Range Test) 104 6-5. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Culex nigripalpus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple Range Test) 105 6-6. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Aedes aegypti on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0. 1 0, Duncan's Multiple Range Test) 1 06 6-7. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Aedes albopictus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0. 1 0, Duncan's Multiple Range Test) 107 6-8. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Anopheles quadrimaculatus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple Range Test) 108 xiii

PAGE 14

Figure Page 6-9. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Culex nigripalpus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple Range Test) 109 xiv

PAGE 15

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctorate of Philosophy LIGHT COLOR ATTRACTION AND DIETARY SUGAR COMPOSITION FOR SEVERAL MOSQUITO (DIPTERA: CULICIDAE) SPECIES FOUND IN NORTH CENTRAL FLORIDA By Douglas Arthur Burkett August 1998 Chair: Daniel L. Kline Cochair: Jerry F. Butler Major Department: Entomology and Nematology The behavioral response of mosquitoes to different wavelengths of light and an evaluation of sugar meals was studied for several species of mosquito found in north central Florida. Gas chromatography (GC) was used to analyze dietary sugars of Anopheles quadrimaculatus, Coquillettidia perturbans, Culex nigripalpus, Culiseta melanura and Psorophora ferox. GC was also used to determined whether carbohydrases are present in the mosquito diverticula. A wide range of sugars was found in wild mosquitoes including fructose, glucose, sucrose, maltose, turanose, melezitose, raffinose. erlose, arabinose, rhamnose, and several unknowns. Laboratory time course studies with Aedes albopictus showed rapid hydrolysis of sucrose within 2 hours of ingestion. xv

PAGE 16

Conversely, melezitose remained relatively unchanged after 8 hours. The frequency of wild mosquitoes containing sugars ranged from 10-1 1% in An. quadrimaculatus and Ps. ferox to 48% in Cq. perturbans. All species tested contained honeydew sugars including An. quadrimaculatus (55%), Cs. melanura (33%), Cx. nigripalpus (15%), Cq. perturbans (10%) and Ps. ferox (7%). Light trap capture numbers for woodland mosquitoes were evaluated using light emitting diodes (LEDs) of different colors. Analysis of data by species showed significant differences in color attraction (green and blue best) for^e. atlanticus,Ae. dupreei,Ae. infirmatus,An. crucians, Cs. melanura, Cx. nigripalpus, Ps. columbiae and Uranotaenia sapphirina. Attraction was also evaluated using different orientations (light reflected of off lid, or 360 radius) of green/blue LEDs and found significant capture number differences for only An. crucians, Cs. melanura, Cx. (Melanoconion spp.) and Ur. sapphirina. In the lab, a ten port visualometer was constructed to evaluate the feeding response of 4 mosquito species over various wavelengths of light. Color preferences were based on feeding durations obtained electronically over 4 hour intervals and numbers of fecal specks after a 1 6 hour exposure. Feeding duration and fecal speck data were collected for female Ae. aegypti, Ae. albopictus, An. quadrimaculatus and Cx. nigripalpus in the visualometer on an artificial host illuminated from below with light of equal intensities in 50 nm increments (700-350 nm). Significant color preferences based on feeding durations were detected iox Ae. albopictus, An. quadrimaculatus and Cx. nigripalpus. Fecal speck numbers significantly correlated with feeding durations for An. quadrimaculatus and Cx. nigripalpus. xv i

PAGE 17

CHAPTER 1 MOSQUITO SUGAR FEEDING AND VISUAL ATTRACTANTS REVIEW Introduction To Sugar Feeding An increased understanding of mosquito biology, improved surveillance and effective control practices are the main driving forces that has made Florida and other coastal states habitable. Blood loss, annoyance and their ability to transmit a variety of serious diseases have given mosquitoes dubious notoriety. Research on the sugar sources of mosquitoes can have great potential returns to biologists and mosquito control personnel. Determination of crop carbohydrate composition would allow researchers to make inferences about the identification of plants and plant-derived products being used to provide metabolic demands of wild mosquitoes. These inferences, however, are only possible if common plant-derived oligosaccharides remain unhydrolyzed and identifiable in the crop for a significant amount of time. A better understanding of these facets of mosquito ecology can be applied directly to the development of new attractants and improving control methods. It is currently unknown as to whether natural sugars are an unlimited and universally available resource, or if suitable sugars are scarce and their lack can adversely affect mosquito survival and reproduction. In general, mosquitoes obtain sugars from a wide range of sources (Gadawski and Smith 1992, Grimstad and DeFoliart 1974, 1

PAGE 18

2 Magnarelli 1977, 1979, 1983). Furthermore, not much is known about species-specific preferences for natural sugar hosts. Are mosquitoes generalists and opportunistic, obtaining sugar from sources that are readily available or do they use a narrow set of discrimination criteria to choose their natural sugar sources? Plant taxa differ in the proportion and to a lesser extent, the composition of sugars present in nectar. If sugar meals remain unchanged upon ingestion and storage in the crop, it may be possible to identify specific plants or at least groups of plants with similar properties based on the presence or absence and relative concentrations of sugars present in the mosquito crop. It is currently unknown how many mosquito species contain salivary or crop enzymes which hydrolyze sucrose and other common plant oligosaccharides into simpler molecules. As a precursor to determining sugar sources utilized by various wild mosquito species, one must determine if, how much, and how fast the plant-derived monodiand trisaccharides are broken down in the crop. If these ingested sugars remain unchanged in the crop for any length of time, inferences could be made as to their sugar source utilization of wild mosquitoes. Ultimately, we could determine if certain mosquito species preferentially utilize various plant families with certain nectar characteristics, or if they utilize sugary exudates from Homoptera (e.g. honeydew) or some other sugar source. For a given mosquito species, what proportion of the population feeds on nectars or other plant juices and what proportion feeds on honeydew?

PAGE 19

3 Carbohydrate and Amino Acid Sources Much of the evidence for mosquito nectar feeding comes from direct field observations rather than through indirect chemical qualitative analysis of crop contents. West and Jenkins (1951), Sandholm and Price (1962), Gadawski and Smith (1992), Breeland and Pickard (1961, 1967), and Magnarelli (1977) have observed some mosquito species feeding during the day. Most direct observations, however, have been made at night, dawn or dusk (Grimstad and DeFoliart 1974, Magnarelli 1983, Andersson and Jaenson 1987, Bowen 1992, Vargo and Foster 1984, Yee et al. 1992). Interestingly, not all common mosquito species have been observed feeding on sugar in the field. Bidlingmayer and Hem (1973) found evidence of fructose in the crops of Culiseta melanura (36%), Culex nigripalpus (17%), Psorophora ferox (20%) and Anopheles quadrimaculatus (15%). Likewise, Magnarelli (1978) obtained similar results for Cq. perturbans (57%). There have been no published accounts of field sugar feeding for Aedes albopictus, An. quadrimaculatus, or Cs. melanura and only a few for Cx. nigripalpus (see Nayar 1982). Regardless of species, most direct field observations of sugar feeding have occurred at ground level nectar sources on flowering herbs and low growing shrubs. These observations do not incorporate sugar feeding in trees or at the potentially numerous honeydew, extrafloral nectary or other plant exudates scattered up in the canopy and throughout the environment. As indirect evidence for "nectar" feeding, dozens of mosquito species have been chemically analyzed for the presence or absence of fructose. Fructose is a ubiquitous

PAGE 20

monosaccharide present in almost all plant sugars. Many authors (Andersson and Jaenson 1987, Edman et al. 1992, Van Handel et al. 1972, 1994, Bidlingmayer and Hem 1973, Magnarelli 1978, 1979, 1980, 1983, Reisen et al. 1986, Nasci and Edman 1984, Smith and Kurtz 1994) have used the cold anthrone test (Van Handel 1972) to qualitatively establish the presence or absence of fructose (or oligosaccharides containing fructose or other reducing sugars). According to Van Handel (1967), the cold anthrone reagent reacts with fructose, inulin, sucrose, melezitose, and raffinose. These sugars all contain the fructose moiety and give a positive or negative response based on a resulting color reaction. Unfortunately, the cold anthrone test does not provide information about the origin of crop sugars or the exact carbohydrate composition. Many other common mono-, diand trisaccharides are present in potential natural sugar meals, but few attempts have been made to categorize the individual sugars and relate the meals back to the original source. Although usually used as evidence for "nectar feeding," the cold anthrone test will also react positively for insects having fed on honeydew or other non-floral sugar sources. Few attempts have been made to determine the exact crop sugar concentration and composition for various Dipteran crops using modern chromatographic techniques. Thin Layer Chromatography (TLC) has been used by Magnarelli and Anderson (1977), Magnarelli (1980), and Van Handel (1984). High performance liquid chromatography (HPLC) was successfully used by MacVicker et al.( 1 990), to examine the contents of crops of five wild Italian sand fly species. Hoppe (1983), used HPTLC for determining the sugars found in horse flies. In spite of its common use in other systems, Moore et al.

PAGE 21

5 (1987), Chang et al. (1977), and Alexander (1988), were among the first (and few) to use modern and sensitive gas liquid chromatography (GC o r GLC) for investigating dipteran diets. The Need for Dietary Sugars With few exceptions, mosquitoes generally require dietary sugars for survival, longevity, host finding and reproductive success. In terms of survival, female Ae. aegypti, Cx. quinquefasciatus and Cx. tritaeniorhynchus lived much longer (Briegel and Kaiser 1973, Harada et al. 1976) and laid more eggs (Nayar and Sauerman 1975a) in the laboratory when provided with both sugar and blood meals than when given blood alone. When fed individual sugars, Galun and Fraenkel (1957) reported ihdXAe. aegypti survived the longest (at least 25 days) on glucose, fructose, sucrose, maltose, trehalose, melibiose, raffinose or melezitose. With the exception of melibiose, similar results were found for Cs. inornata (Salama 1 967). Other sugars reduce the longevity or have no effect on survival or fecundity (Nayar and Sauerman 1971a). The sugars enhancing survival are commonly found directly or indirectly with plants. Nayar and Sauerman (1975a) reported duration of mosquito survival to be directly related to the rate of disappearance of glucose which followed an exponential law of decline. Furthermore, the disappearance rate differed between mosquito species. Ae. aegypti and An. quadrimaculatus, for example, used glucose faster than Ae. sollicitans and Ae. taeniorhynchus under similar conditions.

PAGE 22

6 Some species or populations of some species may require minimal or no sugar to survive and reproduce. An Ae. aegypti population from Thailand was found to rarely, if ever take sugar meals, apparently obtaining all their sugar requirements from multiple blood meals (Day et al. 1994, Edman et al. 1992). Interestingly, this Ae. aegypti population produced as many eggs as those that took sugar meals. In another similar case, a Puerto Rican Ae. aegypti population contained only 5% of the mosquitoes positive for fructose (Van Handel et al. 1994). Whereas, in a Florida population, 50-75% of the sampled mosquitoes were positive for fructose. Some authors find sugar vital for stimulating ovarian development. For example, Klowden (1986), found that laboratory reared Ae. aegypti did not develop a batch of eggs without obtaining a sugar meal before a blood meal. Much remains to be learned about sugar feeding. A lack of sugar feeding by some species or populations of some species may be more common than portrayed in the literature. Indeed, many species have only been rarely observed sugar feeding in the wild and of those species examined, some contain only a very low percentage of sugars in their crops. Foster (1995) summarizes that some species have only 1-4% of the population containing fructose. In most mosquito species, however, greater than 40% contain fructose and a few species have more than 80%. In terms of host-finding, female sugar feeding mosquitoes may enhance their vectorial competence and improve their success of obtaining a blood meal. Walker and Edman (1985), reported that female Ae. triseriatus and Ae. aegypti that have had access to sugar sources were more persistent (thus more successful) in their attempts at blood feeding than were sugar-deprived females. However, in terms of frequency blood feeding

PAGE 23

7 by An. quadrimaculatus was not affected by the availability of sugar (Foster and Eischen 1987). Floral Nectaries The most documented source of carbohydrate for mosquitoes is floral nectaries. Haeger (1955) ranked floral nectar as most important followed by honeydew and extrafloral nectars for natural sugar sources. In a recent review article, Foster (1995) agreed that floral/extrafloral nectars and honeydew are probably mosquitoes' main source of sugar for mosquitoes, but stated that their importance was difficult to determine. What exactly is nectar? Kevan and Baker (1983) state that nectar is a phloem-sap derivative produced in a series of complex physiological processes in special glands of plants. It is composed mostly of sugars, but also may contain free amino acids, proteins, lipids, antioxidants, alkaloids, vitamins, organic acids, allantoin and allantoic acid, dextrins and inorganic materials such as minerals. Crop sugar contents of mosquitoes were 20-50% w/v as reported by Hocking (1953, 1968). The majority of evidence indicates that mosquitoes are selective about the source of sugar meals and may not feed on whatever is present. All authors agree that the most common nectar sugars consist of sucrose, glucose, and fructose (Wykes 1951, Van Handel 1972, Baker and Baker 1983a, 1983b, Kevan and Baker 1983, Percival 1961). There tends to be disagreement, however, as to the exact composition and relative abundance of some of the less common sugars. Wykes (1951), Percival (1961), Baker and Baker (1983b) state that the monosaccharides galactose and

PAGE 24

8 arabinose, the disaccharides maltose and melibiose, and the trisaccharides melezitose and raffinose are present in some plant species. It is rare to find a nectar with only one detectable sugar and none contain only fructose, or sucrose and fructose in the absence of glucose (Baker and Baker 1983b). Some plant families may be characterized by the relative proportions of the three major sugars. For example, certain plant families such as Lamiaceae (mints) and Ranuculaceae (buttercups) are characterized by sucrose-rich nectars (Baker and Baker 1983a). Other families such as Brassicaceace (mustards) and Asteraceae (composites) have hexose-rich nectars. Wykes (1951), Percival (1961), and Van Handel (1972) examined many plant species in several families and concluded the constituent sugars and their relative proportions in nectar tend to remain constant from any one species, while their occurrence appears to be characteristic for certain families. Percival (1961) tabulates the major plant families likely to contain the less common nectar sugars such as raffinose, maltose, and melibiose. If salivary or crop enzymatic activity is not found in the mosquito species being tested, it may be possible to provide generalizations on the relative abundance of each of the dominant sugars present in wild sugar sources. A highly relevant, but almost completely ignored aspect of mosquito sugar feeding, is that nectar also contains detectable quantities of amino acids. If indirect determination of sugar feeding origin is not possible by examining sugar composition alone, it may be possible that examining both the amino acid and composition sugar will provide valuable clues about where mosquitoes obtain their sugar meals. As listed in descending order of commonness, Baker and Baker ( 1 983a), report floral nectary amino

PAGE 25

acids as follows (percentage in ()): alanine (96), arginine (90), serine (89) proline (87), glycine (84). isoleucine (73), threonine (67), valine (66), leucine (66), glutamic acid (62), cysteine, etc. (55), phenylalanine (55), tyrosine (52), tryptophan (48) lysine (41), glutamine (41), aspartic acid (32), asparagine (27) methionine (20), histidine (19), and nonprotein amino acids (36). Baker and Baker (1973) report the vast majority of the plant species tested to have nectar amino acids in detectable quantities. Furthermore, they state that the amino acid compliments, as well as the concentration of these acids in the floral nectar, will help to determine the nectar's "taste" to the flower visitor in addition to whatever nutritional significance it may have. Similar to sugar concentration/composition, amino acids vary depending on the nutritional requirements of the pollinator. In general, the concentrations of nectar amino acids seem to be greater if nectar is the only or the predominate source of protein for the flower visitor (Baker and Baker 1983a). For example, ants can act as selective agents, and favor plants with particular amino acids in their nectar (Lanza and Krauss 1984). Amino acids also appear to extend life expectancy of insects. Eischen and Foster (1983) reported that nectar amino acids extend survival, but are of insufficient quantity to trigger ovarian development in mosquitoes. It remains unclear as to whether mosquitoes choose their plant sugar sources based on the amino acid contents. Many species of mosquitoes have been observed feeding on a large variety of flower species. Flowers or some plant sugars may serve as attractants. Indeed, sugarcontaining flower or honey extracts are attractants for An. quadrimaculatus, and Cx. nigripalpus (Kline et al. 1990, Hancock and Foster 1993, see Foster and Hancock 1994),

PAGE 26

10 Breeland and Pickard (1961, 1967), observed several flood water species, Ae. vexans, Ae. trivattatus, Ps. cyansens, Ps. ferox, Ps. ciliata feeding on composites such as goldenrod (Solidago) and ageratum (Eupatorium), dock (Rumex) and ironweed (Vernonia). The composites, ox-eye daisy (Chrysanthemum), yarrow (Achillea), golden rod (Solidago), and the common milkweed (Asclepias) were the dominant feeding sources of An. earlei, An. walkeri, Cq. perturhans, Ae. canadensis, Ae. communis, Ae. sticticus, Ae. stimulans, Ae. vexans, Cs. silvestris and Cx. restuans (Grimstad and DeFoliart 1974). With the exception of milkweed, the dominant nectar sources all contained high levels of glucose (considered hexose-rich). The univoltine snow pool mosquito, Ae. provocans, preferred feeding on trees and shrubs (particularly Rosaceae) over herbaceous plants (Gadawski and Smith 1992). Magnarelli (1983) observed Ae. canadensis andAe. stimulans commonly probing white baneberry (Actaea pachypoda), bird's rape (Brassica rapa), maple leaved viburnum (Viburnum acerifolium) and less commonly on yarrow (Achellea millefolium), evening lychnis (Lychinis alba), wild lily-of-the valley (Maianthemum canadense) and false Solomon's-seal (Smilacina racemosa). Direct field observations made by Sandholm and Price (1962), found 10 species of Minnesota mosquitoes feeding on a wide variety of flowering species from a local arboretum. Ae. vexans was present feeding on 39 species of plant, the most common being common milkweed (Asclepias syriaca), wild cucumber (Echinocystis lobata), fall phlox (Phlox paniculata), goldenrod (Solidago latifolia), and meadowsweet (Spiraea latifolia). Ae. vexans and Cx. resturans were the common species of mosquito observed by Vargo and Foster (1984), on Canada goldenrod (Solidago) and white snakeroot (Eupatorium rugosum). The most common

PAGE 27

11 floral nectaries visited by Ae. vexans, and Ae. trivittatus included Queen Anne's lace, (Daucus carota), common milkweed (Asclepias), dogbane (Apocynum medium), and oxeye daisy (Chrysanthemum)Yee et al. (1992). Interestingly, although blooming yarrow was also common, mosquitoes were not observed feeding on it as in other aforementioned studies. Based on the large percentage of mosquitoes that contained crop sugars and the relatively few direct observations of flower sugar feeding, Magnarelli (1980) concluded that Psorophora ferox obtains most of its sugars from sources other than floral nectaries. Both male and female mosquitoes seek a sugar meal shortly after emergence and continue to take nectars throughout adulthood (Yuval 1992). Males probably feed nearly every day (McCrae et al.1976, Magnarelli 1979, 1983, Reisen et al. 1986). Field observations and chemical analyses (cold anthrone test) indicate that females of mosquito species obtain sugar meals throughout the gonotrophic cycle (Magnarelli 1978, 1979, 1983, Andersson 1990, Nasci and Edman 1984, Vargo and Foster 1984, Reisen et al. 1986, Andersson and Jaenson 1987, Haramis and Foster 1990). According to Pappas and Larsen (1978). mosquitoes may remain feeding on a floral nectary for up to 18 minutes. Similarly, lab work by Jepson and Healy (1988), found that Ae. aegypti feed on flowers for an average of 2 1 minutes. Extrafloral Nectaries Plant nectaries located outside the flowers are termed extrafloral nectaries. Most extrafloral nectaries do not involve pollination, and their function is not as uniform as that of floral nectaries (Koptur 1 992). Extrafloral nectaries may be present on virtually every

PAGE 28

12 vegetative and reproductive structure. They occur on the petiole, rachis, upper and lower surfaces of the blade, leaf margin, stipules and on most external parts of the flower. Although inconspicuous and often widely scattered, extrafloral nectars are readily available on assorted ferns and about one quarter of all angiosperms (Koptur 1992). They are unknown in gymnosperms (Keeler and Kaul 1984). Koptur (1992) provides an extensive list of plant species and the locations of their extrafloral nectaries. The sugar and amino acid contents of extrafloral nectars contain slightly different constituents compared with their floral counterparts. Extrafloral nectar has a wider range of sugars, usually being poor in sucrose (Koptur 1992). That sugar concentrations of extrafloral nectars vary over a much wider range than floral nectars of the same species. Similar to the sugar composition, the amino acid composition within a species usually also differs slightly from that of the floral nectaries (Baker et al. 1978). Most notably, the cysteine group (lysine, asparagine, and tyrosine) are more common in extrafloral nectar. Differences in amino acid composition presumably relate to attraction of different insects such as ant or wasp "guards" which may have different nutritional requirements than those of pollinators. Baker et al. (1978) reported that nonprotein amino acids are more frequent in extrafloral nectars. There have been only a few direct observations of mosquitoes feeding on extrafloral nectaries. Of the >20 nectar sources reported by Gadawski and Smith (1992), for Aedes provacans, only one source was extrafloral {Crataegus, hawthorn). Haeger (1955) observed Aedes taeniorhynchus feeding on the tender branch tips of buttonwood (Conocarpus erectus) on Sanibel Island Florida. More recently, Taylor and Foster ( 1 996)

PAGE 29

13 observed Cx. nigripalpus feeding on an extrafloral nectary on caster beans as well as several mosquito species feeding on extrafloral nectars on cashew trees. Honeydew Perhaps the single most underrated and common sources of sugar in dipteran biology are honeydews. The sugary exudate, "honeydew," refers to the liquid excretions from the alimentary tract, as released through the anus by aphids, coccids, and many other plant sucking insects. Auclair (1963) reports that honeydew (a.k.a. mannas) are usually complex mixtures of a large variety of chemical compounds, including several sugars, amino acids, amides, organic acids, alcohols, auxins, and salts. Honeydew typically contains about 1 1% dry matter, of which 88 % are carbohydrates and 7 % nitrogenous compounds. Honeydew is probably extremely common and may be used by mosquitoes and other flies far more often than what the literature reports. Auclair (1963) also summarized that although honeydew production varies widely between Homoptera species, in general, aphids produced 1 .7-20 drops per aphid every 10 hours with an average volume and weight of 0.5 mm 3 and 0.9 to 8.6 mg, respectively. Although, undoubtedly varying by species, it is currently unknown what proportion of the medically important mosquitoes of Florida utilize honeydew regularly as a sugar source. It is possible that honeydew feeding may have at least partially influenced mosquito evolution. Downes and Dahlem (1987), convincingly argue that Diptera (Cyclorrhapha in particular) evolved ingesting honeydew. They conclude that honeydew-producing Homoptera (present in the Permian era) occurred prior to the first

