University of Massachusetts Amherst ScholarWorks@UMass Amherst
Doctoral Dissertations 1896 - February 2014
1-1-1985
Reproductive ecology and host-seeking behavior of the black fly, Simulium venustum Say (Diptera: Simuliidae).
Kenneth Raymond Simmons University of Massachusetts Amherst
Follow this and additional works at: https://scholarworks.umass.edu/dissertations_1
Recommended Citation Simmons, Kenneth Raymond, "Reproductive ecology and host-seeking behavior of the black fly, Simulium venustum Say (Diptera: Simuliidae)." (1985). Doctoral Dissertations 1896 - February 2014. 5630. https://scholarworks.umass.edu/dissertations_1/5630
This Open Access Dissertation is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Doctoral Dissertations 1896 - February 2014 by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact [email protected].
REPRODUCTIVE ECOLOGY AND HOST-SEEKING BEHAVIOR OF THE
BLACK FLY, SIMULIUM VENUSTUM SAY (DIPTERAs SIMULIIDAE)
A Dissertation Presented
By
KENNETH RAYMOND SIMMONS
Submitted to the Graduate School of the University of Massachusetts in partial fulfillment of the requirements of the degree of
DOCTOR OF PHILOSOPY
May 1985
Department of Entomology Kenneth Raymond Simmons © All Rights Reserved
National Institute of Health
AI—07109—06, AI—13981, RR-074B
World Health Organization
United States Department of Agriculture
NE-118
i i REPRODUCTIVE ECOLOGY AND HOST-SEEKING BEHAVIOR OF THE
BLACK FLY, SIMULIUM VENU5TUM SAY (DIPTERA: SIMULIIDAE)
A Dissertation Presented
By
KENNETH RAYMOND SIMMONS
Approved as to style and content by:
X -c MZ,, JS . John D. Edman, ! Chairman of Committee
Dr. Joseph S. Elkinton, Member J*. Dr. John T. Finn, Member Xyy,'/X Dr. John G. Stoffolano, Jr., Member
Dr. Ring T. Card£, Department Head Department of Entomology
iii To mom and dad - Carolyn and Leon (Doon) Simmons: your support, understanding, and love are my first and yet still my most cherished memories. There is no reason to archive them. They will always inspire me.
iv ACKNOWLEDGEMENTS
Special thanks to my advisor Dr. John Edman. Over the years he has provided immeasurable support, both emotional and -financial. I truly appreciate it. His willingness to let me pursue many of my own interests made this whole project a reality. I also am grateful to my committee members, Drs. Joe Elkinton, John Finn and John Stoffolano.
They made insightfull suggestions that helped me cure a couple bouts of "tunnel-vision". I owe the existence of this dissertation to Drs. K. Rothfels and R. Ferraday.
They willingly scored larvae for cytospecies identifica¬ tion. I thank them for this, even if they did make me eat duck feet. Dr. Steve Bennett allowed me to identify some of his engorged black flies.
I will always be thankful1 to Drs. Bob Coler and
Warren Litsky for inflicting me with an inspiration about nature during my Stockbridge years that has not wained with time. Dr. T.M. Peters carries all the weight of my future as a entomologist on his shoulders . He talked into making it my undergraduate major. I harbor no regrets and thank him for the persuasion. I also will be forever grateful to Mary Ssala. She really is a graduate student's best friend, and not just on payday!
v Special thanks to Chris Gates, Wren Withers, Jane Bain,
Nancy Nut 1 ie-McMenemy and Ernie. They helped me in ways -for which gratitude is not sufficient thanks. Sandra Allan
came to my rescue during the host-seeking studies. Jon Day
made helpful suggestions for identifying blood meals. I
am lucky to have Ned Walker and Dennis LaPointe as friends
and colleagues. They are 2 special folks. Two other
notable folks that were more than just 'drinking club'
companions are Don Eaton and Gaylen Jones.
Sassafrass and Boomerang deserve special credit. They
sat through hours of experiments without a word of com¬
plaint. They also are darn good buddies, even if they had
a habit of taking off.
Stephen Fel1ers'generosity, friendship and good nature
helped make this project a success, as well as enjoyable.
The rest of the 52 gang, particularly Warren and Martha,
also fit into this category. The debt of gratitude I accum¬
ulated during my graduate years for the patience and love
extended to me by my parents, grandmothers, and my siblings
and their families cannot be repaid.
Anne Averill's love and companionship made this task a
joy.
Financial support was provided by NIH grants AI-07109—
06, AI—13981 and RR-0748, USDA Regional Project NE-11B and
the World Health Organization. I am grateful for it.
vi ABSTRACT
REPRODUCTIVE ECOLOGY AND HOST-SEEKING BEHAVIOR OF THE
BLACK FLY, S.IMULIUM VENU5TUM SAY (DIPTERA: SIMULIIDAE)
(May 1985)
Kenneth Raymond Simmons
A.S., Stockbridge School of Agriculture B.Sc., University of Massahcusetts M.Sc., University of Massachusetts Ph.D., University of Massachusetts
Directed bys Professor John D. Edman
The blood-feeding and host-seeking behavior of black flies was studied in Franklin Co-, Massachusetts- Empha¬ sis was on the interaction between humans and Simulium venustum s.1 -, the most common and widely distributed spring species in the county.
Blood engorged black flies, including S. venustum, were captured by spraying trees with quick knockdown insec¬ ticide and by truck trapping- Ninety V. of the blood sources identified in 12 species of black flies were from mammals, in particular equines, bovines and racoons.
Vision was found to be important in long- and close- range orientation of Prosimulium mixtum/fuscum and S. venustum. Both species oriented to the undersides of host models but this behavior was most pronounced in S-
venustum. Close range orientation was affected by model
vii posture, shape, color, color intensity and legs suggesting the major cues -for orientation are associated with areas of maximum contrast both within the spehere of the host and with the host's background. UV reflecting pigments were least attractive and dark, low intensity pigments were most attractive to host seeking flies.
Simuliurn venustum began to blood feed when 24 hrs old but feeding was low until they were 120 hrs old. Maximum feeding was by 168 hr old females, the oldest group as¬ sayed. Comparison of follicle lengths and stage of wild caught host—seeking females with females of known ages suggested most females do not host seek until 120 hr old.
Simulium venustum developing at a lake outlet were significantly larger than individuals developing 1 km down¬ stream from the lake. These smaller flies had lower blood feeding rates in the laboratory and field, were less fe¬ cund, and did not survive as well as larger flies. Less than 207. of the S. venustum that landed on humans in nature blood fed.
The mermithid parasite Neomesomermis f1umenalis inhi¬ bited host-seeking, blood-feeding and ovarian development
of S. venustum but not "oviposition".
viii TABLE OF CONTENTS
DEDICATION . iv
ACKNOWLEDGEMENTS . v
ABSTRACT . vi i
LIST OF TABLES . xi
LIST OF FIGURES xv
CHAPTER
I. INTRODUCTION . 1
II. COLLECTION OF BLOOD ENGORGED BLACK FLIES AND IDENTIFICATION OF THEIR SOURCE OF BLOOD . 9
Introduction . Materials and Methods .. 10 Results .-. 13 Discussion . 1*? Conclusions . 24
III. BLOOD FEEDING BEHAVIOR OF SIMULIUM VENUSTL1M ON HUMANS IN NATURE . 26
Introduction . 26 Materials and Methods .- 27 Resul ts .. 29 Discussion ....-. 36 Conclusions . 45
IV. HOST-SEEKING BEHAVIOR OF SIMULIUM VENUSTUM AND PROSIMULIUM MIXTUM/FUSCUM . 46
Introduction .-. 46 Materials and Methods .-. 50 Results . ^*2 Discussion . ^1 Conclusions .. 1°8
V. CHRONOLOGICAL AGE AT BLOOD FEEDING AND FOLLICLE GROWTH IN FEMALE SIMULIUM VENUSTUM . HI
Introduction ... 1H LIST OF TABLES
Specific antisera used to identify black fly blood meals ....