PAGE 30

14 undisputed fossil evidence of Diptera, (occurring later in the Triassic era). Honeydew feeding and/or perhaps the piercing and sucking of sugar-containing plant juices must have occurred. Flowering plants, with their associated nectaries, did not occur until much later in the Cretaceous era. Only a few floral or extrafloral nectars contain sugars, amino acids, or other chemicals unique to specific plant families or genera. This can make generalizations about specific mosquito and plant relationships very difficult if these are based solely on comparative chemical analysis. Honeydew, however, contains oligosaccharides uncommon in plant nectars, and which are largely unique to honeydew and their presence may be indicative of feeding on honeydew. A review of the literature on the sugars commonly found in honeydews produced by Homoptera is summarized by Auclair (1963). Fructose, glucose and sucrose were present in all species tested, glucosucrose (a.k.a. fructomaltose, erlose, 4-alpha-glucosylsucrose, alpha-maltosylfructoside) and melezitose were present in most species, and maltose and maltotriosucrose were present in a few species. A melezitose hydrolysis product, turanose, is also indicative and uniquely associated with homopterous honeydew (Hudson 1946). Using several modern chromatographic techniques, Lombard et al. (1987), found fructose, glucose, sucrose, maltose, turanose, trehalose, melibiose, raffinose, melezitose and some other unknown oligosaccharides in the honeydews of several homopterous species. More recently, Bates et al. (1990) used N.M.R. and Byrne and Miller (1990) used HPLC to characterize trahalulose (a disaccharide) as the dominant honeydew sugar produced by white flies. Further analysis by Yee et al. (1996) ranked white fly honeydew sugar components as

PAGE 31

15 follows: trehalulose > melezitose > sucrose > fructose > glucose. Some sugar alcohols are also commonly found in honeydew. Stachyose has been reported from honeydew but not in the host plant (Byrne and Miller 1 990). Ribitol, dulcitol, and mannitol have also been reported from honeydew (Ewart and Metcalf 1956). These alcohols were not considered unique products of honeydew, however, but were thought to pass unchanged from the plant through the insects alimentary tract. Melezitose, glucosucrose, turanose, stachyose, and trehalulose are not, or only rarely found in plant nectar. By association, the presence of any of these sugars in the crops of mosquitoes is indicative of honeydew feeding. Amino acids are also a common component of honeydew, comprising up to 315% of the total composition (Auclair 1963, Ewart and Metcalf 1956). The most common amino acids present in honeydew are alanine, asparagine, aspartic acid, glutamic acid, glutamine, leucine, phenylalanine, proline, serine, threonine and valine (summarized by Auclair 1963). Unfortunately, in most cases, the amino acids present in honeydew were also present in the host plant. There do not seem to be any amino acids uniquely associated with Homoptera excretion. Mosquitoes can, and probably do, routinely ingest honeydew sugars. Honeydew probably ranges from liquid to a semi-liquid syrup to dry solid residues on stems and leaves. The latter may not pose a problem for ingestion by some species. Mosquitoes have been shown to ingest solid sugars by first liquefying with saliva. For example, Eliason (1963) found Cx. tarsalis, Cx. pipiens, Cs. incidens and Cs. inornata fed readily

PAGE 32

16 on solid, crystalline sugar; however, An. freeborni, Ae. sierrensis, Ae. aegypti and Ae taeniorhynchus did not readily feed on solid sugars. Just as mosquitoes appear to have preferences for certain species of plant nectars, there is no reason to believe they do not also discriminate between different insect/plant honeydews. There have been a few direct observational accounts or chemical analyze proving that mosquitoes ingesting honeydew. Nielsen and Greve (1950), observed Ae. cantans feeding on honeydew and concluded it was probably its main sugar source. They also observed Ae. taeniorhynchus, Ae sollicitans, Cx. nigripalpus, and An. atropos feeding on green aphid honeydew on Bidens spp.. Honeydew feeding by Cx nigripalpus was also observed by Haeger (1955) who found honeydew second in importance compared to nectar. Killick-Kendrick (1987) conducted laboratory experiments demomstrating that the sand fly, Phlebotomus ariasi, ingests honeydew from aphids on plants, and presented strong circumstantial evidence indicating that wild flies feed on honeydew. Using wild sand flies, MacVicker et al.(1990) categorized sugars in the crop using HPLC and found significant levels of melezitose in five sand flies species from different habitats. Other Sugar Sources The piercing-sucking mouthparts of mosquitoes are utilized for feeding on animals, and could possiblly be used for feeding on plant tissues. Patterson et al. (1969) showed plant tissue feeding by mosquitoes, but was insufficient to sustain the life of mosquitoes and thought to be a means of obtaining water. Joseph (1970) observed

PAGE 33

17 mosquitoes feeding on fruit. Other researchers (McCrae et al. 1976), Mogi and Miyagi 1 989) observed mosquitoes imbibing exudates from leaves damaged by feeding insects. By products of fermenting sugars may be attractive to some mosquitoes. Worth (1975) noted that a wide variety of mosquito species were attracted to a sugar-beer-rum bait painted onto the sides of trees. One interesting theory of plant juice ingestion was proposed by Schlein and Muller (1995), who found evidence of plant tissue feeding by sand flies and mosquitoes and concluded that although low in sugar content and questionable as an energy source, the plant juices may be a means of helping to control internal parasites. Diel Sugar Feeding Activities Sugar and blood feeding may not be mutually exclusive events. Field data from Grimstad and Defoliart (1975), Yee et al. (1992), and Yee and Foster (1992) reported that nectar feeding and blood feeding activities occur at or about the same time of day with respect to photoperiod. Although not directly observed, both Reisen et al. (1986), and Yuval et al. (1994), found the percentage of mosquitoes containing fructose changed depending on what time of night the mosquitoes were sampled. The highest percentage occurred around sunrise for Cx. tarsalis and male An. freebornii. Conversely, however, in laboratory studies by Jepson and Healy (1988), Ae. aegypti fed on flowers most frequently at dusk or dawn, slightly different from its normal blood-seeking times. Habitat and trap type may also bias sugar feeding results. A significantly higher percentage of Cs. melanura were positive for fructose at the perimeter of its swamp

PAGE 34

18 breeding habitat than closer to the center (Nasci and Edman 1 984). For Cx. nigripalpus, differences in fructose levels depended on the habitat collected, and the type of trap used in sampling (Bidlingmayer and Hem 1973). Conversely, MacVicker et al. (1990) reported that sand flies captured from different sites and habitats did not differ significantly in the crop sugar composition between any of the species examined. Morphology and Physiology of Mosquito Alimentary Tract Several factors appear to govern the destination of sugar meals. Nearly all of the mosquitoes investigated to date, store sugar solutions obtained from plant juices in the crop diverticula and not in the midgut (Clements 1992). The alimentary tract of the mosquito is typical of most Diptera. Near the posterior end of the foregut are two small dorsal and one large ventral diverticula or crop. Valves separate the three diverticula and the posterior foregut and midgut (see Clements 1992). In general, sugar meals, depending on their composition and concentration, initially go to the crop and are gradually released into the midgut (Friend 1981). The crop is considered a storage organ with no known secretory capabilities (Christophers 1960). Sugar meals may be partially converted into glycogen or fat, but unlike other insects and birds, mosquitoes do not use fat as fuel for flight. Mosquitoes can only use glycogen and simple sugars for flight fuel (Nayar and Van Handel 1971). Likewise, Yuval et al. (1994) showed that only simple sugars and glycogen are used during swarming. Friend et al. (1988, 1989) and Schmidt and Friend (1991), studied the effects of sugars on ingestion and diet destination in Cs. inornata. They found two response systems regulating the crop, one controlling the

PAGE 35

19 amount ingested, the other controlling diet destination. Apparently, the destination depends on the molecular configuration of the meal. At a concentration of 0.5 M for example, common plant sugars like sucrose, maltose, fructose, alpha-glucose, and isomaltose induce crop valve opening and closing of the midgut valve. The valve closing results in the meal being shunted almost exclusively into the crop in the majority of the insects tested. Conversely, sugars not commonly found in nectars such as cellobiose, (5glucose, lactose and gentiobiose induce much less crop opening and almost no midgut closing. As for controlling the amount consumed, Friend et al. (1989), found ingesting large quantities of sucrose directed the meal to the crop, smaller sucrose meals were directed to both the crop and midgut. When sucrose concentrations were 0.4 M or more, most insects ingested large amounts and deposited it into the crop. At concentrations below 0.4 M the mosquitoes did not consume as much, and the meal went to both the crop and midgut. Contrary to popular belief, crop contents are not always strictly sugar meals. Trembly (1952) found crops of several species to contain remnants of blood meals. In order to accurately determine the natural sugar sources of the mosquitoes, samples must be taken as soon as possible after feeding. The crop volume varies depending on species. Aedes communis andAe. punctor have volumes of about 0.91 and 3.39 microliters, respectively (Hocking 1953). Sugar content for a typical crop as in Cs. melanura contains about one microliter of liquid (Friend et al. 1988) and according to Schaefer and Miura (1972), typically contains about 0.087 to 0.45 mg of sugar per crop. Shortly after sugar feeding, the sugars may be wholly or partially broken down

PAGE 36

20 enzymatically (see later section) and begin to slowly empty as needed energetically by the mosquito. In general, sugars appear to be digested in a species-specific manner and the half-life may depend on the ambient temperature and the individual energy requirement for the mosquito. As an example, Andersson and Jaenson (1987) found fructose in 90% of wild mosquitoes immediately after being caught, 30% in mosquitoes held and tested 1 0 hours later, and undetectable 20 hours after sampling. Likewise, Smith and Kurtz (1994) found Aedes triseriatus could empty its crop in as little as 12 hours when fed 10% glucose. If, as stated by Foster (1995), temperate and subtropical mosquitoes only ingest sugar every 2-5 days, collecting large enough samples to determine the relative proportion of the mosquito population feeding on preferential sugars sources (honeydew for example) or amino acids may prove difficult indeed. Clements (1992) states that nearly all pentoses, hexoses, and diand trisaccharides stimulated feeding; however, the acceptance thresholds varied between sugars or mosquito species. Conversely, trioses, tetroses and heptoses failed to stimulate feeding. Clements (1992), also states that the common nectar sugars, fructose and sucrose had low median acceptance thresholds of 0.020 and 0.023 M for Ae. aegypti, but for glucose was considerably higher at 0.1 1 M. The most potent phagostimulant for Cs. inomata was sucrose followed by a 1:1 mixture of glucose and fructose, maltose, fructose, and alphaglucose (Friend et al. 1988). The trisaccharides, raffinose and melezitose also stimulated gorging (Clements 1992).

PAGE 37

21 Crop Enzymatic Activity It is often assumed that the crop is simply a storage organ with little digestion of sugars or amino acids occurs there. Enzymatic activity in the saliva and/or crop may greatly and quickly alter the sugar and amino acid composition of a nectar or honeydew meal making it impossible to determine the exact sugar meal source based on composition and concentrations. To complicate matters further, crop and salivary enzymatic activity may vary by species. Several hematophagous Diptera contain carbohydrases in their saliva (Gooding 1975) which are presumably shunted to the crop upon ingestion of sugar meals. Marinotti and James (1990) and Marinotti et al. (1990) show Ae. aegypti to possess an alpha-glucosidase in the saliva which breaks down some sugars in the crop. Of the sugars tested, Marinotti and James (1990) indicated that sucrose was broken down the fastest, followed by maltotriose, maltopentaose and maltose. Melezitose, trehalose, raffinose, starch and others were only minimally affected. Schaefer and Miura (1972) also found carbohydrate enzymes in the saliva of Cx. tarsalis as well as invertase, maltase, melezitase, amylase and lactase. Indirect evidence for salivary and/or crop enzymes also exists. Crops of sucrose fed sand flies were analyzed with a GC and found to contain the hydrolysis products of fructose and glucose (Moore et al. 1987, Alexander 1988). However, there was no evidence of hydrolysis products for melezitose-fed flies (Alexander 1988). Interestingly, the problem of crop microbes digesting and altering a sugar meal may be reduced. Mosquito saliva contains a

PAGE 38

22 bacteriolytic factor in the glands of both sexes and this substance is present in the crop after a sugar meal (Pimentel and Rossignol 1990, Rossignol and Lueders 1986). Introduction To Mosquito Visual Attractants Review Insects have long been known to have an attraction to certain wavelengths of light (Herms and Ellsworth 1934, Gui et al. 1942, Pfrimmer 1957) The attraction of many haematophageous Diptera to light is thought to be a host-seeking response. Indeed, the majority of insects haematophageous insects attracted to light traps are host-seeking females (see Service 1995). Host-seeking responses of female mosquitoes or other blood feeding Nematocera is largely due to the physiological requirement for a blood meal required to stimulate and complete ovarian development. The initiation of mosquito hostseeking is a least partially due to visual cues which are used to detect illumination levels and perceive objects (Laarman 1955, 1958). Mating, dispersal, appetitive flight, location of sugars, hosts, resting, oviposition and overwintering sites are all governed to some degree by vision (Allan et al. 1987). Many authors' have examined the important visual components of host finding and divided them into shape, color (reflected and transmitted), size, contrast, light intensity, texture and movement. All of these factors alone or in combination appear to play an important role in the ability of a blood-feeding fly to discern preferences and successfully find a suitable host or other resource.

PAGE 39

23 Reflected Light Although this paper is concerned mostly with host-seeking preference as it relates to various wavelengths of filtered transmitted light, much of the work done with mosquitoes and other dipteran color vision has focused on host-seeking behavior toward various wavelengths of reflected light. In general, research on reflected light research has found flies to be most attractive to darker, less reflective colors. Brown (1951, 1954) reported attractiveness to vary inversely with reflectivity or brightness between 475 and 625 nm wavelengths and found darker colors (those with lower reflectivity) more attractive than lighter colors for a variety of wild Canadian mosquito species. Browne and Bennett (1981 ) reported host-seeking female Aedes spp. and Mansonia perturbans were most attracted to black followed by red, blue, white and yellow. Similar results for An. maculipennis andAe. aegypti respectively (Brett 1938). O'Gower (1963) reported black moist surfaces more attractive than gray moist surfaces for Ae. aegypti during hostseeking. Gilbert and Gouck ( 1 957) also found darker shades to attract the most Ae. aegypti, and lighter shades more attractive to Ae. taeniorhynchus. Bracken et al. (1962), Granger (1970), Bradbury and Bennett (1974), Browne and Bennett (1980), and Allan and Stoffolano (1986) all found similar results for tabanids or black flies where low reflective colors like blue, black and red were more attractive than white, yellow and ultraviolet. For nocturnal mosquito species, Barr et al. (1963) reported light trap color to have some effect on catches, but concluded intensity was the most important factor.

PAGE 40

24 Transmitted Light Most dipteran visual research, has focused on diurnal species host-seeking preferences as a function of reflected light color. Significantly less has been written about day or night blood feeding responses to various wavelengths of transmitted light. Ultraviolet lamps have long been known to increase the numbers of host or resource seeking mosquitoes captured at light traps (Headlee 1937, Weiss 1943, Williams et al. 1955, and Breyev 1963). Electroretinograph (ERG) studies consistently show most flies possess a bimodal spectral response. ERG's conducted by Muir et al. (1992) found Ae. aegypti to have spectral sensitivities ranging from ultraviolet (323 nm) to orange-red (621 nm) with sensitivity peaks in the ultraviolet (323-345 nm) and green (523) wavelengths. Similarly, Smith (1986) reported that several tabanid species to have a peak activity of 400-600 nm and a smaller peak between 330-400 nm. No ERG references could be found for nocturnal mosquitoes. Using a different approach, Browne and Bennett (1981) tested filtered light of known wavelengths to equate host preference with landing rates for Mansonia perturbans. Shorter wavelengths (400-600 nm or blue-green) attracted significantly more mosquitoes than longer wavelengths, particularly those above 600 nm (infrared). In fact, for all wavelengths tested, the greatest numbers of mosquitoes landed on colored filters with the infrared wavelengths removed. In Georgia, Bargren and Nibley (1956) found several local species to have varying levels of attractiveness to New Jersey light traps using different color bulbs of similar intensities. Aedes vexans and Cx. salinarius were most attracted to blue (447 nm) lamps followed by red (670 nm), yellow

PAGE 41

25 (570 nm) and white. Culex pipiens fatigans, however, was attracted to red, followed by yellow, white and blue. Culex nigripalpus showed no preference for any of the four colors tested. In field tests, Vavra et al. (1974a,b) tested several types and colors of light and with no significant difference in the numbers of mosquitoes attracted to each of the colors. Similarly, in a lab test using Culex tarsalis, Culex pipiens fatigans and Anopheles sierrensis, Gjullin et al. (1973) tested New Jersey light traps with ceramic dipped bulbs colored red, green, blue, orange and white incandescent bulbs and an ultraviolet light. No convincing difference in attraction between any of the colored lights tested were observed. Conversely, Wilton and Fay (1972) tested Anopheles stephensi, a nocturnally active mosquito, against a clear incandescent bulb and monochromatic light of various wavelengths. This species was highly attracted to 290 and 365 nm in the ultraviolet region and 690 nm in the infrared. Blue, green and yellow (490, 540, and 590 nms) were found to be repellent compared to the clear bulb.

PAGE 42

CHAPTER 2 ANALYSIS OF SUGAR MEAL COMPOSITION OF WILD MOSQUITOES BY GAS CHROMATOGRAPHY Introduction Adequate dietary sugar is a critically important aspect of mosquito and other dipteran survival that is often overlooked. Some species have been commonly observed feeding on natural sugar sources including floral nectars (Haeger 1955, Breeland and Pickard 1961,1967, Sandholm and Price 1962, Grimstad and DeFoliart 1974, Magnarelli 1980, 1983, Vargo and Foster 1984. Yee et al. 1992, Yuval 1992, Foster 1995), extrafloral nectars (Haeger 1955, McCrae et al. 1976, Gadawski and Smith 1992, Taylor and Foster 1996), honeydew (Nielsen and Greve 1950, Haeger 1955), fruit (Joseph 1970), and other sources (Worth 1975, Patterson et al. 1969, Mogi and Miyagi 1989). Several aspects of mosquito sugar feeding have been reviewed by Yuval (1992) and Foster (1995). Few attempts have been made to determine the exact sugar composition within dipteran crops using modern chromatography. Thin layer chromatography (TLC) has been used for pooled samples of tabanids by Magnarelli and Anderson (1977) and of mosquitoes by Magnarelli (1980). Hoppe (1983) and Burgin and Hunter (1997) used high performance thin layer chromatography (HPTLC) for qualitative determination of 26

PAGE 43

27 sugars found in tabanids and black flies, respectively. High performance liquid chromatography (HPLC) was successfully used by MacVicker et al. (1990) to examine the crops of five wild Italian sand fly species. Despite its common use in other systems, only 4 studies have used gas chromatography (GC) for investigating dipteran diets, including Schaefer and Miura (1972) for Culex tarsalis; Moore et al. (1987) and Alexander (1988) for phlebotomine sand flies and Chang et al. (1977) in tephritid fruit flies. All of these studies analyzed pooled crop samples. As indirect evidence for "nectar" feeding, thousands of mosquitoes and other Diptera have been tested using the cold anthrone method that determines the presence or absence of reducing sugars (e.g. fructose) found in almost all plant nectars. Many authors (Van Handel et al. 1972, Bidlingmayer and Hem 1973, Magnarelli 1978, 1979, 1980, Reisen et al. 1986, Andersson and Jaenson 1987, Nasci and Edman 1984, Smith and Kurtz 1994 and Edman et al. 1992) have used the cold anthrone test developed by Van Handel (1972) to qualitatively establish the presence or absence of crop "fructose." According to Van Handel (1967), the cold anthrone reagent reacts with fructose, inulin, sucrose, melezitose, raffinose and others. These sugars all contain the fructose moiety and give positive or negative responses based on a resulting color reaction. Unfortunately, the cold anthrone test does not provide any information about the origin or exact composition of sugars found in the dipteran crop and provides no quantitative information. Indeed, providing there is no enzymatic breakdown of sugars in the crop, it would be possible to determine if flies have preferences for certain families of plants or types of nectar (sucrose-rich for example). Furthermore, we suspect that many of the

PAGE 44

28 reports for "nectar" feeding due to the presence of "fructose" in the crop, may actually have been a positive anthrone reaction to melezitose (a common trisaccharide found in honeydew) or other reducing sugar not commonly found in floral nectaries. GC is an excellent and powerful tool for qualifying and quantifying the exact dietary sugar preferences, occurrence and composition of wild flies. Processing crop contents for GC is only slightly more labor intensive than the methods used for the cold anthrone test. A quantity of less than a microliter is all that is required. The technique is relatively rapid, employing silylation (using a TSMI derivitizing agent) of crude crop contents and subsequent GC analysis. We describe here a method using gas chromatography allowing researchers to determine the individual sugar meal components for mosquitoes and other flies, or the apparent presence or absence of carbohydrases in the crops of mosquitoes and/or other flies by monitoring the decline in parent sugar concentrations, and the increase in the metabolites over time. Methods and Materials Specimen Preparation Representative samples of wild mosquitoes were vacuumed from resting sites shortly after sunrise in a ca. 1 hectare cypress swamp surrounded by pine flat woods located north of Gainesville, FL. Mosquitoes were kept alive, chilled, and processed within four hours of capture. Specimens were sacrificed by laterally inserting a # 0 or smaller insect pin just above the mesothoracic spiracle. Legs and wings were removed

PAGE 45

29 using fine forceps. The crop (ventral diverticulum) was exposed by grasping third or fourth abdominal segment and pulling back slowly so that the crop would emerge between the abdominal segments. For crops containing liquid contents, 1 pi or less of material was sampled using a fine-tipped 1 0 pi capillary tube made from heated and pulled glass tubing. The capillary tube contents were transferred to a 200 pi glass insert tube inserted into a 3 ml glass GC sampling vial (National Scientific Company, Lawrenceville, GA) secured with a teflon lined cap. A separate vial was used for each specimen. Sample Preparation for GC Analysis Sugars are highly polar compounds and their analysis by gas chromatography requires silylation to derivitize the polar carboxyl and hydroxyl groups. The derivitizing agent Tri-Sil Z* (Pierce Chemical Company, Rockford IL) composed of TMSI (N(TMS) imidazole), in dry pyridine was used to process crop contents. Tri-Sil Z (100 pi) was added to each vial containing one crop extract and each vial was vortexed, heated at 60-70 C for 15 minutes and frozen until analysis. Using a 1 .0/100 pi aliquot sample, GC was performed using an Hewlett Packard 6890 instrument with an on-column auto injector, flame ionization detector, and equipped with a DB-5 fused silica capillary column (30 m X 0.25 pm, J & W Co., Folsum, CA). The column was heated from 60 to 300C at a ramp of 20C/min for 20 minutes. Pyridine and acetonitrile were used as solvents to clean the syringe between samples. The following sugars (Aldrich Chemicals, Milwaukee, WI) were made up as 0.01% standards in distilled water: D(-)fructose, D-

PAGE 46

30 glucose, sucrose, maltose, D(+) melezitose, L-arabinose, L-rhamnose. D(+)melibiose, D(+)raffinose, turanose (a hydrolysis product of melezitose), and trehalose. Trehalose is a sugar present in insect hemolymph (Friedman 1985); the others have all been found associated directly or indirectly with plants (Percival 1961, Van Handel et al. 1972). Melezitose and turanose are known to be associated with honey dew (Auclair, 1963). The data and resulting chromatograms and integrations were recorded and processed using a PE Nelson 900 Series (970A) interface with Turbochrome software (Ver 4.1, 1995 [Perkin-Elmer Corp., Cupertino, CA]). Crop Sucrose and Melezitose Hydrolysis Thirty to fifty laboratory reared Aedes albopictus (July 1995, USDA, Gainesville FL) pupae were placed in clean 35 mm film canister lids and allowed to emerge in new 200 ml urine cups with fitted fine mesh screen lids. A total of 25 cups with pupae were assembled. The adult mosquitoes were deprived of sugar and water for two days and kept in a rearing chamber at 28C, 80% RH, and 14:10 L:D photoperiod. The mosquitoes then were provided with two hour unlimited access to 1 0% standard solutions of sucrose and melezitose. Blue food coloring was added to the sugar solutions to aid detection in the crop. Mosquitoes were fed by placing 1 0 drops of the standard sugar solutions atop the screens of each cup. Mosquitoes from each cup were sacrificed until several samples of males and females were found to have contents in their crops. Individual mosquito samples were taken at 2, 4, 8, and 20 hours after ingesting one of the sugar solutions. A

PAGE 47

31 sample of the dyed standard sugar solutions was used as a control (time 0). Mosquitoes were otherwise processed and analyzed as discussed above. Results and Discussion Our preliminary analyses show a wide variation in the types of sugars present in wild mosquitoes. Our chromatographic analysis of combined standard sugars indicate that these common plant mono-, di-, and trisaccharides have unique retention times (Fig. 2-1). Some of the sugars, such as fructose, glucose and others, are anomeric molecules and thus display one peak for each form. Qualitative analysis of crude crop contents of wild Culiseta melanura (Coquillett) (Figs. 2-2) and Anopheles qaadrimaculatus s.l. Say (Figs 2-3), indicate that most of the peaks are identifiable by GC. Figures 2-2 and 2-3 show that these female mosquitoes had recently fed on sugar sources containing fructose, glucose, sucrose, turnanose, melibiose, melezitose, raffinose, and a few unknowns. Later GC runs identified the unknown trisaccharide as erlose (glucosucrose). Honeydew has been found to contain oligosaccharides (e.g. melezitose, erlose, turanose, trehalulose) largely unique to homopteran exudates and are uncommon in plant nectars (Auclair 1963, Hudson 1946, Lombard et al. 1987, Bates et al. 1990, Byrne and Miller 1990, Yee et al. 1996). The presence of a large sucrose peak in Cs. melanura (Fig. 2-2), suggests that either salivary sucrase is not present in this species, or more likely, was not secreted at the time of feeding. Another possibility is that the sugar source contains carbohydrase-

PAGE 48

32 inhibiting compounds that prevents the immediate hydrolysis of sucrose into glucose and fructose. Much of the evidence for mosquito nectar feeding has come from direct field observations rather than through specific chemical qualitative crop content analysis. West and Jenkins (1951), Sandholm and Price (1962), Gadawski and Smith (1992), Breeland and Pickard (1961, 1967), and Magnarelli (1977) have observed some mosquito species feeding at sugar sources during the day. Others have observed them feeding on sugar sources at night, dawn or dusk (Grimstad and DeFoliart 1974, Magnarelli 1983, Andersson and Jaenson 1987, Bowen 1992, Vargo and Foster 1984, Yee et al. 1992). Many mosquito species have been found to contain "fructose" in their crops (Bidlingmayer and Hem 1973), yet some of these species have not been observed sugar feeding. Sucrose and Melezitose Fed Mosquitoes Extensive work has been done characterizing protein acquisition and the enzymatic processes of protein (blood meal) digestion in the mosquito midgut. Relatively little, however, has been done to assess the relative importance of sugar acquisition and its subsequent digestion. GC analysis provides a rapid method of indirectly determining the presence or absence of carbohydrases in the crop (from saliva shunted to the crop). A total of 45 samples of sucrose and melezitose-fed Ae. albopictus were processed. Figure 2-4 shows the time course of crop sucrose and melezitose,

PAGE 49

33 respectively, for males and females following ingestion. Almost complete hydrolysis (>90%) of sucrose occurred within 2 hours of ingestion, while melezitose remained relatively unchanged even 8 hours after ingestion. There was no significant difference between the response of the males and females fed sucrose or melezitose (p = 0.61 and 0.65 respectively). In general, the crops were largely empty 16 hours after ingestion and the number of unidentified peaks increased after 4 hours in the crop. Figure 2-5 shows a representative chromatogram of crude crop sucrose extract 2 hours after ingestion. This chromatogram shows sucrose broken down into the anomers of fructose and glucose. Conversely, melezitose-fed Ae. albopictus remained relatively unaltered 2 hours after ingestion (Figure 2-6) and was found to remain largely unhydrolyzed even 20 hours after ingestion. Crop and salivary enzymatic activity undoubtably varies by species. Depending on the species and type of sugar, GC analysis can reveal important primary sugar meal preferences. Only a few hematophagous Diptera have been reported to contain salivary carbohydrases (Gooding 1 975), which are presumably shunted to the crop upon ingestion of sugar meals. Marinotti and James (1990) and Marinotti et al. (1990) demonstrated that Ae. aegypti possessed an alpha-glucosidase in the saliva which works to break down some sugars in the crop. Of the sugars tested, they found that sucrose was broken down the fastest, followed by maltotriose, maltopentaose and maltose. Melezitose, trehalose, raffinose, starch and others were only minimally affected. Indirect evidence for salivary and/or crop enzymes also exists. Schaefer and Miura (1972) found carbohydrases in the saliva of Culex tarsalis including invertase, maltase, melezitase, amylase and lactase.

PAGE 50

34 Crops of sucrose-fed sand flies were analyzed with GC and found to contain the hydrolysis products fructose and glucose (Moore et al. 1987, Alexander 1988). Likewise, in preliminary GC runs, there was no evidence of hydrolysis products for melezitose-fed flies (Alexander 1 988). At least for some species, it is possible to use GC to determine the proportions and sugars present in the crops of wild mosquitoes and the relative importance that melezitose or other unhydrolyzed sugars play in the diets of medically important Diptera. In most cases, sugar-fed female mosquitoes should be able to enhance their survival, host finding ability, and ultimately improve their chances of vectoring pathogens. A few studies have shown that female sugar-fed mosquitoes of 3 species have lower host avidity than starved or waterfed females (Foster and Eischen 1 987, Xue and Barnard 1997). Other authors, however, have found that females with access to sugar sources are more persistent in their attempts at blood feeding than sugar and water starved females (Walker and Edman 1985). Likewise, Kelly and Edman (1996) found higher parasite transmission rates in sugar-fed Ae. aegypti. This paper proposes a technique allowing an accurate representation of sugars present in individual mosquito meals. Unlike other studies which typically use pooled samples, our GC technique, when applied to sufficient sample sizes, can provide valuable insight into the occurrence, composition, and importance of certain life-sustaining sugar sources for wild mosquito populations. From this information, attractants or other methods could be developed to help control or improve sampling of mosquito populations.

PAGE 51

35 CO CD CO C_? CJ> CO CO 3SO}|Z3|3|/\| asouyjey 3SO)!Z3|3|/\| CO CD 3SOiqi|3^J asoiqi|a|AI CJ> CO CO aso|EM8JX asoue-inj. o o m m o CN asojons CO CO CO in CD CO asoonio asouue^M asoonio asojoeieo asojoruj asojonj-j iiii|iiii|iiii|iiii|iin[iiii|iiii[iiii|iiii|iiii|iiii|iiii|iiii|iiii|iiii|iiii|iiii|iiii|iiii|iiii|ii cd 00000000 in o o d in CD o in 00 o o 0 o o 00 000 000 000000 r~ co m to 01 "a. 8 5 e o o 0 ls — C/5 1/5 -o o q S E E o o (4-1 o U u n g 20 3 M i 09 o s E o -c £ CN (AUJ) asuodsey

PAGE 52

36 asouyjey '9S0J|B^/9S0UBOini asoujjjey <, an 3S0|eqaij. asojcns 2 c -2 53 <3 9SOon|Q • 9S0PITJJ gsopruj asopruj gsopruj esopny o o o o o o o o o o o o 0) 6 <3 esoonio f> CM r (Alu) esuodsey

PAGE 53

37 aso|J3 -2 o to i CO =3 cr xn < asoono asojoruj ssoonio MR xn asopruj asojonjj asbpruj xn 9S0}!Z9|3^ 9S0|J3 gsoutyey 9S0|EL)8J1 •9so)|e^/9soueujnj_ xn asojong CM -2 o CO xn asoonio asopnjj asoonio 3 1 aso pnjj asopnj-j asojoay^ gsopruj oooooooooo ooooooooo moo r
PAGE 54

38 0 5 10 15 20 Hours Post Ingestion Figure 2-4. Timed feeding trials for male and female Ae. albopictus showing the hydrolysis of melezitose and sucrose occurring in the crop (n = 2 to 5).

PAGE 55

39 DO c 1 asojons — = CD O b to CM asooniQ asoonio asojomj asojoruj I i i i i I i i i i | i i i i | asoioruj in o in o CD in o o a in a. o in CO o (AUj) asuodsejj % o o oo £ -a § 03 o -J_ I— > | J v> CO P 3 O 1/1 3 J> O c/i 12 ~0 C/i 3H -J, > ^ 1 S 1 a & .2 a. P OJ co .8 in g 3,

PAGE 56

40 (Aw) asuodsey

PAGE 57

CHAPTER 3 SUGAR MEAL COMPOSITION OF FIVE NORTH CENTRAL FLORIDA MOSQUITO SPECIES AS DETERMINED BY GAS CHROMATOGRAPHY Introduction That male and female mosquitoes and other medically important Diptera feed on a variety of sugar sources of plant origin has long been established. For many species, however, their sugar sources remain speculative. Direct evidence of sugar feeding is documented by observing mosquitoes feeding on a variety of natural sources. These sugar sources include floral nectars (Haeger 1955, Breeland and Pickard 1961,1967, Sandholm and Price 1962, Grimstad and DeFoliart 1974, Magnarelli 1980, 1983, Vargo and Foster 1984, Yee et all 992), extrafloral nectars (Haeger 1955, McCrae et al.1976, Gadawski and Smith 1992, Taylor and Foster 1996), honeydew (Nielsen and Greve 1950, Haeger 1955), fruit (Joseph 1970), and other sources (Worth 1975, Patterson et al.1969, Mogi and Miyagi 1 989). As indirect evidence of sugar feeding, most researchers have used the cold anthrone test (Van Handel 1 972), to determine the occurrence of crop fructose (or other reducing sugars) in wild populations. According to Van Handel (1967), the cold anthrone test reacts with sugars containing the fructose moiety (sucrose, melezitose, raffinose, etc.) giving positive responses based on a color reaction. Many authors (Magnarelli 1978, 1979, 1980, Reisen et al. 1 986, Andersson and Jaenson 1987, 41

PAGE 58

Nasci and Edman 1984, Smith and Kurtz 1994, Edman et al.1992, Holliday-Hanson et al. 1 997) have used the cold anthrone test to qualitatively establish the presence of crop "fructose." However, the cold anthrone test does not provide any ecological information about the origin, distribution or exact composition of sugars found in the dipteran crop and provides little quantitative information. Recent reviews by Yuval (1992) and Foster (1995) covered much of what is known about mosquito sugar-feeding ecology. Few attempts have been made to use modern chromatography to determine the composition of sugars found in dipteran crops. Thin layer chromatography has been used for pooled samples of tabanids (Magnarelli and Anderson 1977, Hoppe 1983), mosquitoes (Magnarelli 1980) and individual black flies (Burgin and Hunter 1997). High performance liquid chromatography was used by MacVicker et al. (1990) to examine Italian sand fly crops. Despite gas chromatography's (GC) common use in other systems, only 4 studies have used GC for investigating dipteran diets, including Schaefer and Miura (1972) for Culex tarsalis; Moore et al. (1987) and Alexander (1988) for phlebotomine sand flies and Chang et al. (1977) in tephritid fruit flies. All of these GC studies analyzed pooled crop samples. Burkett et al. (1998a) presented a method and included representative chromatograms for analyzing sugar meal composition and hydrolysis in individual mosquitoes and found that GC can be used to show honeydew feeding based on crop sugar contents (e.g. melezitose, erlose, turanose). However, they did not assess the importance of honeydew in the diet in different mosquito species. Honeydew is speculated to be a common sugar source for mosquitoes, but its relative importance

PAGE 59

43 among different taxa has not been established. This study identifies many of the mono-, di-, and trisaccharides found in field collected individuals of 5 mosquito species and shows that honeydew feeding is an important dietary component of wild mosquito populations. Methods and Materials Specimen Preparation Wild adult mosquitoes (Anopheles quadrimaculatus s.l. (Say), Coquillettidia perturbans (Walker), Culiseta melanura (Coquillett), Culex nigripalpus Theobald, and Psorophora ferox (von Humboldt)) were collected from several locations in and around Gainesville, FL over the course of 2 seasons (April-October). Ps. ferox were aspirated while attempting to take a blood meal. All other species were vacuumed from typical resting sites (i.e., under bridges, in tree holes, bases of trees, etc.). All collections were made from 0700-0900 and mosquitoes were kept alive, chilled, identified, and processed within four hours of capture. Individuals were sacrificed by laterally inserting a #0 insect pin just above the mesothoracic spiracle. Legs and wings were removed using fine forceps. The crop (ventral diverticulum) was exposed by grasping the third or fourth abdominal segment and pulling back slowly so that it would emerge between the segments. For crops containing liquid contents, 1 pi or less of material was sampled using a fine-tipped 10 pi capillary tube made from heated and pulled glass tubing. Capillary tube contents were transferred to a 200 pi glass insert tube inserted into a 3 ml

PAGE 60

44 glass GC sampling vial (National Scientific Company, Lawrenceville, GA) secured with a teflon lined cap. A separate vial was used for each specimen. Sample Preparation for GC Analysis Sugars are highly polar compounds. Their analysis by GC requires silylation to derivitize the carboxyl and hydroxyl groups. The derivitizing agent, Tri-Sil Z (100 ul) (Pierce Chemical Company, Rockford IL), was added to each vial containing one crop extract. Each vial was vortexed, heated at 60-70 C for 15 minutes and frozen until analysis. Using a 1 .0/100 ul aliquot sample, GC was performed on one of the following instruments. Some samples were run using an Hewlett Packard 6890 instrument with an on-column auto injector, flame ionization detector (FID), and equipped with a DB-5 fused silica capillary column (30 m X 0.25 mm, 0.25 urn J & W Co., Folsum, CA). The column was heated from 60 to 300C at a ramp of 20C/min. Pyridine and acetonitrile were used as solvents to clean the syringe between samples. Depending on availability, other sample runs were made using either an HP 5890 GC, split-splitless injector (250C), FID, J&W DB-5 (30m X 0.32 mm, 0.25 urn), heated from 125 to 300C at 5C/min., 15 min. hold or a Varian 3700 GC, split-splitless injector (250C), J&W DB-1 (30 m X 0.32 mm, 0.32 urn) heated from 125-285 C at 5C/min., 18 min hold. The resulting chromatograms and integrations were recorded and processed using a PE Nelson 900 Series (970A) interfaced with Turbochrome software (Ver 4.1, 1995 [Perkin-Elmer Corp., Cupertino, CA]).

PAGE 61

45 Standards The following sugars (Aldrich Chemical, Milwaukee, WI) were made up as 0.01% standards derivatized as above: D(-)fructose, D(-)glucose, sucrose, maltose, D(+) melezitose, L-arabinose, L-rhamnose, D(+)melibiose, erlose (glucosucrose), D(+)raffinose, turanose (a hydrolysis product of melezitose (Hudson 1 946)), and trehalose. Trehalose is a sugar present in insect hemolymph (Friedman. 1985). but also found in some honeydews (Hendrix and Wei, 1 994). Turanose, melezitose and erlose are known to be associated with honeydew (Hudson 1946, Auclair 1963, Wolf and Ewart 1955). The other sugars have been found associated directly or indirectly with plants (Percival 1961, Van Handel et al. 1972). Results and Discussion The sugars reported here were unambiguously identified by comparison of retention times on 2 GC columns of slightly different polarities. Crop samples contained fructose, glucose, sucrose, maltose, turanose, melezitose, raffinose, erlose, and traces of arabinose, melibiose and rhamnose (Figures 3-1, 3-2, 3-3, 3-4 and 3-5). Fructose and glucose were found in all samples that contained sugars. All species tested contained trehalose, a sugar not found in nectar, that could have originated from hemolymph (Friedman 1985) or honeydew (Hendrix and Wei 1994). Maltose and turanose had similar retention times and could not be distinguished with confidence. A total of thirty unknown peaks were observed across all samples. Most of these unknown compounds

PAGE 62

46 eluted during times atypical for mono-, dior trisaccharides and were probably either incompletely silyated sugars or other miscellaneous crop components. Rarely were the unknown peaks consistent between species, and few constituted a significant percentage of any of the samples. The percentage of mosquitoes containing detectable crop sugars ranged from 10-1 1% in An. quadrimaculatus and Ps.ferox to 47.7% in Cq. perturbans (Table 3-1). Crop sugar occurrence was significantly greater in females than males in An. quadrimaculatus (chi-square, p=0.0006) and Cs. melanura (p=0. 00004). These findings agree with those for other Florida mosquitoes. Bidlingmayer and Hem (1973), for example, found similar sugar positivity results using the cold anthrone test for An. quadrimaculatus (9=15%); Cs. melanura (9=36%, cf=29%); Ps.ferox (=20%); and Cx. nigripalpus (9 and o"=17%). Moreover, Magnarelli (1978) obtained similarly high values for Cq. perturbans (9=57%). Our results differed only in that more female An. quadrimaculatus contained sugars than males. The diverse sugar composition, proportion, and occurrence found in the crops of all species tested show that these mosquitoes tend to be opportunistic and that a variety of sugar sources are used. Ecological Significance of Honevdew in Mosquito Diet The widespread and often invisible nature of honeydew complicates direct feeding observations (Foster 1995). Chromatography has been used to identify honeydew sugars in mosquitoes (Schaefer and Miura 1971, Burkett et al. 1998b) and other Diptera (Hoppe 1983, MacVicker et al. 1990, Moore et al. 1987, Burgin and Hunter 1997. However, until

PAGE 63

47 now the proportion of mosquito populations feeding on honeydew and its ecological significance has only been speculative. All species tested (Figs. 3-1, 3-2, 3-3, 3-4 and 35) contained melezitose and/or erlose (An. quadrimaculatus (55%), Cs. melanura (33%), Cx. nigripalpus (15%), Cq. perturbans (10%), and Ps.ferox (7%)). These data show honeydew to be an extremely important resource for both Cs. melanura and An. quadrimaculatus comprising 1/3 and 1/2 of the sugar meals, respectively. These observations make sense, because unlike Cq. perturbans (Grimstad and DeFoliart 1 974, 1975), Ps.ferox (Magnarelli 1980) and Cx. nigripalpus (see Nayar 1982), which have been observed feeding on a variety of natural sugar sources, An. quadrimaculatus and Cs. melanura have not been reported to feed on floral/extrafloral nectars. Of the mosquitoes investigated in this study, only Cx. nigripalpus had been previously observed feeding on honeydew (Haeger 1955). Chromatography on a relatively small number of Homoptera has found honeydew to be a diverse complex of 20 or more sugars (Hendrix and Wei 1 994), many of which are not found in plant tissues (Byrne and Miller 1990, Tarczynski et al. 1992). A comparison of honeydews among different Homoptera shows that unique sugars are found among various families (Gray and Fraenkel 1953, Wolf and Ewart 1955, Ewart and Metcalf 1956, Lombard et al. 1987, Bates et al. 1990, Hendrix and Wei 1994, Wei et al. 1996,1997, Yee et al. 1996). It is reasonable to expect that different Homoptera produce different sugars and ratios of sugar components that would be identifiable using GC or other modern chromatographic techniques such as HPLC. Although yet to be determined, mosquitoes may prefer honeydew from certain groups of Homoptera.

PAGE 64

48 Percival (1961), and Van Handel et al. (1972) characterized nectar sugars and ratios for many plant species. We initially hoped that GC could be used for determining which plant(s) or even family of plants that mosquitoes use for sugar meals based on direct comparisons with published reports. Unfortunately, unlike melezitose and other trisaccharides, sucrose is rapidly hydrolyzed into fructose and glucose in insect crops (Schaefer and Miura 1972, Burkett et al. 1998b) negating any possibility of determining from which plant a sugar meal came. One interesting observation, however, is that some crop samples from An. quadrimaculatus, Cs. melanura and Cx. nigripalpus contained relatively large amounts of sucrose in their crops, showing that either some sugar sources contain carbohydrase (sucrase) inhibiting compounds, or that salivary sucrase is not always produced and shunted to the crop with sugar meals. Sugar Feeding Field Observations From 1900-1 100 hrs in July and August 1997, two species of plants were commonly observed as hosts for sugar-feeding mosquitoes. Ps. ferox, Ae. albopictus, Cx. nigripalpus, Cx. (melanoconion) spp., and Uranotaenia sapphirina were observed feeding on the extrafloral nectaries of partridge pea (Cassia fasciculata Michx.). More interesting from a public health perspective was the observation of many insects including mosquitoes such as Cx. nigripalpus and Ae. albopictus feeding on the sugary exudates from the developing fruits of bahiagrass (Paspalum notatum Fluegge). Analyses of sugars from samples of bahiagrass exudates matched those from the crops of Cx. nigripalpus sampled while feeding on bahiagrass, showing raffinose to be one of the

PAGE 65

primary sugars. Bahiagrass is a commonly planted turf and pasture grass whose characteristic "V" shaped seed heads are a familiar sight in residential areas in north central Florida. The presence of bahiagrass around human domiciles, and the absence of other suitable sugar sources may influence epidemiological factors for some of Florida's medically important mosquitoes. GC has proved to be an excellent and powerful tool for identifying and quantifying the occurrence and composition of sugar feeding in Diptera. In some cases, the sugar source can be determine (i.e., honeydew). Extracting/analyzing crop contents is not difficult, requiring only small quantities (0.1-1 ul) of crude crop content. Future efforts could address/utilize larger sample sizes, test for seasonal differences, run a variety of different honeydews, and incorporate mass spectrometry to determine unknowns. From this information, attractants or other methods could be developed to help control or improve sampling of mosquito populations.