Host blood sources of engorged black flies collected in spring and summer, 1980 — 1982 ...
Identification of blood sources in black fly mixed meals .
Percent engorgement of female black flies captured by truck trap and insecticide spray ..
Percentage of S. venustum that landed and fed on a human host
Test to determine if biting of humans by S. venustum during a 5—min sampling period occurred randomly .....
Time spent by S_. venustum females crawling on the skin of a human host before either feeding or flying away
Biting of humans positioned in an upright versus quadruped stance by E>. venustum ......
Percentage of parous female EL venustum captured biting humans versus those flying near the host ...*.
Number of black fly bites on 2 dogs of the same breed and percentage of total bites on each body region .
Number of _P. mixturn and S. venustum engorging on dogs versus those attracted to humans at the same time .
Number and percentage of P. mixtum and S. venustum females captured per body region on human hosts positioned in an upright versus quadruped stance . 13. Percentage of the total JP. mixtun> and S. yenusturn captured on the circumferential and longitudinal axis of black horizontal and upright model hosts ..
14. Number of P. mixtum and S. venustum females captured on vertical and horizontal panels baited with carbon dioxide ..
15. Percentage of the total P. mixtum and S. venustum females captured on vertical black panels baited with carbon dioxide 74
16. Number of P. mixtum and S. venustum females captured on the top and bottom of horizontal black panels baited with carbon dioxide ... 75
17. Percentage of the total female P. mixtum and S. venustum captured on each longitudinal section of black model hosts placed in the woods or field . 76
18. Number of P. mixtum and S. venustum captured on black horizontal models baited with or without carbon dioxide . 78
19. Mean and percentage of the total S3. venustum captured per body section on blue and black model hosts baited with and without carbon dioxide .....
20. Mean and percentage of the total P. mixtum and S. venustum captured on each of the 3 longitudinal and 6 circumferential sections of black horizontal models with and without
21. Percentage of the total P. mixtum females captured on each circumferential and longitudinal section on horizontal model hosts painted neutral colors with varying intensities 83
22. Percentage of the total S3, venustum females captured on each circumferential and longitudinal section on horizontal model hosts painted neutral colors with varying intensities 84
23. Mean number and percentage of the total .P- mixtum captured on variously colored horizontal models ... 85
xii 24. Mean number and percentage of the total S. venustum captured on variously colored horizontal models .... 86
25. Mean number of P. mixtum and S. venustum females captured on upright models painted various colors ... 90
26. Percentage of the total P. mixtum and S. venustum females captured on 4 sections of upright models painted various colors .... 92
27. Follicle growth and chronological age at blood feeding of S. venustum . 115
28. Follicle size and stage of S3, venustum females feeding on humans in the laboratory ... 117
29. Condition of follicular intima of parous j3. venustum captured while biting or flying near a human between 0600 - 2000 hr . 118
30. Ovarian development in wild S. venustum held at various intervals at 22 C . 120
31. Ovarian development of non-blood-fed S. venustum held in the laboratory 5+ days .... 121
32. Body length of larvae and subcostal length of adult S. venustum from SI and S2 . 137
33. Subcostal length and number of oocytes on non-blood fed S. venustum from SI and S2 .... 138
34. Least squares regression of subcostal length against number of oocytes matured in S. venustum after blood feeding . 139
35. Percentage blood feeding of 96+ hr old S. venustum females on humans in the laboratory 141
36. Subcostal length, fecundity and percentage blood feeding of S. venustum females ...... 142
37. Subcostal lengths of human-biting and non-biting S. venustum from SI and S2 . 143
xi i i 38- Subcostal lengths and number of oocytes of j3. venustum reared from SI and S2 field collected pupae ... **4
39. Mean longevity at 22 C of different size classes of S. venustum reared from pupae collected from SI and S2 ..-. 149
40. Laboratory blood feeding on humans by mermithid parasitized and non-parasitized S. venustum reared from wild pupae .. 165
41. Effect of mermithid parasitism on ovarian development and blood feeding behavior of S. venustum ...
size of S. venustum females non-parasitized and mermithid parasitized . *67
43. Percent daily survival of mermithid parasitized and non—parasitized S. venustum females . 172
44. Collection sites in Franklin Co. where S. venustum CC were positively identified in 1981 samples . 198
45. Number of times S. venustum and P. mixtum were captured biting humans in a 5—min test period .-... 201
46. Black fly species cpatured in 5-min landing collections and overhead net sweeps near Lake Wyola, May 1982 . 202
47. Species of Frankling Co. black flies collected biting humans at least once and species in overhead net sweeps but never collected biting humans in 161 samples, 1980 - 1983 .... 203
xi v LIST OF FIGURES
1. Percentage of the total nulliparous and parous S. venustum females captured biting vs. flying overhead of a human in hourly samples between 0600 and 2000 hours .. 37
2. Percent parous S. venustum females captured per hourly human biting vs. overhead net-sweep sample between 0600 and 2000 hours . 39
3. Horizontal and upright cylinder host models . 52
4. Relative spectral reflectance of pigments and polyethylene used to cover the host models ..... 59
5. Diagram of modified Dayno Aqua-Lab . 134
6. Percentage daily survival of S. venustum females from various populations and rearing regimes . 147
7. Percentage of the total non-parasitized gravid and mermithid parasitized female S. venustum captured in hourly net-sweep samples over an oviposition site at the outlet of Lake Wyola . 169
xv CHAPTER I
INTRODUCTION
Economic importance and control of black flies
Black flies are of great medical and veterinary impor tance. Toxic and allergic reactions from their bites and general annoyance from hordes of attacking females can lead to severe discomfort; sensitized individuals may re¬ quire hospitalizaton (Fredeen 1969, Newson 1977). Outdoor related industries (e.g. mining, lumbering and recreation) lose revenue during outbreaks (Fallis 1964, Jamback 1973,
1976). Black flies are responsible for decreased milk production in cattle and decreased weight gain in cattle and sheep in Canada and the United States (Fredeen 1956,
1959, Steelman 1976, Shemanchuck 1977, Annonymous 1979).
Black flies transmit filarial parasites to humans, domesitc and wild animals (Crosskey 1973, Addison 1980).