PAGE 66

50 a 83 09 8 9 S — 5 CO 9 s u, 8 a a (0 o CL CO C\ in oo rt vi oo vi ^ <— i m tN ^ ~-h oo m t> -r „ rl r-~ r-~ ^ vo K o ^ N M h (N v Tf (N r<-> ~ -h ^ r--. _^ £ 2 ^ O ~* Tj" TjrII oo m f"> ^ Tf ~H Si I ^ s ^ ^ ^ O O a s s: s

PAGE 68

52 O < o 9S0)IZ8|8y\| 9S0M3 U Jasouyey O O (O9S0IBM9J1 O sscuons 1 i S asopruj £]asouiuBLjy £y esouiqejv o o o o o o o oo cd eg c c s s tt > H at T3 2 > c a> .2 3 O -*- 'N V 3 § oo £ "2 c S 2 • C C/3 O c
PAGE 69

53 < asoiizaiaw asouyjey QH DO CUMasoieqaji (Q)(MWew/UBjni O OO O0S>OO O O e <3 asojons asoonio — asopruj asoujqejv u -*-* o o w § r> C C ^ § T3 as ~> 0 8 1 1 o o a. T3 C c a> 22 > .2 M 3 X) 6 j so S So _. C 2 S £| fe o i u • C g i e g-J 2 <2 H 22 8 o o o o 00 CD CM oi!nbso|/\| J8d je6ns dojQ % u,

PAGE 70

54 O O <]aso|J3 Eh asouyey <3Hh aso|BM9Ji }|B|/\|/UBJni O O 3^ as0J3n s < < 1 I esopruj esouujBuy (gh asoujqejv 8 § S ^ CM o^nbso|/\| J9d je6ns dojQ %

PAGE 71

55 [>asoijzaia^ 8S0|J3 asouyjey <^>asojons — asoonio asopruj asouweijy asouiqejv o o o o o o O 00 CD ^ CM ojjnbso|/\i J9d je6ns dojQ % a £ I O -o 0) *" c > a. I 2 c H 5 1-1 J II { *-< O 4> c >~> o 2 I o js o o 5 v •o c c w CO T3 23 F C o .9 3 X> 09 1 1 ••3 8 •3 I GO ^ c a. ca 2 fe C co O u>
PAGE 72

CHAPTER 4 FIELD EVALUATION OF COLORED LIGHT EMITTING DIODES AS ATTRACTANTS FOR WOODLAND MOSQUITOES AND OTHER DIPTERA IN NORTH CENTRAL FLORIDA Introduction Light emitting diodes (LED) were evaluated as an alternative light source for use as an adult mosquito attractant. Much of the research on Dipteran color preference is based on reflected light (Brett 1938, Bracken et al. 1962, O'Gower 1963, Granger 1970, Bradbury and Bennett 1974, Browne and Bennett 1980, 1981, Allan and Stoffolano 1986). Many authors have shown that mosquitoes are attracted to transmitted light (Headlee 1937, Weiss 1943, Williams et al. 1955, Breyev 1963, Bargren and Nibley 1956, Gjullin et al. 1973, Wilton and Fay 1972, Browne and Bennett 1981). Few reports detail the response of individual species. Several colors (100 nm width) of highly efficient, low cost, "super bright" LEDs have recently been developed. These colored LEDs when used in CDC traps have a greater intensity and require significantly lower amounts of energy (ca. 0.125 ma/hr vs. 150 ma/hr for standard CM-47 bulb). We evaluated the relationship between transmitted light color and its attractiveness to woodland mosquito and other dipteran species.

PAGE 73

57 Methods and Materials Three field trials were conducted using standard CDC-type traps (John W. Hock Company, Model 512, Gainesville FL) modified by replacing the standard bulbs with the LEDs. The LED was secured into a piece of 2 X 2 cm plexiglass and fastened to the "screen" atop the lid assembly 3 cm below the aluminum trap lid (Fig. 4-1). Trial 1 used 9 kg compressed-gas (carbon dioxide) cylinders equipped with double-stage (Victor Equipment Company, Model VTS453B-320, Denton TX) and microregulators (Nupro Inc., Series M, Wiloughby OH) to maintain a constant gas flow of 200 ml per minute. Carbon dioxide was delivered to the trap through a 3 m piece of 8 mm clear plastic tubing secured with a rubber band so the top of the tubing was even with the top of the trap opening. Gas flow was checked each morning and evening using an in-line flowmeter (Gilmont Instruments, Great Neck NY, no. 12). Mosquitoes attracted near the trap intake were drawn in by the trap fan, blown through a screen funnel and into a quart polypropylene jar containing a 3 X 6 cm piece of dichlorvos impregnated vinyl strip used as a killing agent. Batteries consisted of 6V, 1 0 ampHrs rechargeable gel cell batteries (Powersonic Corp., San Diego CA) which were used to run the fan motor and standard incandescent light. Six Colors With and Without Carbon Dioxide (Trials 1 and 2) From July 15-20 (Trial 1) and July 22-27 (Trial 2), 1996, different colored lights were used as attractants in standard CDC-type surveillance traps using a 6 X 6 Latin

PAGE 74

58 square design. Light, day, and position effects were evaluated using a three-way ANOVA (SAS Institute, 1995) for the total number and most common species represented in the traps. Multiple comparisons were made using the Ryan-Einot-GabrielWelsh multiple range test (alpha = 0.05). Trial 1 used carbon dioxide as an additional attractant and Trial 2 used the same randomization, but did not use carbon dioxide. Four different colored "super bright" LEDs (Toshiba Tosbright, Martech Optoelectronics, Latham, NY) were compared to no light and a standard incandescent bulb (John W. Hock Company, CM-47, 6.3 V, 520 millicandela [mcd], Gainesville, FL) used as controls. The diodes tested were red (613 50 nm, 1600 mcd, 22 ); orange (605 50 nm, 2000 mcd, 22); yellow (587 50 nm, 2300 mcd, 22) and green (567 50 nm, 2400 mcd, 8). Each LED was powered by 2 alkaline D cell batteries at 2.8 0.2 volts and 1 8 2 ma. A 10 ohm resistor was placed in series to prevent over driving the LED. Trials 1 and 2 were conducted at the University of Florida's Austin Cary Memorial Forest, a research area located north of Gainesville, FL. The habitat consisted of a cypress swamp surrounded by pine flatwoods. Traps were hung 165 cm above ground level and placed every 30 meters near the banks of a seasonal forest stream that runs through the middle of the swamp. Due to the thick vegetation, none of the traps were visible to each other. The trapping period followed a 24.5 and 6.83 cm rainfall on July 5 and 9, respectively. These trials ran from 1 800 to 0800 hr for six days in a row using a 6 X 6 Latin square design. The trap and motor assembly remained stationary, but the lights or diodes were changed nightly so each light would occupy each position during the six-day period.

PAGE 75

59 After each trap night, the captured mosquito adults were identified and counted. All Culex (Melanoconion) Theobald, Anopheles quadhmacalatus Say, An. crucians Wiedemann and Aedes atlanticus Dyar and Knab/vk. tormenter Dyar and Knab species were pooled, as these taxa could not be distinguished with confidence. Eight Colors With Carbon Dioxide (Trial 3) Trial 3 was conducted from August 12-21 (Trial 3), 1996. Due to the fluctuating water levels and mosquito populations at the Austin Cary Forest site, a similar, but more permanent cypress swamp habitat was chosen north of Gainesville. In addition to the four LED colors and two controls previously discussed, two additional LED wavelengths, infrared (940 50 nm, 22[Martech Optoelectronics, Latham NY, model MTE1080]) and blue (450 50 nm, 800 mcd, 22 [Panasonic, Digikey Corp, Thief River Falls, MN]]) were evaluated also. Using an 8 X 8 Latin square design, traps were placed around the perimeter of the swamp. Each trap was baited with 200 ml/min carbon dioxide as in trial 1 and otherwise treated as before. Results Six Colors With and Without Carbon Dioxide (Trials 1 and 2) During the six trap-nights of trials 1 and 2, 32,059 and 1,916 specimens of mosquitoes were collected respectively. The mosquito species composition attracted to the incandescent light trap agree with those found by Mann (1993). Response of the most numerous mosquito species are shown in Figures 4-2, 4-3, 4-4 and 4-5. Means, standard

PAGE 76

60 errors, p-values, and significant differences for species represented in large enough numbers are shown in Tables 4-1 and 4-2. As noted in the tables, there were significant trap-position and day effects for some species. No significant differences were observed for the total number of mosquitoes captured at different colors in either the C02 baited or unbaited trials (p = 0.08, p = 0.24, respectively). Differences were observed for individual species however. In trial 1, Aedes dupreei (Coquillett), Ae. infirmatus (Dyar and Knab), Anopheles crucians s.L, Culiseta melanura ((Coquillett), and Uranotaenia sapphirina (Osten Sacken) showed significant color preferences. In trial 2, only Ae. atlanticus, An. crucians s.L, and Ur. sapphirina showed significant preferences. Aedes dupreei was the predominant species and was the only species preferring the carbon dioxide baited trap using no light. This species was also abundant in the no light control. Three female Lutzomyia shannoni (Dyar) and 1 Lutzomyia vexator (Coquillett) were collected during trial 1 None were collected in trial 2. Eight Colors With Carbon Dioxide (Trial 3) During the eight trap-nights, 4,668 specimens of mosquitoes, 1,189 tabanids (Diachlorus ferrugatus Osten Sacken), 3,667 chaoborids (Corethrella spp.), and 3 phlebotomine sand fly specimens were collected. Response of the most numerous mosquito species is shown in Figures 4-6 and 4-7. There was a highly significant difference in the total numbers of mosquitoes captured for the different colors (p = 0.0001). Means, standard errors, p-values, and significant differences for species

PAGE 77

61 represented in large enough numbers are shown in Table 4-3. As seen in trials 1 and 2, there were significant trap-position and day effects for some species. Overall capture of mosquitoes was significantly greatest with the standard white broad spectrum incandescent, followed by blue, green, orange, yellow, red, no light control, and infrared respectively. When collections were classified by mosquito species, clear preferences were seen between species. Anopheles crucians s.L, Cs. melanura, Cx. nigripalpus Theobald, Ps. columbiae (Dyar and Knab) and Ur. sapphirina showed significant color preferences. White light captured the most An. crucians s.L. The greatest numbers of Cs. melanura were captured in traps with white, green, and orange. The most Ps. columbiae were collected in traps with blue and significantly more Ur. sapphirina were captured in traps with standard white or blue. No colors were found to be repellent to mosquitoes when compared to the no light controls. No significant difference (p = 0.26) for color attraction were obtained for the tabanid, Diachlorus ferrugatus. Conversely, the chaoborids, Corethrella spp., showed significant color attraction (p = 0.002), preferring white and blue over the other colors. Discussion Many common Florida woodland mosquitoes are medically important. Although one of the primary means of evaluating the presence/absence and relative abundance of certain mosquitoes is through the use of light traps, few studies have evaluated mosquito response to different wavelengths of transmitted monochromatic light. Even fewer studies have detailed the response of individual species. Browne and Bennett (1981) tested

PAGE 78

62 filtered light of known wavelengths to equate host preference with landing rates for Coquillettidia perturbans (Walker). They found shorter wavelengths (400-600 nm or blue-green) attracted significantly more mosquitoes than longer wavelengths. Their results correspond well to ours in Trial 3. In Georgia, Bargren and Nibley (1956) found Ae. vexans, Cx. salinarius, and Cx. quinquefasciatus to have varying levels of attractiveness to New Jersey traps using different color bulbs of similar intensities. Other species, such as Cx. nigripalpus, showed no preference for any of the four colors (447, 570, 659, 670 nm) tested. This finding agrees with ours in Trials 1 and 2, but differs from those of Trial 3 that found significant color preferences (blue, green, white) for Cx. nigripalpus. Vavra et al. (1974a) tested several types and colors of light and found no significant differences in the total numbers of mosquitoes attracted to each of the colors. Attraction of individual species of mosquitoes was not examined. In a laboratory test using Cx. tarsalis, Cx. quinquefasciatus and An. sierrenis, Gjullin et al. (1973) tested New Jersey light traps equipped with either ultraviolet light or ceramic dipped bulbs colored red, green, blue, orange, and white. They found no significant differences in attraction between any of the colors tested. Wilton and Fay (1972) tested An. stephensi against a clear incandescent bulb and monochromatic light of various wavelengths. They found this mosquito highly attracted to 290 and 365 nm in the ultraviolet region and 690 nm, but that blues, greens and yellows (490, 540, and 590 nms) were found not as attractive as the clear bulb. Allan et al. (1987) stated that crepuscular and nocturnal biting flies are unlikely to have well-developed color vision, but their abilities to detect differences in intensity

PAGE 79

63 contrast are likely to be well-developed. Spectral sensitivity (i.e. relative sensitivity of the retina to light of different wavelengths) studies consistently showed most flies possess a bimodal spectral response with peaks around 340 and 525 nm (White 1985). Considering the variation in mosquito species attractiveness to light traps (Huffaker and Back 1943, Bidlingmayer 1 967), it is not unreasonable to expect differences in color preference based on variations in their spectral sensitivities. Alternatively, attractiveness may not be due to color per se. Different wavelengths may be physiologically perceived as more intense and subsequently more attractive. Barr et al. (1963) concluded that more intense light (up to a point) is more attractive than less intense light. Electroretinographs (ERG) conducted by Muir et al. (1992) found Ae. aegypti to have spectral sensitivities ranging from ultraviolet (323 nm) to orange-red (621 nm) with sensitivity peaks in both the ultraviolet (345 nm) and green (523 nm) wavelengths. By studying the relationship between illumination and suitability of an oviposition site for Ae. aegypti, Snow (1971) found a similar bimodal response. Unfortunately, all mosquito spectral sensitivity studies have focused on Ae. aegypti, none have been conducted on nocturnal or other species actually commonly attracted to artificial light. Unlike the standard bulbs which radiate light in a 360 pattern, the diodes emit a narrow beam (8 or 22), which in these tests was oriented upwards and reflected off the shiny surface of the aluminum pan covering the CDC trap. If the mosquitoes are responding to differences in perceived intensity, an assortment of several of the most promising LED colors (e.g., blues and greens) could be arranged to produce a 360 pattern

PAGE 80

64 and perhaps greatly increase the efficiency of the trap. Several LEDs in series would be many times brighter and still use significantly less battery power than a single incandescent bulb. Several of the ERG studies previously mentioned have shown peak dipteran spectral activity from 450 to 550 nm. Based on the numbers of certain mosquito species that were attracted to the blue and green wavelengths, an LED producing wavelengths between 450 and 550 nm may produce excellent results. Currently technology limits production of LEDs producing these wavelengths. Super-bright blues (450 nm) have only recently become available, and perhaps future technology will produce a blue-green diode peaking at about 500 nm. LEDs run on significantly lower amounts of energy (ca. 1 ma/8 hrs) than incandescent bulbs resulting in substantial savings in battery life and expense. Hours of use (means SEM) with the no light control (697.5), blue (639) and green (635.7) were found to last significantly more hours (n= 4, p=0.02) than the standard white (360) bulb. For convenient use, LEDs can be soldered in series directly into the existing trap circuitry. Best results for all colored LEDs except blue (100 ohm) were obtained using 200 ohm resistors connected to the light motor assembly. Future studies should focus on combinations of colors oriented in different directions. The use of "superbright" LEDs warrants serious consideration as a mosquito replacement for standard incandescent bulbs used in light traps. These results have potential for use in population dynamics studies or for enhancing the attractivity of certain species.

PAGE 81

65 CM r-r t1 1 <0 CXI m CD > — i rr O iD ro ro o o 1 i — i CD 1 1 CD m O o l o z: o Q o : o : o o o o a III m m Si .a rH ro on a 03 01 ro m a in ro ro CO ro CJ CM o CD a) co O to CO CM o 31 C'l r.1) 01 o 1 J I — 1 CH rH c to : r•r : CN rH 3 rH C -.r o c 1 : > — I Hi +1 +i +i +i H + -H +1 +1 + -H HH + +i -h + HH HH CD l> c~ O CM 00 tn O N to O in CM CM CD o p* 01 D 0 m • o rCO O CN O r* a i — I O i O* CO ro ci CM ul 11CM rro CM ro ro A ro ro ro O ro ro XI -i O O o if o CO rH o H CN o CO CM o CM O o > — i CM o in 3 o : — 1 1 H •H o 1 C 1 o c — i +1 +1 +4 -H +1 +i -H + 1 -H +! • HH + +1 HH -H +i -H X in CM o riX o O CO O rCN O o 111 ro CM ro in c < — i : :> o CO i CM O O iri -J' o cc CM CO ro ID CM CM X' OJ I rH rH .-i z T> Z -. rH o rH O o '71 in a o i,i : C rH -H +1 M HH +l HH tH + 1 • HH +i +1 t +j HH -H HH 4H HH if) o tn oo i-i) O o ro CO CM o CO CM O CO in CM r> CM : i.ii rH m O o o rH o rO rH rr rH ro O oi ro H eg Oi m — 1 ro ro o CO IT) O UO CM If) A ro CD If) O HH HH CO CM rH O HH HH CO CO oo co cn XI CO CO rH n a £ rH m -r ro ro ro ro ro ro ro ro CO m ro CO CM CM CM rH a *f 00 CM Xi o 00 CM rH rIS (0 CO CO CD <§ rH Z "'I O in CM : ir rH o o O rH m CM o rH HH -H -i-i + i i -H +i +1 H H +1 +1 H H M +-i Y\ +i HH rCM O CM 00 o r^ CO CM ro o CM hi rO al O ai c 1 : 111 m : 00 co rH -' a o rH X CO z C*J rH 10 CO CM rH 111 CO CM on ro on 111 ro a JH XI ro o UI ro o ro r^ m ro in ro CO ro m ra ro X o ro o CM o o o O r~ in 31 ro O Xi CD O T o c§ o rH CD rH rH rH CM : o o •r o rH CO D 3 rH T3 +! +i +! +i H + 1 +1 +1 +l +i H +1 HH HH HH +l HH HH o e rhi CM CM CM CM o O m o CM O CO CM CM C) O CM rC V CA o cn o CD O rH C ID to c ii i rH o rH c 3 10 ci O rH IT, rH o O 1 9 m ro ai V >1 ~ -Si J! I CU ^ •2 s: -Si u c^ =: *— -2 5 S I 3 .11 la I lliil 3 oi C C r H X X *i tri CN 1^ CU "3 R. 5 in O d I o f-

PAGE 82

66 en co in h to CM o o o 8 m n 10 a a m (0 tn ;n re 10 .-i o t- o ci T CM 00 *r CI o o o : —1 in : H o o o o 3 — i +1 co H m o +i ci -H m 4-1 CM +i IT) H oo +1 o +1 o f! O H in 4H r~ CM rO CI o o 00 ci — i c a o Z 10 u re Xl re m ra a 10 1.0 o CM o 0 1 o o o CM o o CM o o 00 o — 1 o 0 o O c o O o c +1 +1 +1 + 1 +1 -i-i +4 +4 +1 -H +1 +1 +1 — CM o CO o CO o o o CM O o CM o rm o o o O o o o oo ci o s o I o XI CM a 10 re 10 to 10 m O CD O 00 — 1 o o 1 1 C-J CM CM o CM : • : c 1 — 1 :> H o : c Cl H +1 +H 4H -H -H t-i H 44 +1 +1 +i +4 +i cj o 00 O rm o m o CM CM CO CM CO c Cl CN — i 5 C J O c c UD IS U n jQ (0 a 10 re 10 re XJ CM o CM o o o CTt CM CO o o CI CM rCM o H J o O o o o O CP -H + +i +1 H + 4-! +H 4-1 +1 + 4H H H uo o CM o r~ o CI CM r* o O rOD lO 1 p! 3 CM o iH o o o c C : CM s 73 a .0 n rin (13 a re ra n M XI o Cft CM a\ Cl O cl r~ CJ o CJ C) XI C) t-H O CJ : ^< O o c CJ H +1 +: +l -H H 1 4-1 +1 4-1 H "1 +1 -H cl o ci CM 00 o O rCM o CM 00 CM '-n : in o CM O : o : D CI L1C I u m XI RJ 'X' re 10 XI a m re 10 XI 'X) CM Cl C J pIT) o m ci CM o CM Si — i H o c o c o : 0 .-1 -H +1 + H +1 +1 4: 4-1 H +1 H -H CM Cl CM r~ 00 o 00 Cl CM o CM o <3C 00 o i-i o o o o o o P* o l C/3 C3 3 g '65 | I ojcu i i *i s: B >3 a 1 a R S S -S a .Sis 2 Cj CJ. g I s s S. 8 "ft* <3 R do s ^ -5 u u a, q; o Z 1

PAGE 83

67 3 3 | 2 3 .1 A 0"1 to CM ffl -r to to m r£1 co O n GO 9bc 7ab to en CD CM 5 C ro H o ro <-> 1 O H ro (M rH CM H -H ll 44 to +1 if) 10 +1 t i H H o -11 m -H CO +1 fl CM 0 H C > — 1 rH : CM rH rr— i Kj a tn .'1 CM IO CM rO O m o rH -H 4 +i 4 41 to ro m If] Q : ro m u ro 7ab m M a") CM vO O 4-1 CM rH -H CJ +1 CO H ffl O "d" rH lO CM CO rH H re el in in h cm r^ o o t -H -H -H -M -H m m to ro to O tH o 0 n to 4J .u £1 H U 10 rr-4 rH O H o 0 J rH 4-1 rH +1 (0 -H H rH 10 O CO -4 O CM 10 CM rH' 4H rH O 0 U 10 H n to n CM 04 rn CM ^* ro rro ^4 o en o O 44 4-1 +i + 41 +1 CD H rH DO to o c *T o in H CM 4 re CJ 10 10 n m to m 00 C4 rH *r c 3 D CI o o H 4-1 -H 4-1 +4 +1 41 10 10 o -4 rH ffl ffl CM rH O n ^4 1 10 T1 a 0 o U to rro H H 1 +1 o c i rH 44 o 4-1 m 41 +1 rH 10 rH 0> o S "J rH ^ 0 io CM O C a s: e = -3 S S ? 3 J. ? ~ .3 cu cu cu cu ^ ""3 § to § 65 a s Hi 1 a o c -5* o a a s: -<3 C ^ rH 0 C in a m a •t 10 CO 9bc rH 10 XI <0 CM lab n o fl o H Q m ro to 10 lb "T 41 a\ c 41 c 4 ro O 41 rH rH 3 1 I o ^1* -i ro o M o I o 4-1 3 rH 4! to o rH +i o 10 +1 (O CO 41 O 41 rH 41 o\ hH .1 H 1 1 'Xi 1 o rH 4-1 O 41 It) a. rH m o ; o m o in CM 10 c H a o o .H CM 9) i ^ I § rn CM + to U 10 a ... rH CM Ci c J 10 O +1 H rH 10 Oh rH m m CM rH -4 p XI 2 ^ CM CM u Hi U >U Ml Ml C St CSI Crj >< V H &rj ^ ^ n 8 3 "3 I 5 o Z a 6 f£ 0.05). I o V i d B distill: a(p< effect not be asition

PAGE 84

68 3 I i Oh m m tn od cj o o o o -H O -H o ,1J 10 a ro vo rj O o a o : H +1 +! f -H o m O o K H ro C J o ] c ro 9 c c r 4 XJ cs H 3 8. t/3 a a 6c e u Q I e C a •^a a ^ £ -~ S Is -S> 5 a y w w C ^ ^ a a 5 I 1 s o y 5 § i-H i— I O ro coo o ro INI ro m ro r~ a ro iH T3 If) o ro rXI 'T 7ab ,0a H O o a ra ro T lb rn 01 CJ +H : ci o +i 10 If) H 00 tH + 1 T 5 +1 O J H o o +i — 1 H m o t ID -H : +l H -H o o +1 o o +1 m — ( H If) < — i H c c\ c If) — i O C J 0\] -T m i — i : o eg S 2 3 a -5 5 a U g c t <3-' M MM M sS no to v> J< 1 I o d z 1 s 8 i i 8 £ o d v a 3 I 1 5 5 ill

PAGE 85

Figure 4-1. Modified CDC-type trap equipped with LED.

PAGE 86

70 o S c c o a> ab s c £ s E CD CQ J c n CD CD 8 I CD o •3 S3 i op g B § £ i u -a 0 "a X) N o u s "3 U G 2 | is a E Cm O c o :5l 9'3 i M s o 00 H 9 8 3 o a 3 0) o I £ 3 O CO m CM o CM m m i u u

PAGE 87

71 S c to eg c TO •e £ 0) er O c o c o u o c -2 <3 n .a c £ to C .a 3 C o O CD O LO o o CO o C\J c Q M -J ~o u o o DO c 55 3 ge o. a i u Ok >. I U Q U -o t> £ 8 oo I -a J2 II C 00 r. I p CO O fi St l— Oh 00 "3 ga V ^ 1 1 3 cr t o E H-* oj 3 X3 go 8 w "S c od >^ a ps r<-> 4 CJ H— p9jnjdt3Q SGjoeds %

PAGE 88

72 CO 03 |_J ~ t e f s cc o > a 2 S ea I/a 'sjyffiyyy, I • 1 O o CO o C\J p9jn;dBQ sepads %

PAGE 89

73 m 9 o o CD O lo o o CO o CM c o c O S o c TO I d TO .a c C d c .TO g JS i c u o CO U -a c cO O C o CO Q W -o u a o I o c W i c CO 5 c i c CO O 00 .8 CO ItS 3 5 co 00 C CO is o Q. O g Q U c CD •*-* S I 3)

PAGE 90

74 pejrndeo sepeds %

PAGE 92

CHAPTER 5 EVALUATION OF BLUE AND GREEN LIGHT EMITTING DIODES AS ATTRACTANTS FOR WOODLAND MOSQUITOES AND OTHER DIPTERA IN NORTH CENTRAL FLORIDA Introduction Field trials by Burkett et al. (1998a) evaluated capture numbers and species composition for light traps equipped with different colored light emitting diodes (LEDs) and compared them to those using standard incandescent bulb typically used in CDC-type traps. Results of these trials showed blue (450 50 nm) and green (567 50 nm) "superbright" LEDs attracted nearly as many mosquitoes (both in numbers and species composition) as the standard bulb. The LEDs used in those trials were oriented "up" to reflect the LED "light" off the aluminum trap lid. The number of mosquitoes captured in light-baited traps is directly proportional to the intensity of the light (Barr et al. 1963). Our objective was to determine if trap efficiency could be improved by orienting the LED's away from the trap in a 360 pattern rather than reflecting the light off the trap lid. Orienting the LEDs away from the trap could increase the visible radius and thus allow the traps to be detected from a greater distance. Aedes aegypti (Muir et al. 1992, Snow 1971), and other Diptera (Agee and Patterson 1983, Allan and Stoffolano 1986, Smith 1986) have bimodal spectral sensitivities with slight variations in X max occurring in the green and blue areas of the 76

PAGE 93

77 electromagnetic spectrum. Not all mosquitoes are attracted to light. Likewise, not all species are attracted to the same wavelengths. Much of the research on dipteran visual attractant research is based on the response of diurnal species to reflected light (Brett 1938, Bracken et al. 1962, O'Gower 1963, Granger 1970, Bradbury and Bennett 1974, Browne and Bennett 1980, 1981, Allan and Stoffolano 1986). Visual attractant research based on transmitted light is less common (Headlee 1937, Williams et al. 1955, Breyev 1963, Bargren and Nibley 1956, Gjullin et al. 1973, Wilton and Fay 1972, Browne and Bennett 1981). Few reports detail the response of individual species and often focus on species easily reared in the laboratory, but not commonly captured in light-baited traps. Several colors ( 1 00 nm width) of highly efficient, low cost, "super bright" LEDs have recently been developed. LEDs are an improvement over standard incandescent bulbs because they produce greater light intensity and require significantly lower amounts of energy (ca. 0.125 ma/hr vs. 150 ma/hr). In this study, we evaluated the relationship between orientation of the LED on a light trap and capture numbers and species composition for woodland mosquitoes and other dipteran species. Methods and Materials Field trials conducted 1800 to 0800 hrs from August 9-14, 1997 in a 6 X 6 Latin square design evaluated light-trap captures of common woodland haematophageous Diptera including mosquitoes, Corethrella spp. (Diptera: Chaoboridae), Diachlorus ferrugatus Osten Sacke n (Diptera: Tabanidae), and phlebotomine sand flies (Diptera: Psychodidae). Trials were conducted in a ca. 1 hectare cypress swamp surrounded by

PAGE 94

78 pine flat woods located north of Gainesville, FL. Light, day, and position effects were evaluated using a three-way ANOVA (SAS Institute, 1995). Multiple comparisons were made using the Ryan-Einot-GabrielWelsh multiple range test (alpha = 0.05). The trap and motor assembly remained stationary, but the lights were changed nightly so each would occupy every position during the six-day period. After each trap night, trap collections were separated, weighed, identified and counted. Standard CDC-type traps (John W. Hock Company, model 512, Gainesville FL) were modified by replacing the standard bulbs with the LEDs. LEDs oriented "out" in a 360 pattern were compared to those oriented "up" and reflecting off the aluminum trap lid. LEDs were soldered in series to insulated 22 gauge wire shaped to form a 40 cm diameter circle (Fig. 5) and were secured with rubber bands to the 1/4" mesh screen atop the trap lid assemblly. The green or blue LEDs oriented up or out were compared to controls consisting of no light (fan only) and a standard incandescent bulb (John W. Hock Company, CM-47, 6.3 V, 520 millicandela [mcd], Gainesville, FL). The LEDs used were green (567 50 nm, 2400 mcd, 8 [Toshiba Tosbright Martech Optoelectronics, Latham NY ]) and blue (450 50 nm, 800 mcd, 22 [Panasonic, Digikey Corp, Thief River Falls, MN]). Power for the green and blue LEDs were supplied by 2 and 3 alkaline D cell batteries at (2.8 0.2 volts, 1 8 2 ma) and (4.3 0.2, 20 3 ma) respectively. A 1 0 ohm resistor was soldered in series to prevent over-driving the LED. Trials used 9 kg compressed-gas (carbon dioxide) cylinders equipped with double-stage (Victor Equipment Company, model VTS453B-320, Denton TX) and microregulators (Nupro, Series M, Wiloughby OH) to maintain a constant gas flow of

PAGE 95

79 200 ml per minute. Carbon dioxide was delivered to the trap through a 3 m piece of 8 mm transparent plastic tubing secured with a rubber band so the top of the tubing was even with the top of the trap opening. Gas flow was checked each morning and evening using an in-line flowmeter (Gilmont Instruments, no. 12, Great Neck NY). Mosquitoes attracted near the trap intake were drawn in by the trap fan, blown through a screen funnel and into a quart polypropylene jar containing a 6 X 6 cm piece of dichlorvos impregnated vinyl strip used as a killing agent. Six volt, 10 ampHrs rechargeable gel cell batteries (Powersonic Corp, San Diego CA) were used to run the fan motor and standard incandescent light. Traps were hung 165 cm above ground level and placed every 30 meters around the outer perimeter of the swamp. Due to the thick vegetation, none of the traps were visible to each other. The trapping period followed a 1.7, 2.3, 1.2, 2.4, and 1.9 cm rainfall on July 28, 29 and August 1 2 and 7 respectively. The following species groups, Cx. (Melanoconion) spp. Theobald, Anopheles quadrimaculatus Say, An. crucians Wiedemann and Aedes atlanticus/tormenter Dyar and Knab were pooled, as these taxa could not be distinguished with confidence. Results During the six trap-nights, 102,917 mosquitoes, 94 tabanids (Diachlorus ferrugatus Osten Sacken), 2,841 chaoborids (Corethrella spp.), and 26 Lutzomyia shannoni (Dyar) were collected respectively. Proportions of trap collections captured at the different light color/orientation combinations for the most common species are shown in Figures 5-2 and 5-3. Means, standard errors, p-values, and significant differences for

PAGE 96

80 the common species are shown in Table 5-1 As noted in the tables, there were significant trap-position and day effects for some species. No significant differences were observed for the total number of mosquitoes captured at the light color/orientation combinations (p = 0.21). However, there were significant differences in capture numbers for individual species including An. crucians s.l. (p=0.0001), Culiseta melanura (Coquillett) (p=0.02), Cx. (Melanoconion) spp. (p=0.03), and Uranotaenia sapphirina (Osten Sacken) (p=0.001). Anopheles crucians and Ur. sapphirina were captured in the largest numbers at the incandescent bulb and the blue LEDs oriented out. Conversely, Cs. melanura was captured in greatest numbers in the traps baited with the incandescent bulb, and blue and green LEDs oriented up. Blue LEDs "out" were most attractive for Cx. (melanoconion) spp. Although female Lutzomyia shannoni (Dyar), were collected, no significant color differences were observed given the small sample size. The tabanid, Diachlorus ferrugatus Osten Sacken, also showed no significant color preference. Chaoborids (Corethrella spp.), however showed significant color preferences (p=0.0009) for the incandescent, and both blue orientations. No colors were found to be repellent to mosquitoes when compared to the no-light controls. The mosquito species composition attracted to the incandescent light trap agree with those found by Mann (1993) and Burkett et al. (1998a). The weights of non-target insects collected in the different CDC light trap combinations showed the incandescent bulb to have significantly more (p=0.001) than any of the LED color/orientations.

PAGE 97

81 Discussion Few field studies have scientifically evaluated mosquito trap collections based on using different colors of transmitted light of known wavelength and intensity. This is surprising, considering the primary means of evaluating mosquitoes' relative abundance, presence or absence, and effectiveness of control measures is through the use of lightbaited traps. Most studies use pooled species; few report results of individual species. Significant color preferences for some species were demonstrated from trap collections using LED baited traps (Burkett et al. 1998a). In general, traps with LEDs emitting primarily green and blues wavelengths captured the most mosquitoes. In this study, the attraction to greens and blues were further evaluated by changing the orientation of the LEDs in the traps. Collections of most of the common representative species such as Ae. dupreei, Ae. atlanticus/tormentor, Ae. infirmatus, Ps. ferox, Cx. nigripalpus, Ps. ciliata, and Cq. perturbans did not differ significantly between any of the color/orientation combinations tested. Only Cs. melanura, An. crucians, Ur. sapphirina and Cx. (Melanoconion) spp. were collected in significantly greater numbers to the different trap LED color/orientations when compared to the "no light"/C02 only controls. Allan et al. (1987) stated that while crepuscular and nocturnal biting flies are unlikely to have welldeveloped color vision, their abilities to detect differences in intensity contrast are likely to be well-developed. Even though there were relatively large differences in intensity between the incandescent (520 mcd), blue (700 mcd) and green (2400 mcd) LEDs used in

PAGE 98

82 the traps, there were significant differences in trap collection for only 4 species. This may be due to the heavily vegetated and confined habitat of the cypress swamp. Greater differences in trap collections due to differences in intensity may result in more open habitats. We recommend additional work aimed at species from different habitats and in different geographical locations. Furthermore, the attraction of phlebotomine sand flies deserves additional investigation. Light-emitting diodes require significantly lower amounts of energy (ca. 1 ma/8 hrs) than incandescent bulbs (See Burkett et al. 1998a) saving both battery life and expense during use. For convenient use, LEDs could be soldered directly into the existing trap circuitry (in series) using a 1 00 and 200 ohm resistors for blue and green LEDs respectively. Future trap designs should incorporate a mixture of green and blue diodes oriented both "out and up." The use of "super-bright" LEDs warrants serious consideration as a replacement for standard incandescent bulbs used in light-baited traps. These results have potential for use in population dynamics studies or for enhancing the attractivity of certain species.

PAGE 99

83 o o CM c CM CI a s o m TJ S M a c o to 10 ) -H o id .-i CD -o n eg CM 0! in m < — < +i r: -H r7 + r(M 00 ". — i N a o Z m • o 00 CO r~ • Csl CM -H +1 to m CO o H lu 2 to J 2 h o tl 3 O £ CM ^ 10 XI XI in CM uT C J +1 TP H C +1 t00 +: ci in a c CM CO cl o CO tH CM CO CO CO CO CO o 5s s a 01 10 10 u i — 1 CM as C 1 T? o m o CO C IS H CO +i CD i +1 to o +1 +1 If) +1 +i CM +1 +1 O X> in IO O ci O rCO o H H fO to r H CD H £ H S a u cr .u m to (1 10 Lf> o 01 CI +1 CM o H +i 1 r~ 2 +1 CI CM + CD c 00 t 3 r10 CO -H -H r~ m CM CO .H CM I/) CM u TJ o CM o Q ro CD CO a o a 3 oo S 2 O : o (J o <0 CM lab TP O CI o :36.6a TP CO CM 2ab o 6dc CO 10 H 10 O rH VO c -H 3 +1 u +1 CM c +1 +1 o +1 r+i CM H in -H UT +1 ro to o o > — I -H m on +1 o +i CM c 10 10 O H CM iO CM : z ".r H r~i TP H 10 CM a io CM CM TP O -H -H m to D. P i OQ m bo 10 ^~ CI O on 3 c iO o m H <—t W +i in 1 — 1 o CM o in D CM CO Tf CO r, _l_l TJ U CO X3 (0 tP XI to CM i0 CO CO rH CM CO r 1 +1 O +1 CO 7 o H CM CI fl in TP IO rCM en TP M ^ 1 to m 1J la O XI 10 o CO 00 x z CD CM o — 1 m H -H +i m o H in O CM o o r1396. a) to 8 CO n CO 00.2 jQ 1.7a Cl o 27.7 CM -H m r* -M X' CM -H uT r: Oil CD 10 r* T rCM CO o Oa Cl O c o 10 10 -H m — i +i Cl CM +1 CO 1! CO H CM to +1 o m 6 + rCM r— i H o T — 1 CM m o 10 IX' Q o C! 10 CD t o n Cl to o 00 o CM +1 CM 0 +1 +1 C! +1 — 1 1 — 1 o o c Cl +1 O 8 + J o 'X' o Cl — 1 H o m m c C/3 r s 06 "5.x: 5 9 B 73 io S3 TT. TP. a S a o a aj i "a ? 5 eg to to i i •2 e § s: s: tc s I s si O 5 .2 O U 2§S i $ I £ 1 1 M M M M <>; i>i ,cr> y> *2 X v | C 5, C3 S i ll (A C: | I ir § l Q til s i i 1 8 33

PAGE 100

84 Fig. 5-1. Wiring assemble representation for LED orientations: a) Top view of LED assembly oriented 360 out and b) Top view of LED assemblely oriented "up." For green LEDs: 3 V with separate battery pack or if wired to trap electronics use 200 ohm resistor. For blue LEDs: 4.5 V with separate battery pack or if wired to trap electronics use 100 ohm resistor.

PAGE 101

85 II 2P <= M C „ C "> iTi 81 O 3 J3 o § II O ca I.e 00 Q. "3 in D. c3 S, so Pr a. T3 t/l £ O c 'C a i g 3 .s w 3 cr 35 C S c c E E o o Cm O a _o O a. £ o o CD o s 09 Q s a | § 2 c o 1 c CD •c O c4 ^6 2P "3? j O a CD La CD u 2 so 3 C o o CO o CNJ pejnjdeQ S9p9ds %

PAGE 102

86 O o CO o CN 4> VO •5 ii op c S > m o g.§MI MO. s O. 2 -c > c '£ , C T3 a> c 6 o 'S C/5 -E o M o c u •i. C a) •Ja £ — 13 w c £ =3 S '2 r-i in pamjdeo sepads % 2

PAGE 103

CHAPTER 6 LABORATORY EVALUATION OF COLORED LIGHT AS AN ATTRACTANT FOR FEMALE AEDES AEGYPTI, AEDES ALBOPICTUS, ANOPHELES QUADRIMACULA TUS AND CULEX NIGRIPALPUS Introduction That some species of mosquitoes and other medically important Diptera are attracted to artificial light or other visual stimuli has long been known and exploited in a variety of trap designs. Not all mosquito species respond equally to visual stimuli or to different wavelengths of light. Indeed, many mosquitoes do not respond to light traps at all (Service, 1 995). Mating, dispersal, appetitive flight, and location of sugars, hosts, resting, oviposition and overwintering sites are all governed to some degree by vision. Many authors have examined the important visual components of host/resource finding and have divided them into shape, color (reflected and transmitted), size, contrast, light intensity, texture and movement (Allan et al. 1987). These factors alone or in combination appear to play an important role in a blood feeder's ability to successfully find a suitable host or other resource. Most research on responses of haematophagous Diptera to various colors has evaluated the response of diurnal species to reflected light colors (Brett 1938, Brown 1954, Bracken et al. 1962, Barr et al. 1963, Granger 1970, Bradbury and Bennett 1974, Browne and Bennett 1980, 1981, Allan and Stoffolano 1986). Studies using colored 87

PAGE 104

88 transmitted light are few, and even fewer provide information on individual species or emit light of known wavelengths and/or intensity (Headlee 1937. Breyev 1963. Bargren and Nibley 1956, Gjullin et al. 1973, Wilton and Fay 1972, Vavra et al. 1974a, Browne and Bennett 1981). None of these studies incorporate both reflected and transmitted light. Lack of information about the attractancy of different light wavelengths for different species of mosquito is a serious void in a science where mosquito control/research operations are often largely based on the numbers and types of mosquito captured in light-baited traps. A laboratory method for the evaluation of the relationship between various light colors (wavelengths) of transmitted/reflected light and feeding preference (based on duration of feeding time in seconds) is presented herein for Aedes albopictus Skuse, Ae. aegypti (L.), Anopheles quadrimaculatus, Say (Type A) and Culex nigripalpus Theobald. Color preferences also are evaluated based on fecal speck counts beneath the illuminated artificial hosts. Fecal specks on white cards have long been used to evaluate resting site preferences and provide population indices for a variety of filth flies species (Axtell 1970). Fecal speck cards have not been used to evaluate resting site preferences for mosquitoes, but presumably, could provide an equivalent estimation of feeding or resting site preference in a confined area. Information obtained about mosquito responses to different wavelengths of light can be used to further exploit insects' attraction to artificial light and enhance our ability to conduct studies on population dynamics, species specific surveys and/or improve reduction strategies.

PAGE 105

89 Methods and Materials Visualometer and Data Collection A pie-shaped olfactometer (Butler and Katz 1987, Marin et al. 1991, Wilson et al. 1991, Butler and Okine 1995, Okine 1994) electronically quantifies insect feeding activity on 1 0 compounds simultaneously for a set time period. The apparatus (hereafter called a "visualometer") was modified to incorporate 10 different light wavelengths which illuminated from below identical attractive food choices (termed "artificial hosts"). Simplified views of the top and side of the visualometer are shown in Figs.6-la and 6lb, respectively. Each artificial host was illuminated with unique wavelengths (ca 10 nm width) produced using filtered broad spectrum white light. The mosquito feeding response (time) on the illuminated artificial hosts was measured by two methods. First, when the mosquito fed and closed a circuit, mosquito feeding time (total seconds) was computer recorded, logged, and analyzed using touch and bite contact seconds (Fig. 6-1, K and J). Second, feeding preferences were evaluated by counting the number of mosquito fecal specks present within a 5 cm circle around the artificial host/colored light combination. Ten holes drilled into the bottom of the aluminum pie-shaped arena contained the tips of fiber optic cables that emitted light upwards and illuminated the artificial hosts from below (Fig. 6lb, N). The fiber optic tips were covered with recessed interference filters (described below). As an additional attractant, C0 2 (0.5 1/min) was released through Tygon tubing (Norton Performance Plastics Corp., Akron OH), directly below each artificial host (Fig. 6-1, G) for measured time intervals of 4 seconds "on" and

PAGE 106

90 6 seconds "off. The visualometer was located in a temperature-controlled, light-proof, Faraday-cage room (Lindgren Enclosures, Model No. 18-3/5-1). Artificial Host The attractant food source used in the visualometer, termed "artificial host," consisted of fresh, citrated bovine blood mixed with agar and various feeding stimulants/attractants. The following mixture was used as the feeding stimulant/attractant: (J. F.B., unpublished data): For 133 ml: 1.66 gms agar (U.S. Biochemical Corp, Cleveland OH), 33 ml fresh citrated bovine blood, 100 ml deionize water, 7.14 mg sodium chloride, 0.38 mg potassium chloride, 0.154 mg calcium chloride dihydrate, 0.2 mg magnesium chloride hexadydrate, 0.42 mg dibasic sodium phosphate, 2.1 mg sodium bicarbonate, 0.92 mg dextrose, 0.184 mg glutathione disulfide (oxidized glutathione), adjusted to a final pH of 7.4. The blood/agar/feeding attractant mixture was placed into the "cup" on the underside of a 35 mm plastic film canister lid where it was covered with a reinforced silicone membrane (Butler et al. 1984) held in place using a 4 mm retaining ring cut from the top of the film canister. The complete artificial host was then inserted into one of the ten holes cut into the transparent plexiglass visualometer lid. Between trials, the visualometer was disassemble and washed. Artificial hosts were replaced for each replicate and new mosquitoes were used for each trial.

PAGE 107

91 Light Source and Filters The light source used a wide spectrum tungsten-halogen bulb (Sylvania, no. DNF, Danvers, MA) transmitted through fiber optic cables (RTS Industries, Gainesville FL) (Fig.6-lb, N). Seven VIS-NIR broadband ( 5 nm) interference filters (350, 400, 450, 500, 550, 600, 650 and 700 nm) (Fig. 6-lb, L) with appropriate neutral density filters (NDF)(Fig.6-lb, M) to equalize intensities were used for each wavelength tested (Oriel Instruments, Stratford CT). The "white" light from the fiber optic cable (with NDF) and no light were used as controls. Mosquito Species All trials used 1 50, five to eight day old nulliparous, non blood fed females aspirated from cages containing both male and females. Aedes albopictus, Ae. aegypti. An. quadrimaculatus, and Cx. nigripalpus were the species evaluated. All species were tested separately. Laboratory colonies maintained at USDA ARS in Gainesville, FL provided recently colonized (1995) Ae. albopictus, and specimens from a longestablished colony of An. quadrimaculatus. Aedes aegypti were reared as outlined in Gerberg (1970) and obtained from an established University of Florida departmental colony. Wild Cx. nigripalpus were reared from larva and pupae obtained from a sewage lagoon at the University of Florida Swine Research Unit. All mosquitoes were reared and maintained at 25 C, 95% RH and a 14:10 (L:D) photoperiod. All trials were run from 1600 to 0800 hrs.