The most important species of these parasites is Onchocerca volvulus Leuckart, which causes a disease in humans that is called river blindness or onchocerciasis. Humans are the only reservoir and black flies are the only vectors of 0. volvulus. An estimated 20 - 40 million people in Afro- tropical and Neotropical regions have onchocerciasis
1 2
Simulium ienninasi Mailoch is a vector of Onchocerca ^ineal.is Rail let and Henry to cattle in New York (Lok 1981). Simulium ornatum Meigan transmits Onchocerca outterosa (Newman) to cattle in Europe (Eickler 1977). Addison (1980) reported that Simulium venustum s.l. is a vector of Dirofilaria ursi Yamaguti to black bears (Ursus americanus) in Canada. Black flies are also vectors of hematozoan parasites that cause death in avians (Grenier et al. 1975). The most important of the hematozoan parasites transmitted by black flies belong to the genus Leucocvtozoan■ Laird and Bennett (1970) reported that Leucocvtozoan simondi Mathis and Leger was responsible for high mortality of domestic geese in the subarctic. Tarshis (1973) and Herman et al. (1975) reported that Luecocvtozoan sp. caused up to 84% mortality in Canada Geese gosling in Michigan. Luecocvtozoan smithi (Lavern and Lucet) parasitizes domestic turkeys throughout southeastern United States and New York in the northeast (Noblet et al. 1975, Snoddy and Noblet 1976, Cupp pers. comm.). 3 Black flies are not known to vector any disease to humans in the United States. They primarily have been pests in rural regions of the country but they are becoming increasingly abundant in urban areas. Anti-pollution measures to improve the quality of urban waterways may be partly responsible for this problem (Merritt and Newson 1978). Improved water quality of the Miller's River in western Massachusetts has resulted in annoyance of humans in the region by S. jenninqsi complex, including human biting Simulium oenobscotensis Snoddy and Bauer (Simmons unpubl. data). Fifteen years ago no black flies lived in the Miller's (F.R. Holbrook pers. comm.). Similar observa¬ tions have been reported in Maine (May et al. 1977), Minnesota (Simmons and Sjogren 1984) and Pennsylvannia (G. Jones, pers. comm.). Restoration of the beaver (Castor canadensis Kuhl) may have contributed to black fly problems in Massachusetts. Outlets to beaver ponds are preferred breeding sites of Simuliurn venustum Say s.l., a major pest of humans in North America (Davies and Peterson 1956, Fallis 1964, Cupp and Gordon 1982). Larviciding is the most effective way to control black flies (Jamback 1981). Larval control is now feasible in the U.S. due to registration of the endotoxin produced by the bacterium Baci1lus thurinaiensis var. israeliensis. Large scale control programs now exist in Minnesota 4 (Simmons and Sjogren 1984) and Pennsylvannia (G. Jones pers. comm.). The total budget for these 2 programs in 1985 will be ca. 1 million dollars. Renewed interest in con¬ trol has increased the need for information on ecology and behavior of black flies in many parts of the U.S. where little previous work has been done. Black fly sibling species problem Simuliidae is a homogeneous family in which generic and specific identification are often difficult. This is further compounded by the fact that most described species are actually complexes of sibling species which can only be separated by characters on the giant polytene chromosomes of larval salivary glands (Rothfels 1979). This makes routine identification of species within complexes diffi¬ cult. It is possible to separate adults of some complexes by isozyme electrophoresis (May et al. 1977, Meredith and Townson 1981, Synder and Linton 1983, J.L. Peterson pers. comm.) but not in others (Adler pers. comm., Ferraday pers. comm.). Prosimuliurn mixtum/fuscum Syme and Davies complex consists of several species (Rothfels and Freeman 1977, 1983) but it is possible to separate some of these morphologically (Synder and Linton 19B3). The situation within the S. venustum complex is more complicated. It is closely related and morphologically very similar to 5 Sitnal iurn verecundum Stone and Jamback. These two species complexes were considered to be one species until 1955 (Stone and Jamback 1955). Now there are 10 known siblings of S. venustum and 2 siblings of S. verecundum (Rothfels 1981). Simuliurn venustum s.l. is considered to be a man- biting species and S. verecundum is not. The most widely distributed and presumably the major human biter in the S. venustum group is the CC sibling (Rothfels et al• 1978, Rothfels 1981). CC is widely distributed in the north¬ eastern U.S. (Cupp and Gordon 1982). Discovery of the large number of sibling species of S. venustum in North America has created a dilemma in that most behavioral and ecological studies were carried out be¬ fore it was known to be a species complex. No ecological or behavioral studies on a pure population of a sibling have been reported. Different cytospecies may breed in different habitats but characterization of these habitats is difficult (Gordon and Cupp 1980, J.F. Berger pers. comm.). Black flies in Massachusetts Holbrook (1967) found 21 species of black flies in western MA of which £. mixtum/fuscum s.l. and S. venustum s.l. were the most abundant and widely distributed. They also were the only 2 species that he observed to bite humans. I surveyed part of Worcester Co. and most of 6 Franklin Co. for larvae and found 10 species in addition to those that Holbrook (1967) found. Prosimulium mixtum/ fuseurn and §. venustum were the most widely distributed and abundant spring species. Simulium verecundum. S. jenninqsi complex and Simulium fibrinf1atum Twinn were the most widely distributed and abundant summer species (Simmons, unpubl. data; Simmons and Edman, In Cupp and Gordon 1982). Background of the present study I conducted extensive field studies on the larval ecology of black flies between 1977 and 1980 in eastern Franklin Co. I repeatedly observed that S. venustum, at times hordes of them, were attracted to humans but rarely bit them. This was not consistant with reports in the literature that this species is a vicious human biter (Davies and Peterson 1956, Fallis 1964, Crosskey 1981, Cupp and Gordon 1982). I hypothesized that the reason for the discrepancy between my observations and the literature was that more than one sibling species was present in Franklin Co. and the predominant sibling was not a human biter. I planned to test my hypothesis by identifying the cyto- species in my study areas and characterizing the adult isozyme patterns using the methods of May et al. (1977). My hypothesis was rejected since only 1 cytospecies, CC, 7 was -found in any o-f the populations I sampled in my study area (Appendix 1). It also is now known that because of the homogeneity of the §. venustum group electrophoresis may not be useful to separate most siblings, particularly adults (Synder 1982, Ferraday pers. comm.). Specific objectives Because only one sibling of the S. venustum complex breeds in my study area it was possible to conduct field studies on adult populations that were most likely all the same sibling. I chose to focus my investigations on why S. v®Qy=»tum are attracted to humans but rarely bite. The specific objectives of my study were to determine: 1) blood sources of wild engorged females, 2) feeding behavior of females attacking humans, 3) the orientation behavior of host seeking females, 4) the chronological age of the onset of host—seeking and biting behavior, 5) importance of larval habitat on blood feeding behavior and reproductive potential, and 6) the effect of the mermithid parasite Neomesomermis f1umenalis Welch on female biting and re¬ production. Literature relevant to these topics is re¬ viewed in each chapter. General biology of S. venustum s.l. in western Massahcusetts Simulium venustum is widely distributed in western MA. The preferred habitat of larvae is second and third order 8 streams. The largest populations occur at outlets of im¬ pounded waters, including beaver ponds. Eggs hatch during March. Larval development generally is completed by the second week of May, but at some sites it is earlier and at others it is not until mid-June (Simmons unpubl. data). Biting of humans by S. venusturo first occurs during the last week of April and peaks during the second and third weeks of May. There is one generation per year. CHAPTER II COLLECTION OF BLOOD ENGORGED BLACK FLIES AND IDENTIFICATION OF THEIR SOURCE OF BLOOD Introduction Information on black fly host preference is usually based on trapping studies using various animals as bait or collection of engorging flies from wild or domestic animals (Davies and Peterson 1956, Anderson and DeFoliart 1961, Fallis 1964). These studies provide useful data on which hosts are potential blood sources but they do not provide a quantitative assessment of hosts fed on in nature. Identi¬ fication of blood sources in wild caught engorged females is the best way to assess feeding patterns (Washino and Tempelis 1983). Three studies have been published on sero¬ logical identification of blood from wild caught black flies (Davies et al. 1962, Disney and Boreham 1969, Walsh 1980). Downe and Morrison (1957) tested 1,672 engorged females collected in window traps in a one-room barn housing horses, cattle, pigs and chickens. Difficulty in collecting engorged black fly females has caused this lack of serological studies. Little is known of black fly resting sites, particularly those of 9 10 engorged females (Service 1977). Wolfe and Peterson (1960) suggested that nearctic species rest in tree canopies at night but this has not been confirmed. Service (1977) proposed use of aerosol insecticides in tree canopies as a way to evaluate this possibility. Tree canopies sprayed with quick knock down insecticide (Resmethrin) with a ULV back-back sprayer resulted in the collection of small num¬ bers of black flies, including males and engorged females. Small numbers of engorged females also were captured by truck trapping. In this chapter, I report the identity of blood sources from the engorged black flies collected by the two methods. Data on the comparative sampling effi¬ ciency of the 2 techniques also are given. Materials and Methods Study area. Truck trapping was conducted on two, 6.5 km routes in Franklin Co., Massachusetts. One route was in the town of Shutesbury and the other in New Salem and Wendell. The Shutesbury route was along heavily forested rural roads with numerous residences and 2 horse and 1 small cattle farm. All but 1.5 km of the route was along the Saw Mill River, Lake Wyola and Dudleyville Brook. The New Salem/Wendell route was on a dirt road that abutted the West Branch of the Swift River. There were 5 active beaver ponds and 1 residential dwelling on this route. 11 Truck trap. The truck trap was constructed of 0.25 mm mesh nylon netting and aluminum angle. It was held on the roof o-f the truck cab with suction cups and gutter straps. The opening o-f the trap was 1.3 m long X 1 m high. The cone- shaped net was 2 m long with a 4 cm opening at the tip of the cone. Insects were collected in 4 cm dia. X 30 cm long plastic vials with bottoms replaced with nylon net. Vials were secured in the tip opening by velcro. They could be removed from the net, capped and replaced with an empty one in a few seconds. Fly collection procedures. During 1980, trapping was con¬ ducted on the Saw Mill River route once a week from 0400 - 2200 hr during July and August. During 1981, both routes were sampled once a week from 0600-2200 hr from late April through June. In 1982, the Saw Mill River route was sam¬ pled weekly between 1400 and 2100 hr from late April - mid- September. Vials were changed at 4 points along the Shutesbury route and 5 points on the Wendell route. Actual subdivision of the routes was based on habitat, but each was ca. 1 km long. Captured black flies were stored in ice until returning to the laboratory. In the laboratory, they were counted, identified, and the engorged females frozen at -40 °C until identification of blood sources. The lower canopy of a maple and pine tree at the edge 12 of a horse pasture were sprayed with Resmethrin by back pack sprayer (Kioritz model DM 9) equipped with a ULV nozzle calibrated to deliver 200 ml/min. The resmethrin was diluted to a concentration of 3.37. with 54 sec mineral oil. Trees were sprayed for 5 min between 0400 and 0600 hr, generallly when it was too dark or cold for substantial black fly activity. After spraying, the trees were shaken as much as possible to knock dead flies onto yellow ground cloths spread beneath them. Engorged black flies were identified to species and frozen as above until identifica¬ tion of the blood source. Each tree was sprayed on 16 different days between late May and mid-July 1980. Blood source identification. Blood meals were identified using the precipitin technique as described by Edman et al. (1972). Engorged flies were individually crushed in 3.5 ml polyethylene centrifuge tubes containing 0.4 ml of 0.85% phosphate buffered saline at pH 7.2 and allowed to extract overnight at 1.5 °C. Extracts were centrifuged and the supernatants transferred to clear polycarbonate 3.5 ml tubes using a pasteur pipette. Blood meals were identified by the modified capillary precipitin test (Tempelis and Lofy 1963). Each extract was first tested with broadly reacting avian or mammalian anti¬ sera. Extracts which did not react with either antisera were then tested with reptile and amphibian antisera. Once 13 the vertebrate class of the the meal was established, each extract was tested with the specific antiserum listed in Table 1. Results Blood-meal sources. A total of 123 engorged black flies consisting of 12 species from 3 genera were tested. Of these, 105 (857.) reacted at least in screening tests for mammal or avian (Table 2). None reacted with the reptile or amphibian antisera. The majority of meals identified were mammal except for meals tested from Simulium qouIdingi Stone and Simuliurn aureum Fries (Table 2). Specific blood sources of identified meals are summar¬ ized in Table 2. For all black fly species combined 267. of all mammal meals were from bovine, 337 from equines, 207 from racoons, 147 unidentified mammal, 17 unidentified avian and 77 were mixed. One JP. mixtum/fuscum meal was from a dog. The only human meal identified was from S. vernum. Steaopterna mutata and £. gouldinai each had one blood meal identified as passerine. Only 1 engorged S. aureum was captured. It had a ciconiform blood meal. Table 3 is a summary of mixed meal sources. Prosi mul i urn magnum« St. mutata, Si mul i urn crox toni. Nicholson and Mikel, S_. vernum and S.venustum contained blood from 14 Table 1. Specific antisera used to identify black fly blood meals. Mammalian Avian rabbit passerine bovine ciconiform equine gruiform human chicken porcine racoon opossum squirrel skunk canine Host blood sources o-f engorged black -flies collected in spring and summer, 1980 15 16 Table 3. Identification of blood sources in black fly mixed meals. black fly species (no. mixed meals) mixed P. St. S. S. 5- meal maqnum mutata croxtoni vernum venustum bovine/equine 1 - - 1 bovine/racoon i equine/racoon 1 racoon/passerine 1 1 - ~ equine/ciconiform - - - 1 17 more than one host. Racoons were involved in 3 of the 5 different mixed meal combinations. Prosimulium magnum. St. mutata and S. venustum had mixed meals of avian and mammal origin (Table 3). Of the 6 engorged §. venustum captured by knockdown, 5 (83%) were identified as having fed on horse and 1 on both horse and bovine blood. The engorged S. venustum captured by truck trap had different blood sources. Seventeen per— cent were unidentified mammal, 38% horse, 21% bovine, 17% racoon, 3% racoon/bovine and 3% horse/bovine. Of the 6 S. venustum captured that had fed on racoon all were collected from sub-divisions of the truck trap routes that were near water. The same was true for the other species that had fed on racoon (Table 2). Too few engorged females of any one species were captured to draw conclusions on differ ences in blood sources from the Shutesbury vs. Mendel1 routes. Comparison of trapping techniques. Eighteen percent of the females captured by insecticide knockdown were engorged compared to 0.9% of the ones captured by truck trap (Table 4). Engorged females of twelve different species were captured by the the truck trap technique compared to 3 by the knockdown technique. The only species captured by the knockdown technique not captured with the truck trap was S. vernum. 18 Table 4- Percent engorgement of female black flies cap¬ tured by truck trap and insecticide spray. truck trap^ insecticide^ no. 7. no. 7. species captured engorged captured engorged P. mixtum/ fuscum 3501 0.43 14 0 P. magnum 514 3.11 2 50 St. mutata 2024 0.30 8 0 S. venustum 930 2.37 38 21 S. verecundum 210 1.43 0 — S. tuberosum 133 3.01 0 — S. vittatum 91 2.20 0 — S. jenningsi^^ 1263 0.79 7 — S. gouldingi 27 11.00 0 — S. aureum 20 5.00 0 — S. croxtoni 17 6.00 0 — S. vernum 0 — 15 33 totals 8730 0.92 76 18 ♦Combined truck trap catches for 1980, 1981 and 1982. ♦♦Total of 32 trees sampled on 16 days in 1980. ♦♦♦Includes members of S. jenninqsi complex and S. fibrinflatum. 