PAGE 108

92 Statistical Analysis All touch/bite contact seconds and fecal specks were recorded for 8 and 1 6 hrs respectively. All species trials were analyzed using the first 4 hours of feeding activity, with the exception of Cx. nigripalpus, which analyzed the last 4 hrs of feeding times. A 10X10 Latin square design (3-way ANOVA) was used for An. quadrimaculatus. For other species, a randomized complete block (2-way ANOVAs) design with 8 to 10 replications was used. Duncan's multiple range test was used to delineate significant differences between the colored light treatment means. Differences between treatment means were considered significant at alpha = 0.10. Pearson's correlation coefficients (SAS Inst. 1995) were performed to test for relationships between feeding times and fecal specks on each of the artificial host/colored light combinations. Results With the exception of Cx nigripalpus, all species showed a period of "orientation/acclimation" lasting ca. 10-15 minutes after which mosquitoes would begin aggressively probing and feeding on the artificial hosts. Of these, Ae. albopictus was the least aggressive and consequently had the lowest over all feeding times on the different host/color combinations. The wild Cx. nigripalpus presumably still under circadian control did not begin actively feeding until about 4 hours into the trial.

PAGE 109

93 Aedes aegypti Feeding durations (Fig. 6-2) and fecal speck (Fig. 6-6) results for this species were not significant different for feeding times (n=10, p= 0.17) or fecal specks (p=0.16) respectively, for any of the colors tested. No correlation (r=0.02, p=0.84) was found between the number of fecal specks and feeding durations. Significant differences for total seconds of feeding (p=0.04) and fecal speck counts (p=0.003) were observed for different replications (day effect). Aedes albopictus This species showed significant preferences (n=10, p=0.03) for certain wavelengths of light (Fig. 6-3). Aedes albopictus fed significantly longer on yelloworange (600 nm), blue-green (500 nm), white, blue (450 nm), violet (400 nm), and black compared to other colors tested. Significant differences (p=0.001) also were found for the number of fecal specks (Fig. 6-7), with 550 nm (yellow-green) being significantly greater than the other wavelengths. The number of mosquito fecal droplets, however, did not correlate (r=0.06, p=0.55) with the feeding times at respective colors. Aedes albopictus had an overall mean ( SEM) feeding time of 244 44.2 seconds which was significantly lower than the feeding times (p=0.002) and fecal speck (p=0.0005) numbers of the other mosquito species. As with all other trials, significant differences for total feeding durations (p=0.01) and fecal speck counts (p=0.004) were observed for different replications (day effect).

PAGE 110

94 Anopheles quadrimaculatus Results of the visualometer trials (Fig. 6-4) found weakly significant differences (n=10, p=0.094) in feeding times for several of the wavelengths tested. Mosquitoes fed for a significantly longer duration on the white and no light controls over the other wavelengths. Likewise, feeding durations on all other wavelengths were significantly greater than those on 350 nm (UV). Significant feeding duration differences were found on consecutive days (p=0.07), but not for different positions (p=0.73). There were significant differences in fecal speck numbers for An. quadrimaculatus (p=0.003, Fig. 68) with the greatest number of specks occurring on 500 nm, 550 nm, 600 nm and white. The number of fecal specks differed significantly between days (p=0.0001), but not for position (p=0.35). The number of mosquito fecal droplets correlated significantly (r=0.45, p=0.0001) with the feeding times at each of the illuminated artificial hosts. Culex nigripalpus Due to lack of activity during the first 4 hours of the feeding trials, the last 4 hours (2000 2400) were analyzed. Significant color preferences (n= 8, p=0.06) were observed for this species. Culex nigripalpus (Fig. 6-5) showed significantly higher feeding times on blue green (500 nm), orange (600 nm), blue (450 nm), white, red (650), and yellowgreen (550 nm) over the other colors tested. Fecal specks (Fig. 6-9) showed significant preferences (p=0.09) for 500 and 600 nm over 700 nm and white. As with An.

PAGE 111

95 quadrimaculatus, the number of fecal specks was significantly correlated (r=0.47, p=0.0001) with feeding times at each of the illuminated artificial hosts. Discussion Considering the variation in attractiveness of different mosquito species to lightbaited traps (Huffaker and Back 1943, Bidlingmayer 1967), it is not unreasonable to expect that individual species will vary in wavelength preference. Such wavelength preferences (exhibited by behavioral responses) may or may not correspond to spectral sensitivities. For attraction to light-baited traps, intensity is considered more important than color (Barr et al. 1963). As such, many studies of color light preferences in Diptera are criticized because they fail to compensate for intensity (and/or hue) and make interpretation of the results difficult (Allan et al. 1987). These visualometer tests compensated for variations in light intensity by incorporating neutral density filters at each wavelength so that each treatment only varied by color and an accurate assessment of "color" preference could be obtained. Even so, different wavelengths may be physiologically more stimulating and result in greater behavioral responses. For mosquitoes, electroretinograph studies for determining spectral sensitivities have only been published fox Ae. aegypti (Muir et al. 1992, Snow 1971). These electroretinograph studies both provide evidence of bimodal sensitivities with a small peak at 350 nm and a large peak 550 nm. This bimodal pattern is similar to those found for tabanids (Allan et al. 1991, Smith 1986) and other insects (White 1985) and is assumed, but never tested, to be similar to the spectral sensitivities of other mosquito

PAGE 112

species. Interestingly, spectral sensitivity research has focused mainly on diurnal species that are not generally attracted to standard light-baited traps. In our visualometer trials Cx. nigripalpus is the only species commonly captured in field trials using light baited traps. Results of our trials showed none of the mosquito species tested highly attracted to the assumed peaks in spectral sensitivities (550 and 350 nm) over the other wavelengths. Peak spectral sensitivities of ca. 550 and 350 nm may serve to allow discrimination in a environment dominated by greens and blues (Lythgoe 1979), but do not necessarily correspond with attractive wavelengths. Aedes aegypti andAe. albopictus are not frequently captured in mosquito traps baited primarily with light (Service 1995). As these species are diurnal, reflected light appears to be more important in resource location than transmitted light. In general, most successful Ae. aegyptil albopictus adult traps do not use light, but rather rely on strategic placement and low reflective colors (Fay 1968, Freier and Francy 1991). The relatively small numbers ofAe. aegypti/ 'albopictus captured in light traps suggest that transmitted light is relatively unimportant in host/resource choice. Indeed, the duration of feeding times and fecal speck numbers fox Ae. aegypti did not differ significantly between any of the wavelengths tested. Duration of feeding times and fecal speck numbers for Ae. albopictus however, were significantly greater for 600 nm 500 nm, 450 nm, 400 nm and broad spectrum white light. Field trials with light emitting diodes or other sources of monochromatic light might result in similar attractive colors under field conditions. Although nocturnally active, An. quadrimaculatus is another species poorly collected by light-baited traps (Bradley 1943). In our visualometer trials, An.

PAGE 113

97 quadrimaculatus were most attracted to the strongly contrasting, no light and broad spectrum white controls followed by 550 nm light. Although ultraviolet lamps have long been known to increase the numbers of host or resource seeking mosquitoes captured at light traps (Headlee 1937, Weiss 1943, Williams et al. 1955, and Breyev 1963), 350 nm was the least attractive wavelength fox An. quadrimaculatus and most of the other species tested in our study. If the duration of feeding is a measure of attractiveness, then the feeding time results for An. quadrimaculatus differed slightly with those found in two field experiments using colored light emitting diodes which found no significant trap count differences for An. quadrimaculatus (Burkett et al. 1998a), See Chapter 5). In either case, the color of light does not appear to be important in the host/resource seeking behavior of An. quadrimaculatus based on these studies. The significant differences in color feeding durations/specks found for An. quadrimaculatus in these trials may be an artifact of the close confines of the visualometer, but are unimportant for trapping wild individuals. Correlations between the number of specks around the artificial hosts and the feeding durations shows significant, though not necessarily strong correlations for both An. quadrimaculatus and Cx. nigripalpus. This negates the possibility of using one as a predictor of the other, but does serve as a means of checking the reliability of the visualometer's touch and bite contact second (feeding time) electronics. With the latter being more effective, field trials with CDC-type light traps baited with light and those with light and C0 2 are effective at collecting Cx. nigripalpus (Nayar 1982). Field research using narrow wavelength LED's (Burkett et al. 1998a) also found

PAGE 114

98 this mosquito attracted to light traps, and in one field trial, Cx. nigripalpus was significantly attracted to green (567 nm) followed by blue (450 nm) and white over the other colors tested. This agrees with what was found in the visualometer trials for this species. Other follow-up studies (See Chapter 5) which compared the blues, greens and whites found no significant color preferences for Cx. nigripalpus. Given the weakly significant results, and general lack of supporting field data, light color is largely unimportant in host/resource acquisition for both Cx nigripalpus and An. quadrimaculatus. According to Clements (1992) mosquito excretion and defecation both produce fly specks, the former being a clear to yellowish spot produced soon after feeding; and the later being a dark speck produced several hours after feeding (rate is temperature dependent). Unfortunately, mosquitoes were not observed while in the visualometer and fly speck counts were not differentiated by color. Ae. albopictus showed no correlation between feeding durations and corresponding fecal specks, yet showed significant color preferences for both feeding time and fecal specks numbers on the different colors. Better correlation between feeding durations and fly speck numbers may have been possible if speck counts were differentiated by color within four hours of initiation of the assay. Likewise, the clear specks may have provided feeding color preferences, and the darker specks may have given resting site information. Future trials using the visualometer need to concentrate on species known to be attracted to artificial light. Information obtained about medically important mosquitoes can be used to further improve current light-based trapping methods and, ultimately,

PAGE 115

enhance studies on population dynamics, species specific surveys and improve reduction strategies.

PAGE 116

Figure 6-1 V is u a lorn etc r (a) Top view (b) Side view. (A) 430 x 5 cm dia. transparent plexiglass lid (B) Artificial host, (35 mm film canister lid) (C) 30 cm high plastic side piece with holes for plastic tubing (D) Aluminum base with holes for fiber optic cables (E) Support leg (F) Plexiglass filter support (G) 1 0 cm diameter Tygon tubing for incoming air/C02 (H) Tubing for exhaust (I) Assembly screw (J and K) Probe inserted into top of artificial host and bottom sensor fitted under artificial host (feeding mosquitoes complete circuit logged by computer), one sensor per artificial host (L) Interference (bandbass) filter (M) Neutral density filter (N) Fiber optic cable (attached to light source). Drawing not to scale.

PAGE 117

From Light Source

PAGE 118

102 O 2 3000 Ae. aegypti 350 400 450 500 550 600 650 700 No Light White Wavelength (nm) Figure 6-2. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Aedes aegypti on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0. 10, Duncan's Multiple Range Test).

PAGE 119

103 O £ 700 Ae. albopictus AB AB AB X 1 I AB AB 350 400 450 500 550 600 650 700 No Light White Wavelength (nm) Figure 6-3. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Aedes albopictus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0. 10, Duncan's Multiple Range Test)

PAGE 120

104 a o An. quadrimaculatus I AB AB AB AB AB AB I AB 350 400 450 500 550 600 650 700 No Light White Wavelength (nm) Figure 6-4. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Anopheles quadrimaculatus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0. 10, Duncan's Multiple Range Test).

PAGE 121

105 o o Cx. nigripalpus I BC <\BC AB ABC 1 BC |bc 350 400 450 500 550 600 650 700 No Light White Wavelength (nm) Figure 6-5. Duration of feeding (seconds) during a 4 hour exposure (means SEM) for Culex nigripalpus on artificial hosts illuminated with different wavelengths of light Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple Range Test).

PAGE 122

106 45 Ae. aegypti i i 350 400 450 500 550 600 650 Wavelength (nm) 700 No Light White Figure 6-6. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Aedes aegypti on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0 .10, Duncan's Multiple Range Test).

PAGE 123

107 14 12 Ae. albopictus 10 I JL JL i JL 350 400 450 500 550 600 650 700 No Light White Wavelength (nm) Figure 6-7. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Aedes albopictus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0 10, Duncan's Multiple Range Test)

PAGE 124

108 <£ 40 (A i_ I 00 An. quadrimaculatus 350 400 450 500 550 Wavelength (nm) 600 650 700 No Light White Figure 6-8. Numbers of fecal specks (means SEM) after an 1 6 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Anopheles quadrimaculatus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0.10, Duncan's Multiple Range Test)

PAGE 125

109 52 yo i_ 60 < la 0) o c D to u 0) a CO TO O 0 C 5 Cx. nigripalpus 50 40 30 20 10 AB I AB AB AB AB AB 350 400 450 500 550 600 650 700 No Light White Wavelength (nm) Figure 6-9. Numbers of fecal specks (means SEM) after an 16 hour exposure in a 5 cm circle around each artificial host/wavelength combination for Culex nigripalpus on artificial hosts illuminated with different wavelengths of light. Means within each species group with the same letter are not significantly different (alpha=0. 10, Duncan's Multiple Range Test).

PAGE 126

CHAPTER 7 MOSQUITO HOST/RESOURCE FINDING: THE IMPORTANCE OF SUGAR FEEDING AND ATTRACTION TO ARTIFICIAL LIGHT Introduction Waning public health budgets, drug resistant pathogens, pesticide resistant arthropods, exponential human population growth, deforestation and rapid global transportation are all contributors to the resurgence and maintenance of the world's most insidious arthropod-borne diseases. Employing an effective surveillance program is the primary means used by vector control specialists to anticipate, prevent, or control disease in human or domesticated animals. These surveillance programs however, are biased and limited in their ability to collect representative age classes. Nor do these programs provide an accurate species profile. A recent account of many of the commonly used mosquito sampling methods and traps is detailed by Service (1995). For most mosquitoes and many other medically importance insects, light traps, in one form or another, are the primary means of tracking vector populations. Many mosquito species do not respond well to light traps (e.g., Aedes aegypti/albopictus, Culex pipiens complex, and many anophelines). Furthermore, use of humans as bait for landing/biting collections is becoming increasingly unethical, and potentially dangerous due to the emergence of drug resistant malaria and other arthropod-borne pathogens. 110

PAGE 127

Ill Light traps collect only flying populations, and catches are influenced by light source, wavelength, intensity, trap placement, and many other intangibles (Moore and Gage 1 996). Because of the selectiveness of sampling inherent to light trap collections, surveillance programs are continually in need of improved trap designs. The ideal trap would collect representative age distributions, species composition, and if desired, larger numbers. To enhance collection efficiency, light traps are usually supplemented with additional attractants such as carbon dioxide, lactic acid, octenol, acetone, heat, water vapor and others. In some instances, light is not used, only chemical attractants. Any improvement to mosquito trapping/attractant technology represents a positive contribution to vector ecology. Depending on the trapping/surveillance objective, finding the correct attractant combination can result in improved or even selective trapping. Potential trap improvements and guidance for additional attractants research are provided within this dissertation. These are (1) the use of super bright LED's as replacements for the standard incandescent bulb used in light traps and (2) determining the attractive chemical components contained within natural sugar sources. Utilizing both of these host/resource seeking parameters into future trap designs may allow sampling of additional age classes and/or species not otherwise normally captured in light traps. A major requisite for survival and reproduction for mosquitoes and other haematophageous insects is their ability to find and exploit suitable hosts or other resources. Much has been written and research continues to progress on how mosquitoes are able to find these scattered resources. Sutcliffe (1987) for example, classifies host/resource finding into three elements: (1) Appetitive Flight: an internally driven flight

PAGE 128

112 initiated by an unmet physiological need; (2) Activation and Orientation: acquisition of external olfactory or visual stimuli from the environment (e.g. host), and (3) Attraction: goal-oriented attraction flight to the objective. This three-step paradigm, redefined by Klowden (1996) as "ranging behavior," describes how host/resource seeking behavior can be applied toward light attraction and location of sugar sources. Bidlingmayer (1994) provides additional information on resource seeking females and concludes that long range stimuli used by resource seeking females on appetitive flights are directed toward selected visual targets. This flight continues until terminated by either the start of a longrange attractant flight or a short-range response to avoid the target. A long-range attraction flight is initiated and guided by various goal associated external cues, which may be visual, olfactory, or other, that indicate the existence and direction of the goal. The attraction flight is terminated either by making contact with the goal or by a shortrange response rejecting the goal and a return to appetitive flight. The purpose of this dissertation is to evaluate aspects of visual and sugar feeding ecology and to provide a basis for the incorporation of their attributes into future trapping systems. This is accomplished by evaluating sugar feeding and artificial light attraction as two important behavioral elements in mosquito host/resource seeking. A better understanding of resource seeking behavior will be a positive contribution to vector ecology and allow the development of improved surveillance/control equipment. This dissertation contains five chapters of original research that provide reasonable evidence that mosquitoes (at least for the species studied) are opportunistic, and that dietary sugars used for flight fuel and immediate metabolic needs are obtained from a wide diversity of

PAGE 129

113 sources including homopteran honeydew. Many of these sugar sources have no detectable odor, but likely have similar volatile components warranting additional investigation. Furthermore, these dissertation results show a species-specific response (for common north central Florida woodland mosquitoes) to traps baited with different wavelengths of artificial light. In combination, these dissertation chapters are an important contribution to the science of mosquito/vector ecology. The results from both the sugar feeding and light attractancy chapters form the basis for future trap improvements used by vector control specialists and field ecologists. The wide diversity of sugar sources used by mosquitoes leads this author to conclude that these sources have common, but undiscovered, attractants. Sugar Feeding Blood feeding arthropods including mosquitoes consume blood meals which satisfy protein deficits needed to promote ovarian development (Clements 1992). With few exceptions, blood contains insufficient sugar to sustain immediate metabolic needs and must be obtained from other sources. An extensive review of mosquito sugar feeding is found in chapter 1 ; however, a few additional points not made elsewhere in this dissertation are included below. Laboratory survivorship data for sugar fed females for several of Florida's common mosquito species indicates species specific variations (Nayar and Sauerman 1975a). All species lived for about 24 hours post emergence without a sugar meal, but survival of mosquitoes fed once to repletion on 50% sugar solutions ranged from 144 (An. quadrimaculatus) to 336 hours (Ae. taeniorhynchus). In general,

PAGE 130

114 smaller mosquito species were more efficient than larger species at utilizing their sugar reserves. In a research project not included elsewhere in this dissertation, Burkett (1998, unpublished) found that survivorship length did not differ significantly for Cx. nigripalpus fed once to repletion on 10% solutions of glucose, sucrose, or melezitose. Assuming wild mosquitoes approach anything close to the 6-14 days of survival following a full sugar meal, it seems reasonable that wild mosquitoes can go at least 3-4 days without sugar feeding. Sugar feeding frequency in wild populations has likely evolved to meet the needs of specific species. Some species are known to feed at all stages of their gonotrophic cycle (Nasci and Edman 1 984, Magnarelli 1 978), but sugar feeding frequency data for wild populations species remains unavailable. Sugar feeding likely varies by species, and depends on the mosquito's physiological and environmental conditions. Natural sugar availability may limit successful host; mate, oviposition, resting sites, or other crucial ecological factors needed to maintain even a minimum level of fitness for mosquitoes. As the ancestral stock of mosquitoes must have needed sugar for these same reasons, it is logical to conclude that mosquitoes evolved and finely tuned their ability to locate sugar resources long before they became blood feeders. As many mosquito species are autogenous, their elongate stylets may have initially evolved in response to sugar feeding. The widely scattered, often rare availability of hosts and other resources in the mosquito's environment necessitates periodic sugar feeding to provide flight energy and other immediate metabolic needs. Whether the host/resource seeking insect employs the "opportunistic/passive" or the "actively search" mode of host/resource

PAGE 131

finding (O'Meara 1987), an external source of sugar for energy is required. Unlike other species, mosquitoes do not use fat reserves (triglycerides) for flight and immediate energy requirements (Nayar and Van Handel 1971, Van Handel 1984), but instead use dietary sugars and glycogen. Fat stores are used for long term survival. Adequate dietary sugar then, becomes a critical component of successful host finding. Sugar and blood feeding are mutually exclusive events, yet often share similar activity periods. As with blood-feeding, sugar-feeding for anautogenous mosquitoes is subject to circadian events and associated with metabolic needs. According to Foster (1995), the behavioral context of feeding decisions is not fully understood but can be divided into two possible paradigms. In a fixed-search paradigm, a mosquito in a particular physiological state would be locked into a dominant behavioral mode at some point before the insect encounters food stimuli. Hence, its activity would be part of an appetitive sequence with a specific goal. Alternatively, in a conditional-response paradigm, the mosquito's decision to choose among foods (simultaneous or sequential encounters) would be governed by instantaneous internal state and stimulus strength. When confronted with a choice of sugar or blood, Hancock and Foster (1993) found female Cx. nigripalpus preferred sugar when energy reserves were low and blood when reserves were high. Likewise, Yee et al. (1992) found all sugar feeding ceased in the presence of a host, but resumed once the host-stimuli was removed. Blood-host attraction should dominate earlier in the activity window but then gradually give way to sugar attraction as persistence diminishes and reserves are consumed in continued flight or unsuccessful attempts to obtain blood. This later statement is supported by Xue and

PAGE 132

116 Barnard 1997, who found sugar-starved Ae. aegypti were not as aggressive/persistent (lower avidity) then those with adequate sugar reserves. Although having lower avidity, Foster and Eischen (1987) found sugar-starved Ae. aegypti (but not for An. quadrimaculatus) resulted in a greater frequency of blood feeding over mosquitoes supplied with sufficient sugar. Several papers have linked visual and sugar feeding ecology, as visual cues may be important in finding sugars for some mosquito species. Many authors have noted a propensity for crepuscular/nocturnal mosquitoes to feed from floral nectars of background contrasting pale-colored or white flowers (Sandholm and Price 1 962, Grimstad and Defoliart 1974, Magnarelli 1977, 1978, 1979, 1983, and Gadawski and Smith 1 992). At least for north central Florida, the generalization of feeding on palecolored flowers does not appear to be true. At least from this author's field observations, pale-colored flowers were not an important source of sugar for mosquitoes endemic to the Gainesville area. Indeed, during observations made from sunset through midnight during July and August 1 997, not a single mosquito (and few other insects) was present foraging on pale-colored composites (e.g., Bidens pilosa, Erigeron quercifolius, Eupatorium capillifolium, Rudbeckia hirta, or Cnidoscolus stimulosus (Euphorbiaceae)) which were abundant in one of the study areas. Mosquitoes were only observed feeding on the extrafloral nectaries of partridge pea {Cassia occidentalis) and on the sugary exudates (apparently produced only at night) from the developing fruits of bahaigrass (Paspalum distichum). Neither of these sugar sources had detectable odors; yet photographs taken by the author show these plants were visited by a wide range of insects including several