19 Eighty-two percent of the total number of black flies (n= 76) collected by the knockdown technique were from the pine and lQVm were from the maple tree. Of the 13 engorged females captured by the knockdown method, 12 (927.) were from the pine tree and 1 from the maple. Blood meal sizes were assessed visually as incomplete or complete. All females captured by knockdown generally had fresh, complete meals whereas most captured by truck trap had incomplete meals or meals that were partially digested. Discussion general discussion. I did not collect and identify enough blood meals of any one black fly species to draw definitive conclusions about feeding patterns or preferences. My re¬ sults agree with previous host selection studies on Nearc- tic black flies which report that large mammals, particlar— ly bovines and horses,are common hosts (Davies and Peterson 1956, Anderson and DeFoliart 1961, Fallis 1964). Horses were the most abundant host in my study site, and the highest percentage of meals were from horse (Table 2). Spraying was done adjacent to a horse pasture and all meals of §. collected there were from horse.' Downes and Morrison (1957) found that S. venustum fed more frequently on horses than cattle, pigs or chickens housed together in 20 a 1 room barn. The bovine feeding rate also was quite high, although the anti-bovine sera I used would not have separated cattle blood from white-tailed deer blood. It is interesting that there were not more feedings on dogs (1 meal in P. mixturn/fuseurn) since several species are known to be attracted to them. This may have been due to anti-black fly behavior of dogs (Chapter IV). Simulium venustum and £. mixtum/fuscum are the major black fly biters of humans in Massachusetts and North America (Davies and Peterson 1956, Cupp and Gordon 1982). No human meals were identified from either of these 2 species. These results suggest that despite their availability humans are not important blood sources of S. venustum or P. mixtum. Moreover, most S. venustum females that are attracted to humans are observed not to feed (Schreck et al. 1980, Chapter III, Appendix II). In comparison, £. mixtum/fuscum frequently bit humans in my study areas (Ap¬ pendix II). The only species with a blood meal identified as human was S. vernum (Table 2). Davies et al. (1962) found that (as £usimulium latipes) in Scotland this species fed on birds, humans and other mammals. There are other reports of this species feeding on humans as well (Davies and Peterson 1956, Fallis 1964). The 6 other meals of this species that I identified were from bovine (1), horse (2), 21 racoon (1), unidentified mammal (1) and racoon/passerine (1). As discussed in a previous report (Davies et al. 1962), this species does not follow the pattern of Eusimulium which generally is believed to be a strict bird feeder (Fallis and Bennett 1958). Though widely distri¬ buted, and fairly abundant near my study sites, I have never collected S. vernum biting me, but I have collected it in overhead net samples (Appendix II). Overall, 20% of the identified blood meals were from racoon (Table 2). Eight species were found to have fed on racoons. Four species had racoon blood in meals of mixed sources (Table 3). There are 2 previous reports of S. venustum feeding on racoon (Davies and Peterson 1956, Wright and DeFoliart 1979), but none of the 7 other species were previously known to feed on racoon. All racoon meals were identified from samples collected from sites adjacent to water. Racoons are nocturnal and black flies are diurnal blood-feeders (Wenk 1981). Perhaps the black flies feed on racoons as they rest in open tree cavities during the daytime. Seven percent of all meals were from mixed host blood sources (Table 2). This may be related to a host's anti¬ black fly behavior (Chapter IV). Host defensive behavior causes feeding disruption in mosquitoes and may contribute 22 to multiple host feeding (Edman et al. in press). Results from the knockdown experiment support Wolfe and Peterson's (1960) suggestion that black flies rest in trees at night. My results clearly show that engorged females also rest in trees during daylight. I have sampled understory vegetation, grasses along stream banks, and in fields and leaf litter with power aspirators (Nasci 1982) and sweep nets during day and night and have captured no engorged females and very few males or unengorged females. Along stream banks, the only time I have found large num¬ bers of flies (particularly P. mixtum/fuscum. Cnephia dacotenisis Dyar and Shannon, mutata and decorum) is after eclosion or when they congregate prior to mating or oviposition. These findings do not agree with several pre¬ vious studies which suggest that Nearctic black flies rest in the understory vegetation or along stream banks (Service 1977). I feel most Nearctic species, at least, rest in tree canopies except just prior to oviposition or after eclosion. Most of the engorged (94%) and non-engorged (82%) black flies captured by the knockdown technique were from the pine tree. This may have been due to better dispersal of the insecticide in the pine, a preference to rest in pine over maple, or better location of the pine relative to the maple tree. Much comparative work remains to be done 23 on this. Methods also should be developed to distribute the insecticide higher into the tree canopy. Preliminary evidence suggests that the maximum vertical killing dis¬ tance of the spray was 8 meters (S.R. Bennett, unpubl. data). Many more black flies were captured in the truck trap than by the knockdown method but the percentage of engorg¬ ed females was much higher with the knockdown method (Table 4). All engorged flies captured by the knockdown technique had fresh, complete blood meals whereas most of the engorg¬ ed flies captured in the truck trap had incomplete or partially digested meals. Females captured in the truck trap may have been seeking nectar, additional blood or were moving to different resting sites. Davies and Roberts (1973) reported much higher blood engorgment rates of 3 species of black flies captured by truck trap in England. The knockdown technique also has the advantage of being applicable in areas where vehicle travel is not possible. It would be interesting to identify more meals from S. venustum captured in different habitats. This species is a potential vector of Dirofilaria sp. in Massachusetts: greater than 1 X of the parous S. venustum females cap¬ tured in truck trap collections were found to have LI and L2 filaria in the Malphigian tubules. Addison (1980) re- 24 ported that S. venustum is a vector of g. ursi to black bears in Canada. Bears are rare in the vicinity of my study sites (Dodge pers. comm.). Dog heartworm is endemic in the vicinity o-f Lake Wyola (M.L. Katz pers. comm.). Prelim¬ inary studies on S. venustum -fed on a heartworm infected dog resulted in destruction of Nalphigian tubules but no larvae were recovered (Sevrino and Simmons unpubl. data). Racoons may be possible sources of the infection. They are hosts of 2 filarial species that develop in Malphigian tubules of mosquitoes (Hawkins and Worms 1961). Conclusions 1. Blood engorged black flies were captured by spraying quick knockdown insecticides into tree canopies and by truck trap. This is the first study of precipitin identi¬ fication of blood sources of wild caught Nearctic black flies. 2. Blood engorged black flies rest in trees during day¬ time. 3. More engorged black flies were captured by the truck trap but a higher percentage of the females captured by the knockdown method were engorged. 4. The greatest percentage of blood meals were from mammal sources, particularly equines, bovines and racoons. 5. This was the first study to demonstrate such a high 25 feeding rate on racoons. 6. Seven percent of the meals identified were from mixed hosts sources. This suggests hosts may exhibit anti black fly behavior and interrupt feeding. CHAPTER III BLOOD FEEDING BEHAVIOR OF SIMULIUM VENUSTUM ON HUMANS IN NATURE Introduction The two major black fly pests of humans in Massa¬ chusetts are Prosimuliurn mixtum/fuscum and Simuliurn venustum (Holbrook 1967, Appendix II). Prosimuliurn mixtum/- fuscum is a more aggressive human biter than §. venustum (Appendix II). Simi liar observations have been made in Maine (Schreck et al. 1980). The Maine observations are interesting because S. venustum is often considered a severe human biter in North America (Davies and Peterson 1956, Fallis 1964, Cupp and Gordon 1982). I initially hypothesized that variation in human biting by S. venustum was due to the presence of multiple cytospecies, some of which bite humans more readily than others. However, the only cytospecies found after a sur— vey of 21 breeding sites in Franklin Co., MA was CC (Appen¬ dix I). CC is the most widely distributed of the 10 known cytospecies in the S. venustum complex (Rothfels et al. 1978, Cupp and Gordon 1982) and is reported to be a human biter (Cupp and Gordon 1982). 26 27 These results prompted me to begin a detailed study to examine other factors which may influence human biting by S. venustum. The objectives of this study were to deter— mine: 1) the percentage of §. venustum females which actu¬ ally bite after landing on humans, 2) the time J5. venustum spend on exposed skin or clothing before biting or flying away, 3) the effect on biting of visual orientation cues on biting, 4) the diurnal biting cycle, and 5) parity of biting and non-biting females. Materials and Methods Study sites. Sampling was done at the edge of clearings in forests near S. venustum breeding sites in MA and New Hampshire. The MA site was 400 m from the Lake Wyola outlet to the Saw Mill River (Franklin Co.). The New Hampshire site was near S. venustum breeding sites on White Brook where it flows beneath the Kancamangous Hwy. (White Mtn. Natl. Forest, Grafton, Co.). The NH site was chosen for comparative purposes. Three cytotypes in addition to CC breed in this area (J. Berger, pers. comm.). Quantification of human biting by §. venustum. The proce¬ dure to quantify biting of humans was as follows: 1) a shirtless male wearing blue jeans positioned himself in a shaded spot at the edge of the clearing. 2) after standing 28 motionless for 1 min the number of S. yenusturn that landed on the exposed skin (stomach/chest and arms) and blue jeans was recorded and the time of their stay Mas measured with a stopwatch, and 3) females that began to engorge were collected with a mouth aspirator. All tests lasted for 5 min. No attempt was made to count females that made "touch and go" landings. In the NH tests, the time females spent on the collector was not measured. Tests were done between 1600 and 1800 hr on days with high fly activity (low wind, 17 - 22 °C) during mid- to late-May, the time of peak S. venustum biting in the area (Simmons, unpubl. data). Only one test per day was run at the same site. One individual conducted all tests. It was possible to identify S. venustum with the unaided eye since it was the only "stock¬ ing-footed" black fly that regularly landed on humans in the vicinity of the study site during the experiments (Appendix II and Simmons unpubl. data). Effect of visual cues during orientation of S. venustum. —-—-as_—-- The experimental procedure consisted of the collector positioning himself in a quadruped (hands and knees on the ground) or an upright stance and recording the number of flies that landed and/or bit. Touch and go landings were not included. Paired tests were accomplished by making consecutive 5-min samples for each stance on opposite sides of a field 1 hr apart. The collector left the area during 29 the interim. All tests were done in Franklin Co.. Parity and diurnal biting rhythm. Five minute biting col¬ lections of S. venustum (collector shirtless) were taken hourly from 0600 - 2000 hr one day each week during the last 2 weeks of May and first week of June 1981. At the end of each hourly biting sample, 3 back and forth over¬ head net sweeps were made to collect females hovering overhead. Flies were stored on ice until returned to the laboratory. They were frozen at -40 °C until dissected to determine parity (Detinova 1962). Results Human biting by S. venustum. At the MA site, 207. and 107. of S. venustum females that landed fed in 1982 and 1983, respectively. At the NH site, 117 fed (Table 5). The succession of females landing on the collector during the 5-min sampling period that bit versus did not bite was analyzed for randomness by the run test using a normal approximation (Sokal and Rohlf 1969). Biting did not occur randomly during the sample periods but tended to be clumped (Table 6). This suggests that biting was influ¬ enced by something, possibly whether the previous female that landed on the collector had bitten him. 30 Table 5- Percentage of S. venustum females that landed and fed on a human host. population no. sampled year 1anded no. bite (%) X2 MA - Lake Wyola 1982 110 22 (20) 4.41^ MA - Lake Wyola 1983 150 15 (10) 0. 68^ NH - White Br. 1982 220 24 (11) ♦Chi-square test, not significantly different. ♦♦Chi-square test of 1982 and 1983 Lake Wyola samples combined versus White Br. samples. Not significantly different. 31 C £ TJ L 3 C O e n m n m «B •** i o o o > in Oi ■ • • IB ■o •H o o o »H £ «B x O 3. IB •H u O C E to QJ L to 01 01 tfl r-i £ a oi > ai at a u U L c • L U 0 01 cni ai 3 £ L £ to > 0 M CM IB L >£ 18 C E 3 £ 0 « £ \ L U o at TJ O U tfl 0 C £ C c 0 C £ iB IB Ql E 01 > 3 > O' ■H £ h c 10 £ to 10 4- IB f-* 01 TJ Oi 0 H IB h Ql o in u 01 £ IB TJ in ^ u D> L O E C -H (N 3 C £ 01 IB 10 >< 4- r-< Ql £ rH Ql E L £ 0 0i c O TJ H 3 E 4- C a to L •*-* IB m e tj L 01 L as o 0i O £ 01 10 £ -h C TJ 0 L £ CM £ QJ c c at E L ih a > Ql £ L 3 E £ L O 01 U 01 £ U TJ Z s Oi 0 L CM K> CM 01 TJ TJ IB 00 00 00 N tO C aj O' O' Ch > C 01 a£ .X h* •H c IB IB O to O' 01 £ 0 H H O IB >0 £ • O' TJ O 0 L 2 th •r* NO C £ £ 01 > > CQ £ •H O IB i-« 3 3 X 01 ^ c r-H a 01 IB 4* TJ H a 3 E 01 01 £ C £ E TJ a <8 -X .X f-4 IB £ Females that bit spent a mean of 2.9 secs (range 2 - 5 sec) walking on exposed skin before initiating feeding. Females that did not bite remained for 2.5 sec (range 1 - 8 sec). The difference between these means was not signi¬ ficant (Table 7). Females that landed the collector moved at a rapid pace dragging their labial palps and/or tapping their fore-tarsi on the skin. The speed at which they moved appeared to increase with time, though this was not quantified. Host stance and biting by S, venustum. There was no significant difference in the biting rate or the percentage of females that landed which bit when the subject was in a quadruped stance compared to an upright stance (Table 8). Numerous females made touch and go landings every few centimeters on the collector's chest and stomach when he was in the quadruped position. I did not attempt to quan¬ tify the number of these landings but it was clear that there were many more than when in an upright stance. Parity and diurnal biting cycle. The percentage of parous females which bit the collector during the hourly 5-min sampling periods was significantly higher than the percen¬ tage captured in overhead net sweeps at the end of the test (Table 9). Overall, 62% of the females that bit the collec¬ tor were parous compared to 47 percent parous in the net sweeps. 33 Table 7. Time spent by S. venustum females crawling on the skin of a human host before either feeding or flying away. female X ± S.D. time (secs) on skin behavior n per female 1anded* blood feed 22 2.90 ± 1.18 not blood feed 88 2.54 + 1.70 ♦Difference between means was not significant as determined by Student's t-test. 34 Table 8. Biting humans positioned in an upright versus quadruped stance cr o S, venustum. % of females that body number of X ± S.D, females landed on subject stance reps. biting/rep.* which fed quadruped 6 2,67 ± 2.34 10 upright 6 2,50 ± 1-87 10 *Difference between means was not significant as determined by Student's t-test. 35 Table 9 . Percentage of parous female S. venustum captured biting humans versus those flying near the host. behavioral state of S. venustum when captured* biting flying sample total total 7. date cap. parous cap. parous 21—V—81 48 57 84 35 (N H GO > GO 1 1 49 63 114 54 4-V-81 13 77 65 51 totals 110 62 263 47* ♦Differences in parity between biting and flying females were tested for by the Chi-square test (X2 =6.11, P<0.05). 