PAGE 133

117 moths (especially noctuids and geometrids), flies, green lacewings, earwigs, ants and even German or Asian cockroaches. For Culex nigripalpus, decaying fruits from elderberry {Sambucus canadensis) were also found to be an important sugar source for males and females. Samples obtained near a primary mosquito breeding site in a grove of elderberry contained numerous males and females with distended bright red abdomens which contained elderberry juice as verified using GC. Photographs were also taken of Aedes albopictus feeding on honey dew from aphid infested corn and sorghum at the University of Florida's Santa Fe Beef Unit. These same mosquitoes were observed puncturing the corn leaves with their stylets and apparently imbibing liquid. Based on these field observations and the wide diversity/ratios of sugar species (see chapters 2 and 3) found in the crops of wild mosquitoes, one can safely conclude that mosquitoes are opportunistic feeders, and obtain sugar from a wide variety of sources. Since many of these natural sugar sources (e.g., extrafloral nectars, exudates, decaying fruit, etc.) are not dominated by strong "flowery" scents characteristic of some nectar sources, mosquitoes and the other visiting insects must cue in on olfactory stimuli which have yet to be determined. In any case, isolating the attractive sugar source components and incorporating them into traps would allow for sampling a different age structure and perhaps species composition then would otherwise be possible in a standard attractant baited or unbaited light trap. In this dissertation, gas chromatography (GC) analysis was used to evaluate the composition of sugars found in the crops of five field collected mosquito species from a variety of habitats over the course of two seasons. The percentage of mosquitoes

PAGE 134

118 containing crop sugars ranged from 10-1 1% in An. quadrimaculatus and Ps. ferox to 47.7% in Cq. perturbans. Crop sugar occurrence was significantly greater in females than males in An. quadrimaculatus (p=0.0006) and Cs. melanura (p=0.00004). These findings agree with those of other Florida mosquitoes. Bidlingmayer and Hem (1973) for example, found similar sugar positivity results using the cold anthrone test for An. quadrimaculatus (9=15%); Cs. melanura (9=36%, tf=29%); Ps. ferox (9=20%) and Cx. nigripalpus (9=17%, cf=17%). Moreover, Magnarellli (1978) obtained similar results for Cq. perturbans (9=57%). The results in this dissertation differ only in that more female An. quadrimaculatus contained sugars than males. The diverse sugar composition, proportion, and occurrence found in the crops of all species tested strongly suggests that these mosquitoes are opportunistic and that sugar is not a limiting resource. All species tested contained melezitose and/or erlose (An. quadrimaculatus (55%), Cs. melanura (33%), Cx. nigripalpus (15%), Cq. perturbans (10%), and Ps. ferox (7%)). The data shows honeydew to be an extremely important resource for both Cs. melanura and An. quadrimaculatus comprising 1/3 and 1/2 of the sugar meals respectively. These observations make sense, because unlike Cq. perturbans (Grimstad and DeFoliart 1974, 1975), Ps. ferox (Magnarelli 1980) and Cx. nigripalpus (see Nayar 1982), which have been observed feeding on a variety of natural sugar sources, An. quadrimaculatus and Cs. melanura have not been reported to feed on floral/extrafloral nectars. Percival (1961) and Van Handel et al. (1972) characterized nectar sugars and ratios for many plant species. We initially hoped that GC analysis could be used for

PAGE 135

119 determining which plant(s) or even family of plants that mosquitoes use for sugar meals based on direct comparisons with published reports. Unfortunately, unlike melezitose and other trisaccharides, sucrose is rapidly hydrolyzed into fructose and glucose in insect crops (Schaefer and Miura 1972, Burkett et al. 1998) which negates any possibility of determining the source plant of the sugar meal. One interesting observation, however, is that some crop samples from An. quadrimaculatus, Cs. melanura, and Cx. nigripalpus contained relatively large amounts of sucrose in their crops showing that either some sugar sources produce enzyme inhibiting compounds, or salivary sucrase is not always produced and shunted to the crop with sugar meals. When compared to the traditional use of the cold anthrone test, GC has proved to be an excellent and powerful tool for qualifying and quantifying the important ecological dietary parameters such as sugar source, preference, occurrence and composition of sugar feeding flies. Attraction to Artificial Light Many authors have shown that certain mosquitoes are attracted to transmitted light (Headlee 1937, Weiss 1943, Williams et al.1955, Breyev 1963, Bargren and Nibley 1956, Gjullin et al. 1973, Wilton and Fay 1972, and Browne and Bennett 1981). In conjunction with olfaction, mosquito vision plays a critical role for mosquitoes in locating mates, hosts, sugar sources, and oviposition or resting sites. Chapter 1 reviews light attractancy and the importance of vision in the ecology of mosquitoes and other medically important flies. Bidlingmayer (1994) reviews the visual responses of female

PAGE 136

120 mosquitoes during appetitive and attractive flights. His treatise doesn't address attraction to lights per se, but his ideas can be applied to attractive flights toward artificial lights. Long range orientation behavior toward artificial light is likely stimulated by the extreme contrast of an artificial light against the dark background of the mosquito's environment. Upon approaching the light source, short-range orientation behavior takes over and the mosquito is either sucked into the trap when it gets too close, or else repelled. Mosquitoes' attraction to artificial light has long been known and exploited in a variety of trap designs (Service 1995); however, the relationship between light attraction and host/resource seeking is somewhat ambiguous. As the name implies, "light traps" use light as the primary attractant for baiting insects toward a sampling device. Even though light traps have been used since the 1950's, surprisingly little work has been done to evaluate the importance the individual wavelengths comprising broad spectrum "white" bulbs typically used in light baited traps. Light, often supplemented (or not used at all) with other attractants, is an extremely important surveillance tool used by worldwide vector control agencies. Considering the importance of light traps, it remains somewhat perplexing that more work hasn't been published to establish the behavioral effects toward different wavelengths. Amazingly few studies have evaluated mosquitos' response to different wavelengths of transmitted monochromatic light, and even fewer have detailed the response of individual species. Many mosquito control programs throughout the world, and especially in the US, use light trap capture numbers to dictate control programs. The number of mosquitoes per trap used as a "spray threshold" varies depending on geographical location, species of

PAGE 137

121 interest, trap type/ set-up, and the experience of the control personnel. At a recent short course composed of mosquito personnel from around the state of Florida, I posed the question about which light traps were being used and the numbers they used as a "spray threshold." Few of the programs around the state used the same traps, procedures or threshold numbers. Similar experiences were had while reviewing various Air Force pest management programs. Often, little thought is given to the quality or consistency of "light" used in light traps and comparison with other programs is apparently unimportant. For some programs, any bulb found around the shop sufficed. According to Allan et al. (1987), crepuscular and nocturnal biting flies are unlikely to have well-developed color vision, but their abilities to detect differences in intensity contrast are likely to be well developed. Spectral sensitivity (i.e., relative sensitivity of the retina to light of different wavelengths) studies consistently showed most flies possess a bimodal spectral response with peaks around 340 and 525 nm (White 1985). Considering the variation in mosquito species attractiveness to light traps (Huffaker and Back 1943, Bidlingmayer 1967), it is reasonable to expect differences in color preference based on variations in their spectral sensitivities. Alternatively, attractiveness may not be due to color per se. Different wavelengths may be physiologically perceived as more intense and subsequently more attractive. Barr et al. (1963) concluded that more intense light (up to a point) is more attractive than less intense light. For those species attracted to light, diodes emitting blue (450 50 nm) and green (567 50 nm) were most effective in collecting wild mosquitoes. Uranotaenia

PAGE 138

122 sapphirina, Anopheles crucians, and Psorophora columbiae, were significantly attracted to the blue over the other colors (excluding the incandescent bulb). Likewise, Culiseta melanura, and Culex nigripalpus, were captured in the greatest numbers at green and orange, and green respectively. Spectral sensitivity studies using the laboratory rat of medical entomology, Ae. aegypti (Muir et al. 1992, Snow 1971), and other Diptera (Allan et al. 1 991 Smith 1986, and White 1985) show bimodal spectral sensitivities peaking around blue (350 nm) and green (550 nm) regions of the electromagnetic spectrum. Unfortunately, all mosquito spectral sensitivity studies have focused on Ae. aegypti, and none have been conducted on nocturnal or other species actually commonly attracted to artificial light. It remains unclear as to whether mosquitoes attracted to the traps are responding to "color," or if those colors possess a greater perceived intensity and hence attractive from a greater distance. Unlike standard incandescent bulbs which radiate light in a 360 pattern, LEDs emit narrow beams (8-32). In chapter 5, trap numbers based on LED orientation were compared using LEDs oriented up and reflecting off the shiny aluminum pan covering the CDC trap with those oriented out in a 360 pattern. Because of the narrow, but bright beams of light produced by the LED's, those oriented "out" would be visible from a farther distance than those oriented "up." As such, one would expect trap capture numbers to be significantly greater in the traps having the LED "out" orientation. Surprisingly, of the 1 1 common species collected, only An. crucians and Ur. sapphirina were significantly attracted to blue LED's oriented "out." over the other colors and orientations. Future studies and trap designs should focus on combinations of blues and greens oriented in different directions. Hopefully, future technology will

PAGE 139

123 produce "super-bright" LEDs or other high efficiency lights capable of producing wavelengths between 450 and 550 nm, as these may produce excellent results. Summary Incorporating the practical elements of visual and sugar feeding ecology into future trapping systems is the primary goal of this dissertation. This is accomplished by examining the behavioral response of mosquitoes to different wavelengths of light and by evaluating sugar meal composition of wild mosquito populations and laboratory colonies. Both light attraction and sugar feeding are important components of host/resource seeking and deserve to be included in one dissertation. Ecologically, different wavelengths of light (mostly blues and greens) were found to be most attractive to those species attracted to light. Attraction to these wavelengths corresponds to spectral sensitivity peaks found in most dipteran electroretinograph (ERG) studies. Mosquitoes and other insects attracted to light probably evolved photo receptors sensitive to these colors (ca. 550 and 350 nm) because they live in an environment dominated by greens and blues (Lythgoe 1979). Insects must be able to discriminate among similarly colored objects if they are to find the hosts and sugars they require for survival. In terms of practical use, LED's are a viable option to use as an attractive light source. Their use reduces many of the logistical problems associated with typical surveillance programs. Blue and/or green LED's collect nearly as many mosquitoes (both in numbers and species composition), are more attractive to some species, collect less non target insects, are highly efficient providing up to two times the battery use length when compared to a

PAGE 140

124 standard C-47 incandescent bulb. As for the ecological nature of sugar feeding, mosquitoes were found to contain a diversity of sugars and feed on a wide variety of sources and appeared to be opportunistic. All species were found to feed on honeydew and over 50% of An. quadrimaculatus containing crop sugars fed on honeydew. From a practical stand point (at least for the species studies), sugars did not seem to be a limiting resource from which control practices could be developed. Yuval and Warburg (1989) for example controlled phlebotomine sand flies by spraying host plants and animal burrows with Bacillus thuringiensis var. israelensis mixed with a sugar solution. Mosquitoes must find sugar sources in the wild even though they are diverse and widely dispersed. Of the primary sugar sources discovered during the course of these investigations, none had an odor detectable by humans. The olfactory stimuli used by mosquitoes and other insects to locate sugar sources are unknown but likely have common chemical components yet to be identified. Future research should focus on isolating these compounds and incorporating them into traps. This would allow vector control personnel and vector ecologists to sample additional species (perhaps even autogenous ones) and age classes not previous available.

PAGE 141

REFERENCES CITED Agee, H. R., and R. S. Patterson. 1983. Spectral sensitivity of stable, face, and horn flies and behavioral responses of stable flies to visual traps (Diptera: Muscidae). Environ. Entomol. 12:1823-1828. Allan, S. A., and J. G. Stoffolano. 1986. Effects of hue and intensity on visual attraction of adult Tabanus nigrovittatus (Diptera: Tabanidae). J. Med. Entomol. 23:83-91. Allan, S. A., J. F. Day, and J. D. Edman. 1987. Visual ecology of biting flies. Ann. Rev. Entomol. 32:297-316. Allan, S. A., J. G. Stoffolano, and R. R. Bennett. 1991 Spectral sensitivity of the horse fly Tabanus nigrovittatus (Diptera: Tabanidae). Can. J. Zool. 69:369-374. Alexander, B. J. 1988. Biology of Phlebotomine sand flies (Diptera: Phychodidae) in a focus of cutaneous leishmaniasis in northeastern Columbia. Ph.D. dissertation, University of Florida, Gainesville. 290 pp. Andersson, I. H. 1990. Nectar feeding activity of Aedes mosquitoes, with special reference to Aedes communis females. J. Am. Mosq. Control Assoc. 6:482-489. Andersson, I. H., and T. G. T. Jaenson. 1987. Nectar feeding by mosquitoes in Sweden, with special reference to Culex pipiens and Culex. torrentium. Med. Vet. Entomol. 1 :59-64. Auclair, J. L. 1963. Aphid feeding and nutrition. Ann. Rev. Entomol. 8:439-90. Axtell, R. C. 1970. Integrated fly-control program for caged poultry houses. J. Econ. Entomol. 63:400-407. Baker, H. G., and I. Baker. 1973. Some anthecological aspects of the evolution of nectarproducing flowers, particulary amino acid production in nectar, pp. 243-64. In V. H. Hey wood. [ed.]. Taxonomy and Ecology. Academic Press, London. Baker, H. G., and I. Baker. 1983a. A brief historical review of the chemistry of floral nectar, pp. 126-152. In B. Bentley and T. Elias [eds.]. The biology of nectaries. Columbia University Press, NY. 125

PAGE 142

126 Baker, H. G., and I. Baker. 1983b. Floral nectar sugar constituents in relation to pollinator type. pp. 117-141. In C. E. Jones and R. J. Little [eds.]. Handbook of experimental pollination biology. Van Nostrand Reinhold Co. Inc., New York. Baker, H. G., P. A. Opler, and I. Baker. 1978. A comparison of the amino acid complements of floral and extrafloral nectars. Bot. Gaz. 139:322-332. Bargren, W. C, and C. Nibley. 1956. Comparative attractiveness of colored lights of equal intensity to specific species of mosquitoes. Research report of the Third Army Area Medical Laboratory, SU 3004, Ft. McPherson, GA. 40 pp. Barr, A. R., T. A. Smith, M. Boreham, and K. E. White. 1963. Evaluation of some factors affecting the efficiency of light traps for collecting mosquitoes. J. Econ. Entomol. 56:123-127. Bates, R. B., D. N. Byrne, V. V. Kane, W. B. Miller, and S. R. Talor. 1990. N.M.R. characterization of trehalulose from the excrement of the sweet potato whitefly, Bemisia (abaci. Carb. Res. 201:342-345. Bidlingmayer, W. L. 1967. A comparison of trapping methods for adult mosquitoes: species response and environmental influence. J. Med. Entomol. 4:200-220. Bidlingmayer, W. L. 1994. How mosquitoes see traps: Role of visual responses. J. Amer. Mosq. Asso. 10:288-296 Bidlingmayer, W. L., and D. G. Hem. 1973. Sugar feeding by Florida mosquitoes. Mosq. News 33:535-538. Bowen, M. F. 1992. Patterns of sugar feeding in diapausing and nondiapausing Culex pipiens (Diptera: Culicidae) females. J. Med. Entomol. 29:843-849. Bracken, G. K., W. Hanec, and A. J. Thorsteinson. 1962. The orientation of horseflies and deerflies (Tabanidae: Diptera). II. The role of some visual factors in the attractiveness of decoy silhouettes. Can. J. Zool. 40:685-695. Bradbury, W. C, and G. F. Bennett. 1974. Behavior of adult Simuliidae (Diptera). I. Response to color and shape. Can. J. Zool. 52:251-259. Bradley, G. H. 1943. Determination of densities of Anopheles quadrimaculatus on the wing. Proc. 30th Ann. Mtg. New Jersey Mosquito Exterm. Assoc. pp. 22-27.

PAGE 143

127 Breeland, S. G., and E. Pickard. 1961. Observations on mosquito feeding activity on the flower heads of Eupatorium and Solidago (Compositae). Mosq. News 21 :32-34. Breeland, S. G., and E. Pickard. 1967. Field observations on twenty-eight broods of floodwater mosquitoes resulting from controlled floodings of a natural habitat in the Tennessee Valley. Mosq. News 27:343-358. Brett, G A. 1938. On the relative attractiveness to Aedes aegypti of certain coloured cloths. Trans. R. Soc. Trop. Med. Hyg. 32:1 13-124 Breyev, K. A. 1963. The effect of various light sources on the numbers and species of blood-sucking mosquitoes (Diptera: Culicidae) collected in light traps. Entomol. Rev. 42:155-168. Briegel, H., and C. Kaiser. 1973. Life-span of mosquitoes (Culicidae: Diptera) under laboratory conditions. Gerontologia. 19:24-49. Brown, A. W. A. 195 1 Studies on the responses of the female Aedes mosquito. Part IV. Field experiments Canadian species. Bull. Entomol. Res. 42:575-82. Brown, A. W. A. 1954. Studies on the responses of the female Aedes mosquito. Part VI. The attractiveness of coloured cloths to the Canadian species. Bull. Entomol. Res. 45:67-78. Browne, S. M., and G. F. Bennett. 1980. Color and shape as mediators of host-seeking responses of simuliids and tabanids (Diptera) in the Tantramar marshes, New Burnswick, Canada. J. Med. Entomol. 17:58-62. Browne, S. M., and G. F. Bennett. 1981. Responses of mosquitoes (Diptera: Culicidae) to visual stimuli. J. Med. Entomol. 18:505-521. Burgin, S. G, and F. F. Hunter. 1997. Nectar versus honeydew as sources of sugar for male and female black flies (Diptera: Simuliidae). J. Med. Entomol. 34:605-608. Burkett, D. A., J. F. Butler, and D. L. Kline. 1998a. Field evaluation of colored light emitting diodes as attractants for woodland mosquitoes and other Diptera in north central Florida. J. Amer. Mosq. Cont. Asso. In press. Burkett, D. A., D. A. Carlson, and D. L. Kline. 1998b. Analysis of sugar meal composition of wild mosquitoes by gas chromatography. J. Amer. Mosq. Cont. Asso. In press.

PAGE 144

128 Butler, J. F., W. R. Hess, R. G. Endris, and K. H. Holscher. 1984. In vitro feeding of Ornithodoros ticks for rearing and assessment of disease transmission, pp. 1 0751081. In D. A. Giffiths and C. E. Bowman [eds.]. Acarology VI. Vol. 2. J. Wiley & Sons, NY. Butler, J. F., and I. Katz. 1987. Process for determination of repellency and attractancy. International Flavors & Fragrances Inc. NY, and University of Florida, Gainesville. USPN: 4,759,228. Appl. No. 114,424. Butler, J. F., and J. S. Okine. 1995. Development of attractants and repellents for future management of pest fly populations, pp. 1 17-126. In Proceedings of nuisance concerns in animal manure management: odors and flies. March 1995, Gainesville FL. Byrne, D. N., and W. B. Miller. 1990. Carbohydrate and amino acid compostion of phloem sap and honeydew produced by Bemisia [abaci. J. Insect Physio. 36:433439. Chang, F., R. N. Winters, R. I. Vargas, S. L. Montgomery, and J. M. Takara. 1977. An analysis of the crop sugars in the Oriental Fruit Fly Dacus dorsalis Hendel (Diptera: Tephritidae) in Hawaii, and correlation with possible food sources. Proc. Hawaiian Entomol. Soc. 22:461-468. Christophers, S. R. 1960. Aedes aegypti, the yellow fever Mosquito. Cambridge University Press, Cambrdge. Clements, A. N. 1992. The Biology of Mosquitoes, Vol. 1. New York: Chapman & Hall. 509 pp. Day, J. F., J. D. Edman, and T. W. Scott. 1994. Reproductive fitness and survivorship of Aedes aegypti (Diptera: Culicidae) maintained on blood, with field observations from Thailand. J. Med. Entomol. 31 :61 1-617. Downes, W. L. Jr., and G. A. Dahlem. 1987. Keys to the evolution of Diptera: role of Homoptera. Environ. Entomol. 16:847-854. Edman, J. D., D. Strickman, P. Kittayapong, and T. W. Scott. 1992. Female Aedes aegypti (Diptera: Culicidae) in Thailand rarely feed on sugar. J. Med. Entomol. 29:1035-1038.

PAGE 145

129 Eischen, F. A., and W. A. Foster. 1983. Life span and fecundity of adult female Aedes aegypti (Diptera: Culicidae) fed aqueous extracts of pollen. Ann. Entomol. Soc. Am. 76:661-663. Eliason, D. A. 1963. Feeding adult mosquitoes on solid sugars. Nature. 200:289. Ewart, W. H., and R. L. Metcalf. 1956. Preliminary studies of sugars and amino acids in the honeydews of five species of coccids feeding on citrus in California. Ann. Entomol. Soc. Am. 49:441-447. Fay, R. W. 1968. A trap based on visual responses for Aedes aegypti. Mosq. News 30:20-23. Foster, W. A. 1995. Mosquito sugar feeding and reproductive energetics. Ann. Rev. Entomol. 40:443-474. Foster, W. A., and F. A. Eischen. 1 987. Frequency of blood-feeding in relation to sugar availability in Aedes aegypti and Anopheles quadrimaculatus (Diptera: Culicidae). Ann. Entomol. Soc. Am. 80:103-108. Foster, W. A., and R. G. Hancock. 1994. Nectar-related olfactory and visual attractants for mosquitoes. J. Am. Mosq. Cont. Assoc. 10:288-296. Freier, J. E., and D. B. Francy. 1991 A duplex one trap for the collection of adult Aedes albopicutus. J. Amer. Mosq. Cont. Asso. 7:73-78. Friedman, S. 1985. Intermediary metabolism, pp. 472-475. In M. Blum. [ed.]. Fundamentals of insect physiology. John Wiley and Sons, NY. Friend, W. G. 1981. Diet destination in Culiseta inornata: effect of feeding conditions on the response to ATP and sucrose. Ann. Entomol. Soc. Am. 74:151-154. Friend, W. G., J. M. Schmidt, J. J. B. Smith, and R.J. Tanner. 1988. The effect of sugar on ingestion and diet destination in Culiseta inornata. J. Insect Physiol. 34:955961. Friend, W. G., J. J. B. Smith, J. M. Schmidt, and R. J. Tanner. 1989. Ingestion and diet destination in Culiseta inornata: responses to water, sucrose and cellobiose. Physiol. Entomol. 14:137-146. Gadawski, R. M., and S. M. Smith. 1992. Nectar sources and age structure in a population of Aedes provocans (Diptera: Culicidae). J. Med. Entomol. 29:879886.