36 There was a morning and evening peak of both biting and flying nulliparous and parous S. venustum (Fig. 1). Parity tended to increase in the afternoon for both biters and flyers (Fig. 2). A Chi-square test for goodness of fit of observed vs. expected per hourly sample was significant 2 for parous (X = 25.63, P<0.05> but not nulliparous flies. Discussion Four factors affected the biting behavior of S. venustum: 1) site and year of sample, 2) age of the population, 3) outcome of a previous encounter with the same host by conspecific females, and 4) time of day. I also have found that small, possibly nutritionally stress¬ ed, S3. venustum do not feed on humans in the laboratory and field as readily as larger females (Chapter VI). No more than 20V. of the female S* venustum from MA or NH that landed on the exposed skin of a human fed (Table 5). This low feeding rate is not unique to my study sites. Davies (1953) reported biting rates in Canada that calcu¬ late to 157. (my calculations) feeding. Schreck et al. (1980) reported that few S3. venustum attracted to humans feed, though they did not quantify their observations. Laboratory feeding on humans by £,. venustum reared from pupae collected from breeding sites near my field sites 37 Figure 1-Percentageofthetotalnulliparousand parous 01 Itn . venustumfemalescapturedbitingvs.flyingoverhead of human inhourlysamplesbetween0600and2000hours. OF TOTAL CAPTURED HOUR OFDAY 39 Figure 2. Percentage of the total S. venustum captured biting vs. flying overhead of a human in hourly samples between 0600 and 2000 hours. PAROUS 99 PER SAMPLE HOUR OFDAY 1900 40 41 ranged -from <1% - 44% in MA (Chapter VI) and 18% in NH (Simmons unpubl.data). Bait trapping studies suggest S. venustum/s preferred hosts are horses and cattle (Davies and Peterson 1956, Downes and Morrison 1957, Anderson and DeFoliart 1961, Fallis 1964). No blood meals identified from a small number of wild caught engorged §. venustum captured near my study site in MA were from humans (Chapter II). Most were from equines (38%), bovines (21%), racoons (17%) or mixed combinations of these 3 vertebrates. My previous studies showed that S. venustum orient directly to the undersides of natural or quadruped models (Chapter IV). In the present study, I found that there was no difference in the percentage of the females which landed and bit a human in an upright vs. quadruped stance (Table 7). These data suggest that orientation to a host and biting are separate processes. This does not support Sutcliff and Mclver (1979) and Smith and Friend's (1982) hypothesis that onset of biting "mode" is integrated with host seeking. Interestingly, S. venustum's close relative S. verecundum ACD cytospecies, which is abundant in MA but is rarely attracted to humans, (Appendix II) feed readily on them in the laboratory (Chapter VII). My experiments on the time S. venustum spent walk- ing/probing the collector suggest the flies derive infor— 42 mation about a host after they are in contact with it. Fe¬ males that bit the collector spent more time crawling on the skin than females which did not feed, though this difference was not significant (Table 7). I also have observed that females landing on a collectors blue jeans remain for a longer period of time than females landing on the skin (Simmons unpubl. data). This may have been due to blockage of sensory information by the fabric or chemical cues which stimulate probing accumulate in the fabric. Mosquito blood feeding is influenced by what appear to be olfactory cues from engorging conspecifics, a phenomenon called "invitation effect" (Alekseev et al. 1977, Ahmadi and McClelland in press). Schlein et al. (1984) discovered that engorging phlebotomine sand flies produce an aggrega¬ tion pheromone. My data suggest a similiar effect is produced by engorging S. venustum. Females that landed and bit a collector were not randomly dispersed among females that landed but did not bite the collector during 5-min sampling periods (Table 7). Biting tended to occur in clumps. This suggests that the presence of a biting female may have stimulated subsequent females that landed to also bite. I also have observed that in laboratory assays, biting is not random and is influenced by other females feeding (Simmons unpubl. data). Stokes (1914) described a 43 report by Meigen (1898) that biting by one S. venustum stimulated other ' females to feed. An invitation effect helps to explain why S. venustum make touch and go landings along the stomach of quadruped hosts. Such a foraging strategy would minimize the danger associated with crawling on the body of a defensive host (Chapter IV), while opti¬ mizing chances of locating potential feeding sites. It would also explain why scars on horses from black fly bites are clumped (Simmons, unpubl. data). A significantly greater percentage of the §. venustum that bit humans were parous compared to the females cap¬ tured hovering overhead at the end of a 5 minute bite count (Table 9). Disney (1972) observed similiar results for S. damnosum s.l. attracted to humans and chickens. He offered three hypotheses to account for this phenomenon: 1) host specificity decreased with age, 2) the longer a fly seeks a host the more likely it is that host specificity will decrease and 3) availability of man as opposed to other preferred hosts may vary. The first of these seems to be the most likely explanation for my results with S. venustum. Further studies should be conducted to determine the exact reasons. I observed a bimodal feeding pattern of S. venustum (Fig. 1) , and an increase in the parity rate toward after— noon (Fig. 2). Davies (1963) observed similiar trends for 44 S- venustum in Canada. The increase in parity may be due to the -fact that S. venustum generally oviposit between 1200 and 2000 hr, with a peak around 1700 hr (Chapter VII). Females host-seek and bite the same day they lay an egg batch (Chapter V). Host seeking began in the morning as soon as it was light and temperatures rose above 14 °C and ended in evening when temperatures fell below that level or because it was dark. There have been a number of reports on diurnal biting acitivity of black flies (Lacey and Charlwood 1980). The actual percentage of S. venustum which locate a human in nature and engorge is probably much lower than I report here. Humans are defensive towards black fly at¬ tacks and probably prevent most females from obtaining complete meals. Still, S. venustum is an important pest of humans due to its abundance and high landing rate. Conclusions 1. Simuliurn venustum CC or S. venustum s.l. do not readily feed on humans. Maximum feeding was 20%. 2. Run test analysis of the sequence of biting by S. venustum females on humans suggested the outcome of one event (decision to bite) was influenced by the outcome of a prior event (female(s) already biting the collector). This 45 suggests the possibility of an invitation effect. 3. The time females spent crawling on the skin of the collector was not different for biters vs. non-biters. 4. The stance of a human host did not affect the number of females that bit. However, it did affect the number of females that landed. 5. Parous females bit humans more frequently than nulli- parous females. 6. There was a morning and evening peak in both attraction and biting of humans by nulliparous and parous S. venustum. Parity of biters and non-biters tended to increase during the afternoon. This may be related to the fact that ovipo- sition occurs at this time. CHAPTER IV HOST-SEEKING BEHAVIOR OF SIMULIUM VENUSTUM AND PRQSIMULIUM MIXTUM/FUSCUM Introduction Bradbury and Bennett (1974b) proposed the following hierarchical zones of orientation to describe the in flight host seeking behavior of black flies: 1) long-range orien¬ tation, 2) middle-range orientation, and 3) close-range orientation. Smith (1966) proposed a similiar model to describe black fly host seeking, but he included crawling once the fly was on the host as a component. It is diffi¬ cult to determine the exact sequence of stimuli used by black flies during each phase of host seeking because they do not respond in laboratory wind tunnels or olfactometers. Hence, all interpretations of host seeking behavior are based on field experiments (Wenk 1981). Long-range orientation is the initial phase of flight towards the host. Olfactory stimuli involving host speci¬ fic odors are believed to be a releaser of this behavior (Bradbury and Bennett 1974b). However, only 2 cases of specific host odors being used by black flies for host 46 47 seeking are known. Simulium eurvadminiculum Davies is attracted from long distances to the extracts of the uro- pygial glands of the common loon (Gavia immer Brunnich), its primary host (Lowther and Wood 1964, Fallis and Smith 1964, Bennett et al. 1972). Carbon dioxide alone was found to be a poor attractant for S. eurvadminiculurn but in combination with uropygial gland extract, there was a 3- fold increase over gland extracts alone, suggesting some sort of synergistic effect (Bennett et al. 1972). The "forest" form of S_. damnosum Theobald s.l. is attracted to human sweat, although addition of C02 also increased at¬ traction (Thompson 1976 a,b). Middle-range orientation is defined as occurring when carbon dioxide initiates a more precise orientation to the host, though