PAGE 146

130 Galun, R., and G. Fraenkel. 1957. Physiological effects of carbohydrates in the nutrition of a mosquito, Aedes aegypti and two flies, Sarcophaga bulaata and Musca domestica. J. Cell. Comp. Physiol. 50:1-23 Gerberg, E. G. 1970. Manual for mosquito rearing and experimental techniques. Bull. Am. Mosq. Cont. Asso. No. 5. 109 pp. Gibson, G. 1995. A behavioral test of the sensitivity of a nocturnal mosquito, Anopheles gambiae, to dim white, red, and infra-red light. Physiol. Entomol. 20:224-228. Gilbert, I. H., and H. K. Gouck. 1957. Influence of surface color on mosquito landing rates. J. Econ. Entomol. 50:678-680. Gjullin, C. M., D. G. Brandl, and J J. O'Grady. 1973. The effect of colored lights and other factors on the numbers of Culex pipiens quinquefasciatus, Culex tarsalis and Aedes sierrensis entering light traps. Mosq. News 33:67-71. Gooding, R. H. 1975. Digestive enzymes and their control in haematophagous arthropods. Acta Trop. 32:96-1 11. Granger, C. A. 1970. Trap design and color as factors in trapping the salt marsh greenhead fly. J. Econ. Entomol. 63:1670-1672. Gray, H. E., and G. Fraenkel 1953. Fructomaltose, a recently discovered trisaccharide isolated from honeydew. Science 1 18:304-305. Grimstad, P. R., and G. R. DeFoliart. 1974. Nectar sources of Wisconsin mosquitoes. J. Med. Entomol. 11:331-341. Grimstad, P. R., and G. R. DeFoliart. 1975. Mosquito nectar feeding in Wisconsin in relation to twilight and microclimate. J. Med. Entomol. 1 1 :691-698. Gui, H. L., L. C. Porter, and G. F. Prideaux. 1942. Response of insects to color, intensity, and distribution of light. J. Amer. Soc. Agric. Engineer. 23:51-58. Haeger, J. S. 1955. The non-blood feeding habits of Aedes taeniorhynchus (Diptera: Culicidae) on Sanibel Island, Florida. Mosq. News 15:21-26.

PAGE 147

131 Hancock, R. G., and W. A. Foster. 1993. Nectar or blood? A study of juvenile hormone, energy reserves, and odor preference in Culex nigripalpus mosquitoes, pp. 294301. In D. Borovsky and A. Spielman [eds.]. Host regulated developmental mechansims in vector arthropods. Proc. 3rd symp. Vero Beach: University of Florida-IFAS, Gainesville. Harada, F., K. Moriya, and T. Yabe. 1976. Observations on the survival and longevity of Culex and Aedes mosquitoes fed on the flowers of nectar plants (IV supplement). Jap. J. Sanit. Zool. 27:307-309. Haramis, L. D., and W. A. Foster. 1990. Gonotrophic cycle duration, population age structure, and onset of sugar feeding and insemination of Aedes triseriatus (Diptera: Culicidae). J. Med. Entomol. 27:421-428. Headlee, T. J. 1937. Some facts underlying the attraction of mosquitoes to sources of radiant energy. J. Econ. Entomol. 30:309-312. Hendrix, D. L., and Y. A. Wei. 1994. Bemisiose: An unusual trisaccharide in Bemisia honeydew. Carbohydr. Res. 253:329-334. Herms, W. B., and J. K. Ellsworth. 1934. Field tests of the efficacy of colored light in trapping insect pests. J. Econ. Entomol. 28:1055-1067. Hocking, B. 1953. The intrinsic range and speed of flight of insects. Trans. R. Entomol. Soc. London 104:223-345. Hocking, B. 1968. Insect-flower associations in the high arctic with special reference to nectar. Oikos. 19:359-387. Holliday-Hanson, M. L., B. Yuval, and R. K. Washino. 1997. Energetics and sugarfeeding of field collected anopheline females. J. Vec. Ecol. 22:883-89. Hoppe, K. L. 1983. Natural sugars detected in the crops of two species of horse flies (Diptera: Tabanidae) and the hydrolysis of sugars in the crop and in salivary gland homogenates. M.S. thesis. Oklamoma State University. 201 pp. Hudson, C. S. 1946. Melezitose and turanose. Advances in Carbohydrate Chemistry. 2:236. Huffaker, C. B., and R. C. Back. 1943. A study of methods of sampling mosquito populations. J. Econ. Entomol. 36:561-562

PAGE 148

132 Jepson, P. C, and T. P. Healy. 1988. The location of floral nectar sources by mosquitoes: an advanced bioassay for volatile plant odours and initial studies with Aedes aegypti (L.) (Diptera: Culicidae). Bull Entomol. Res. 78:641-650. Joseph, S. R. 1970. Fruit feeding of mosquitoes in nature. Proc. Ann. Meeting N. J. Mosq. Exterm. Assoc. 57:125-131. Kelly, R., and J. D. Edman. 1996. Infection and transmission of Plasmodium gallinaceum (Eucoccida: Plasmodiidae) in Aedes aegypti (Diptera: Culicidae): effect of preinfection sugar meals and postinfection blood meals. J. Vec. Ecol. 22:36-42. Kevan, P. G., and H. G. Baker. 1983. Insects as flower visitors and pollinators. Ann. Rev. Entomol. 28:407-453. Killick-Kendrick, R., and M. Killick-Kendrick. 1987. Honeydew of aphids as a source of sugar for Phlebotomus ariasi Med. Vet. Entomol. 1:297-302. Kline, D. L., W. Takken, J. R. Wood, and D. A. Carlson. 1990. Field studies of the potential of butanone, carbon dioxide, honey extract, l-octen-3-ol, L-lactic acid and phenols as attractants for mosquitoes. Med. Vet. Entomol. 4:383-391. Klowden, M. J. 1986. Effect of sugar deprivation on the host-seeking behavior of gravid Aedes aegypti mosquitoes. J. Insect Physiol. 32:479-483. Klowden, M. J. 1996. Vector behavior, pp. 34-50. In B. J. Beaty and W. C. Marquardt [eds.]. The biology of disease vectors. University of Colorado Press, Niwot, CO. Koptur, S. 1992. Extrafloral nectar-mediated interactions between insects and plants, pp. 81-129. In E. A. Bernays [ed.]. Insect-Plant Interactions, CRC. Press. Boca Raton, FL. Laarman, J. J. 1955. The host-seeking behavior of the malaria mosquito, Anopheles maculipennis atroparvus. Acta Leiden 25:1-144. Laarman, J. J. 1958. The intake of sugars by females of wild Anopheles gambiae and Anopheles funestus. Acta Leiden 36:137-144. Lanza, J., and B. R. Krauss. 1984. Detection of amino acids in artificial nectars by two tropical ants, Letothorax and Monomorium. Oecologia 63:423-425.

PAGE 149

133 Lombard, A., M. Buffa, A. Patetta, A. Manino, and F. Marietta. 1987. Some aspects of the carbohydrate composition of callaphidid honeydew. J. Apic. Res. 26:233-237. Lythgoe, J. N. 1979. The Ecology of Vision. Oxford: Clarendon. 244 pp. MacVicker, J. A. K.. J. S. Moore, D. H. Molyneux, and M. Maroli. 1990. Honeydew sugars in wild-caught Italian phlebotomine sand flies (Diptera: Psychodidae) as detected by high performance liquid chromatography. Bull. Entomol. Res. 80:339-344. Magnarelli, L. A. 1977. Nectar feeding by Aedes sollicitans and its relation to gonotrophic activity. Environ. Entomol. 6:237-242. Magnarelli, L. A. 1978. Nectar-feeding by female mosquitoes and its relation to follicular development and parity. Med. Entomol. 14:527-530. Magnarelli, L. A. 1979. Diurnal nectar feeding of Aedes cantator and Ae. solicitans (Diptera: Culicidae). Environ. Entomol. 8:949-955. Magnarelli, L. A. 1980. Bionomics of Psorophora ferox (Diptera: Culicidae): seasonal occurrence and acquisition of sugars. J. Med. Entomol. 17:328-332. Magnarelli, L. A. 1983. Nectar sugars and caloric reserves in natural populations of Aedes canadensis and Aedes stimidans (Diptera: Culicidae). Environ. Entomol. 12:1482-1486. Magnarelli, L. A., and J. F. Anderson. 1977. Follicular development in salt marsh Tabanidae (Diptera) and incidence of nectar feeding with relation to gonotrophic activity. Ann. Entomol. Soc Am. 70:529-533. Mann, M. O. 1993. Studies of the activity of adult Culiseta melanura (Coquillett) mosquitoes in Florida. Ph.D. dissertation. University of Florida, Gainesville. 1 60 pp. Marin, A. B., C. B. Warren, and J. F. Butler. 1991. Method for repelling Aedes aegypti using 3,7 dimethyl-6-octenenitrile and/or2(3,3-dimethyl-2norbornylidene)ethanol-l. International Flavors and Fragrances Inc. NY, and University of Florida, Gainesville. USPN: 5,134,892. Appl. No. 157,403. Marinotti, O., and A. A. James. 1990. An alpha-glucosidase in the salivary glands of the vector mosquito, Aedes aegypti. Insect Biochem. 20:619-623.

PAGE 150

134 Marinotti, O., A. A. James, and J. M. C. Ribeiro. 1990. Diet and salivation in female Aedes aegypti mosquitoes. J. Insect Physiol. 36:545-548. McCrae, A. W. R., P. F. L. Borham, and Y. Ssenkubuge. 1976. The behavioral ecology of host selection in Anopheles implexus (Theobald) (Diptera: Culicidae). Bull. Entomol. Res. 66:587-631. Mogi, M., and I. Miyagi. 1989. Sugar feeding of Topomyia pseudobarbus (Diptera: Culicidae) in nature. J. Med. Entomol. 26:370-371. Moore, C. G., and K. L. Gage. 1996. Collecting methods for vector surveillance, pp. 471-491. In B. J. Beaty and W. C. Marquardt [eds.]. The biology of disease vectors. University of Colorado Press, Niwot, CO. Moore, J. S., T. B. Kelly, R. Killick-Kendrick, M. Killick-Kendrick, K. Wallbanks, and D. H. Molyneux. 1987. Honeydew sugars in wild-caught Phlebotomus ariasi detected by high performance liquid chromotography (HPLC) and gas chromatography (GC). Med. Vet. Entomol. 1 :427-434. Muir, L. E., M. J. Throne, and B. H. Kay. 1992. Aedes aegypti (Diptera: Culicidae) vision: spectral sensitivity and other perceptual parameters of the female eye. J. Med. Entomol. 29:278-281. Nasci, R. S., and J. D. Edman. 1984. Culiseta melanura (Diptera: Culicidae): population structure and nectar feeding in a freshwater swamp and surrounding areas in southeastern Massachusetts USA. J. Med. Entomol. 21:567-572. Nayar, J. K. 1982. Bionomics and physiology of the Culex nigripalpus (Diptera: Culicidae) of Florida: An important vector of diseases. Tech. Bull. 827. FL Ag. Exper. St., University of Florida, Gainesville. Nayar, J. K., and D. M. Sauerman. 1971a. The effects of diet on life-span, fecundity and flight potential of Aedes taeniorhynchus adults. J. Med. Entomol. 8:506-513. Nayar, J. K., and D. M. Sauerman. 1971b. Physiological effects of carbohydrates on survival, metabolism, and flight potential of female Aedes taeniorhynchus. J. Insect Physiol. 17:2221-2233. Nayar, J. K, and D. M. Sauerman. 1975a. The effects of nutrition on survival and fecundity in Florida mosquitoes. Part 1. Utilization of sugar for survival. J. Med. Entomol. 12:92-98.

PAGE 151

135 Nayar, J. K., and D. M. Sauerman. 1975b. The effects of nutrition on survival and fecundity in Florida mosquitoes. Part 3. Utilization of blood and sugar for fecundity. J. Med. Entomol. 12:220-225. Nayar, J. K., and E. Van Handel. 1971 The fuel for sustained mosquito flight. J. Insect Physiol. 17:471-481. Nielsen, E. T., and H. Greve. 1950. Studies on the swarming habits of mosquitoes and other Nematocera. Bull. Entomol. Res. 41:227-238. O'Gower, A. K. 1963. Environmental stimuli and the oviposition behaviour of Aedes aegypti var. Queenslandenis Theobald (Diptera: Culicidae). Anim. Behav. 11:189-197. Okine, J. S. 1994. Aspects of oogenesis in the horn fly, Haematobia irritans (Linnaeus) (Diptera: Muscidae). Ph.D. dissertation, University of Florida. Gainesville. 130 pp. O'Meara, G. F. 1987. Nutritional ecology of blood-feeding Diptera. pp. 741-764. In F. Slansky and J. G. Rodriguez, [eds.]. Nutritional ecology of insects, mites, and spiders. John Wiley & Sons, Inc., NY. Owen, D. F. 1978. Why do aphids synthesize melezitose? Oikos. 31:264-267. Pappas, L. G, and J. R. Larsen. 1978. Gustatory mechanisms and sugar-feeding in the mosquito, Culiseta inornata. Physiol Entomol. 3:115-1 19. Patterson, R. S., B. J. Smittle, and R. T. De Neve. 1969. Feeding habits of male southern house mosquitoes on P32-labeled and unlabeled plants. J. Econ. Entomol. 62:1455-1458. Percival, M. S. 1961 Types of nectar in angiosperms. New Phytol. 60:235-281 Pimentel, G. E., and P. A. Rossignol. 1990. Age dependence of salivary bacteriolytic activity in adult mosquitoes. Comp. Biochem. Physiol Ser. 96:549-551. Pfimmer, T. R. 1957. Response of insects to different sources of black light. J. Econ. Entomol. 50:801-803. Reisen, W. K., P. P. Meyer, and M. M. Milby. 1986. Patterns of fructose feeding by Culex tarsalis (Diptera: Culicidae). J. Med. Entomol. 23:366-373.

PAGE 152

136 Rossignol, P. A. and A. M. Lueders. 1986. Bacteriolytic factor in the salivary glands of adult Aedes aegypti mosquitoes. Comp. Biochem. Physiol. Ser. B 83:819-822. Salama, H. S. 1967. Nutritive values and taste sensitivity to carbohydrates for mosquitoes. Mosq. News 27:32-35. Sandholm, H. A., and R. D. Price. 1962. Field observations on the nectar-feeding habits of some Minnesota mosquitoes. Mosq. News 22:346-349. SAS Institute. 1995. SAS/STAT user's manual, ver. 6.03. SAS Institute, Cary, NC. Schaefer, C. H., and T. Miura. 1972. Sources of energy utilized by natural populations of the mosquito, Culex tarsalis, for overwintering. J. Insect Physiol. 18:797-805. Schlein, Y., and G. Muller. 1995. Assessment of plant tissue feeding by sand flies (Diptera: Pyschodidae) and mosquitoes (Diptera: Culicidae). J. Med. Entomol. 32:882-887. Schmidt, J. M., and W. G. Friend. 1991. Ingestion and diet destination in the mosquito Culiseta inornata: effects of carbohydrate configuration. J. Insect Physiol. 37:817-828. Service, M. W. 1995. Mosquito ecology: field sampling methods. Chapman and Hall, London. Smith, W. C. 1986. The retina of the Tabanidae (Diptera) compound eye: an ultrastructural and electrophysiological study. M.S. thesis, University of Florida, Gainesville. 406 pp. Smith, S. S., and R. M. Kurtz. 1994. The use of an anthrone reagent to detect sugar meals and their presence in the mosquito Aedes triseriatus (Diptera: Culicidae). Great Lakes Entomol. 27:49-55. Snow, W. F. 1971 The spectral sensitivity of Aedes aegypti (L.) at oviposition. Bull. Entomol. Res. 60:683-696. Sutcliffe, J. F. 1987. Distance orientation of biting flies to their hosts. Insect Sci. Appl. 8:611-616. Tarczynski, M. C, D. N. Byrne, and W. B. Miller. 1992. High performance liquid chromatography analysis of carbohydrates of cotton-phloem sap and of honeydew produced by Bemisisa tabaci feeding on cotton. Plant Physiol. 98:753-756.

PAGE 153

137 Taylor, R. M., and W. A. Foster. 1996. Spider nectarivory. Am. Entomol. 42:82-86. Trembley, H. L. 1952. The distribution of certain liquids in the esophageal diverticula and stomach of mosquitoes. Am. J. Trop. Med. Hyg. 1:693-710. Van Handel, E. 1967. Determination of fructose and fructose yielding carbohydrates with cold anthrone. Anal. Biochem. 19:193-194. Van Handel, E. 1972. The detection of nectar in mosquitoes. Mosq. News 32:458. Van Handel, E. 1984. Metabolism of nutrients in the adult mosquito. J. Am. Mosq. Cont. Assoc. 44:573-579. Van Handel, E., J. S. Haeger and C. W. Hansen. 1972. The sugars of some Florida nectars. Am. J. Bot. 59:1030-1032. Van Handel, E., J. D. Edman, J. F. Day, T. W. Scott and G. G. Clark. 1994. Plant-sugar, glycogen, and lipid assay of Aedes aegypti collected in urban Puerto Rico and rural Florida. J. Am. Mosq. Control Assoc. 10:149-153. Vargo, A. M., and W. A. Foster. 1982. Responsiveness of female Aedes aegypti (Diptera: Culicidae) to flower extracts. J. Med. Entomol. 19:710-718. Vargo, A. M., and W. A. Foster. 1984. Gonotrophic state and parity of nectar-feeding mosquitoes. Mosq. News 44:6-10. Vavra, R. W., R. L. Frommer, R. R. Carestia, F. L. Harding, and D. D. Linehan. 1974a. Evaluation of new light sources as mosquito attractants. U.S. Army Land Warfare Laboratory. Aberdeen Proving Ground, MD. Tech. Rep. No. 74-35. Vavra, R. W., R. R. Carestia, R. L. Frommer, and E. J. Gerberg. 1974b. Field evaluation of alternative light sources as mosquito attractants in the Panama Canal Zone. Mosq. News 34:382-384. Walker, E. D., and J. D. Edman. 1985. The influence of host defensive behavior on mosquito (Diptera: Culicidae) biting persistence. J. Med. Entomol. 22:37-72. Wei, Y. A., D. L. Hendrix, and R. Nieman. 1996. Isolation of a novel tetrasaccharide, bemisiotetrose, and glycine betaine from silverleaf whitefly honeydew. J. Agric. Food Chem. 44:3214-3218. Wei, Y. A., D. L. Hendrix, and R. Nieman. 1997. Diglucomelezitose, a novel pentasaccharide in silverleaf whitefly. 45:3481-3486.

PAGE 154

138 Weiss, H. B. 1943. Color perception in insects. J. Econ. Entomol. 36:1-17. West, A. S., and D. W. Jenkins. 1951. Plant feeding habits of northern mosquitoes studied with radioisotopes. Mosq. News 1 1:217-219. White, R. H. 1985. Insect visual pigments and color vision, pp. 431-494. In G. A. Kerkut and L. I. Gilbert [eds.]. Comprehensive insect physiology, biochemistry and pharmacology, vol. 6, Pergamon, Oxford, England. Williams, C. B., R. A. French, and M. M. Hosni. 1955. A second experiment on testing the relative efficiency of insect traps. Bull. Entomol. Res. 46:193-204. Wilson; R. A., B. D. Mookherjee, and J. F. Butler. 1991. Electronic insect repellency and attractancy tester. International Flavors and Fragrances Inc. NY, and University of Florida, Gainesville. USPN: 5,134,892. Appl. No. 691,339. Wilton, D. P., and R. W. Fay. 1972. Responses of adult Anopheles stephensi to light of various wavelengths. J. Med. Entomol. 9:301-304. Wolf, J. P., and W. H. Ewart.1955. Carbohydrate composition of the honeydew of Coccus hespehdium L.: Evidence for the existence of two new oligosaccharides. Arch, biochem. Biophys. 59:365-372. Worth, C. B. 1975. Sugaring for mosquitoes. Mosq. News 35:83-84. Wykes, G. R. 195 1 An investigation of the sugars present in the nectars of flowers of various species. New Phytol. 51:210-215. Xue, R. D., and D. R. Barnard. 1997. Influence of sugar availability and partial engorgement in Aedes albopictus on human host avidity and the repellency of DEET. J. Vec. Ecol. In press. Yee, W. L., and W. A. Foster. 1992. Diel sugar feeding and host-seeking rhythms in mosquitoes (Diptera: Culicidae) under laboratory conditions. J. Med. Entomol. 29:784-791. Yee, W. L., W. A. Foster, M. J. Howe, and R. G. Hancock. 1992. Simultaneous field comparison of evening temporal distributions of nectar and blood feeding by Aedes vexans and Aedes trivittatus (Diptera: Culicidae) in Ohio. J. Med. Entomol. 29:356-360.

PAGE 155

139 Yee, W. L., D. L. Hendrix, N. C. Toscano, C. C. Chu, and T. J. Henneberry. 1996. Diurnal field patterns of honeydew sugar secretion by Bemisia argentifolii (Homptera: Aleyrodidae) nymphs on cotton. Environ. Entomol. 25:776-782. Yuval, B. 1992. The other habit: sugar feeding by mosquitoes. Bull. Soc. Vector Ecol. 17:150-156. Yuval, B., M. L. Holliday-Hanson, and R. K. Washino. 1994. Energy budget of swarming male mosquitoes. Ecol. Entomol. 19:74-78.

PAGE 156

BIOGRAPHICAL SKETCH Douglas A. Burkett was born on January 25, 1964 in Libertyville, Illinois. He was the third child of Robert and Betty Burkett. Summer and part-time jobs cultivated his interest in entomology where he held positions working for a mosquito abatement district, the Chicago Botanical Gardens and served as an assistant cooperative extension agent for Cook County Illinois. Two years were spent at College of Lake County after which he transferred and completed a duel Bachelor of Science degree in Entomology and Pest Management from Iowa State University in 1987. In June 1988, he married his long-time girl friend, Laura L. Koeppen, also of Libertyville, IL. He then finished his Masters of Science in Medical and Veterinary Entomology at Kansas State University and was commissioned a First Lieutenant in the U.S. Air Force's Biomedical Sciences Corps in 1990. In his first assignment, he served as active duty advisor/medical entomologist for the Air Force's C-130 aerial spray squadron. He also supervised the DoD Aerial Application Certification Course and managed the research and development of new application techniques including biological insecticides, oil dispersants, and decontamination agents. He has served as a pest management consultant for the Air Force Reserve and taught contingency entomology to the Air National Guard. In July 1996 he was selected by the Air Force to return to school and complete his Ph.D. Upon 140

PAGE 157

completion at University of Florida, he will transfer to Okinawa, Japan and serve as a pest management professional and medical entomologist for the Department of Defense in South East Asia. He and his wife Laura have two daughters, Alexis and Sara. 141

PAGE 158

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. Daniel L. Kline, Chair Assistant 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 Philqj Jerry F /Butler, Cochair 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. Sandra L. Allan Assistant Scientist of Veterinary Medicine 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. EugenVJ. Professor of Entomology and Nematology This dissertation was submitted 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. June, 1998 Dean, College of Agriculture Dean, Graduate School