vision may also be important in this phase (Bradbury and Bennett 1974b). Smith (1966) hypothesized that middle-range orientation also involved vision but only for navigation. There are numerous cases where carbon dioxide has been found to be an olfactory attractant to black flies (Golini 1970, Golini and Davies 1971, Bennett et al. 1972, Bradbury and Bennett 1974b, Fallis et al. 1967, Fallis and Raybould 1975). The distance over which C02 is attractive to black flies is not known for certain. Golini and Davies (1971) captured significantly fewer S. veQustum on traps 4.57 m from a C02 source compared to 48 traps at 0-3 m -from the source. Bradbury and Bennett (1974b) observed a large drop in simuliid captures on traps 7.75 and 8.75 m downwind from a C02 source. Close-range orientation is the final approach by the black fly to a landing site on the host. It is believed to be primarily a visually directed behavior (Bradbury and Bennett 1974b, Wenk 1981). Studies have shown that host or host model color, size, shape and movement can affect close-range orientation. However, for a given species, it still is not known exactly which factors are most critical and how they are perceived during orientation. Most studies have employed 2-dimensional targets of various sizes, shapes and colors. In general, it has been found that black flies are most attracted to dark colors (blues, blacks and reds) of low intensity and that they land more on areas of maximum contrast with the background (e.g. edges, points of convergence) (Wenk and Schlorer 1963, Peschken and Thorsteinson 1965, Bennett et al. 1972, Bradbury and Bennett 1974a, Service 1977, Browne and Bennett 1980, Cupp 1981, Wenk 1981). Many black flies orient to the undersides of their natural hosts (Breyev 1950, Davies 1957, felenk and Schlorer 1963) which Davies (1972) hypothesized was a response to dark, shaded regions. Wenk and Schlorer (1963) and Bennett et al. (1972) conduct- 49 ed the most conclusive experiments suggesting vision was important in close-range orientation. They showed that certain black fly species oriented to regions on life-sized 3-dimensional model hosts that corresponded to feeding sites on natural hosts of those species. Their study did not exclude olfactory stimuli, however. Based on their own data and interpretation of Hocking's (1964) review of the functional morphology of insect eyes, Bradbury and Bennett (1974b) hypothesized that close-range orientation of black flies begins between 0 and 1.8 m from the host. Wenk (1981) pointed to the need for more studies on the role of vision in close-range orientation in order to better understand the host—selection behavior of black flies. Many black flies arrive at hosts but do not land or feed, suggesting that long— and middle—range orientation stimuli are present but close-range orientation, landing, crawling and/or engorging stimuli, are not. In Massachu¬ setts, this is particularly true for St. mutata, S. ienninasi complex, §. vittatum and S. venustum when attack¬ ing humans (Appendix II). Many black fly species locate and blood feed on specific body regions of their host, particularly areas where hair density is low such as the inner ears and under—belly regions (Wenk 1981). It would be of interest to know how these areas are selected. The purpose of this study was to investigate the role 50 of carbon dioxide and visual stimuli in host-seeking beha¬ vior of S. venustum and _P. mixtum/fuscum complex (hereafter referred to as P. mixtum) in MA. Simuliurn venustum is a pest of humans in MA but it is not an aggressive biter (Appendix II). Prosimulium mixtum is an aggressive biter of humans (Appendix II). The specific factors I investi¬ gated were: 1) feeding sites on dogs, 2) landing sites on humans in a normal upright stance and a quadruped stance, and 3) effects of carbon dioxide and host-model shape, visibility, color, and color intensity on the number of flies captured and fly landing sites on the model hosts. Materials and Methods Study sites. Trapping was conducted at 2 locations ca. 2 km apart near Lake Wyola and the West Branch of the Swift River in Franklin Co.. Sites were ca. 1 hectar fields of short grass surrounded by mature mixed coniferous/deciduous forest. Both fields were within 400 m of S. venustum CC and P. mixtum breeding sites. 2-Dimensional model hosts. The 2—dimensional models were 1.2 m long X 0.4 m wide X 1 cm thick plywood boards covered with black polyethylene. They were placed on 2 legs 0.75 m above the ground in either a vertical (board perpendicular to the ground) or horizontal (board horizontal to the 51 ground) position. Legs were 2.5 X 2.5 cm wooden stakes painted black. Each side of the model was divided into 9 sections of equal size (40 cm long X 13 cm high). On the vertical model the number of flies captured on the corres¬ ponding sections from both sides were combined. On the horizontal models, collections on the top and bottom were separated. 5-Dimensional model hosts. The 3-dimensional models were 1.2 m long X 0.4 m diam. cylinders made of chicken wire covered with black polyethylene. The ends were plywood painted the same color as the model (Fig. 3A). On the horizontal models (cylinder horizontal to the ground), 4 legs (same as the 2-dimensional models) were set at 30 degree angles so the bottom of the cylinder was 0.75 m above ground. Legs were painted the same color as was used on the underside of the model being tested. Upright models (cylinder perpendicular to the ground) were the same cylin¬ ders as the horizontal models but without legs (Fig 3B). They were placed on one of the plywood ends 0.5 m above the ground on platforms the same diameter as the cylinders. The circumference of the horizontal models was divided into either 4 or 6 sections of equal size, depending on the test. These were called the circumferential sections. On models with 6 sections they were labelled as follows: u (the underside region where the legs were attached - the > 52 Figure 3. Horizontal cylinder host model (A). Broken lines depict the circumferential and 3 longitudinal sections into which the model was divided. e, longitudinal end section without the C02 canister; ec, longitudinal end section where the C02 canister was placed; m, middle section; u, underside circumference sections; lrf, lower right flank sec^-*on5 urf, upper right flank section; t, top section; ulf, upper left flank section; and Ilf, lower left flank section. Upright cylinder host (B). ts. top section where C02 was placed; ms, middle secti on; and bs, bottom section. The circumference was divided into four sections of equal size, which are not labelled on the diagram. Models were 1.2 m long X 0.4 m diam. 53 A B 54 seam where the polyethylene was attached was in this sec¬ tion as well), lrf (lower right flank), urf (upper right flank), t (top), ulf (upper left flank), Ilf (lower left flank) (Fig. 3A). On the models with 4 sections the flank was not divided into 2 sections. Lengthwise (longitudinal axis), there were 3 sections of equal size labelled as follows: ec (end with C02), m (middle) and e (end without C02). In the experiment comparing the orientation of black flies to horizontal versus upright models, the horizontal model was divided into 4 zones on the circumferential axis to correspond to the same zones on the vertical model (see below). In the experiment comparing all black horizontal models to the bi—colored models the zones were divided into a top and bottom section. Vertical models were divided into 4 sections of equal size around the circumference. They were labelled counter clockwise to correspond to the 4 sections of the horizontal models. The polyethylene seam corresponded to the under— side section. The perpendicular axis was divided into 3 sections of equal size, labelled as follows: ts (top sec¬ tion with C02), ms (middle section) and bs (bottom section) (Fig. 3B). In the experiment comparing the all black up¬ right models to the bi—colored model the perpendicular zones were divided into quarters with 2 sections of each 55 color. These were labelled from top to bottom as follows: tsl (top section 1), ts2 (top section 2), bsl (bottom section 1) and bs2 (bottom section 2). Trapping procedures. Flies were captured on the models with a 3:1 mixture of Tangletrap (The Tangletrap Co.) and mineral spirits. This is the optimal dilution for trapping black flies (S.R. Bennett pers. comm.). The mixture was applied to the models with foam rubber paint brushes. Within one-half hour of applying the mixture, the mineral spirits evaporated leaving a thin layer of pure Tangletrap. The C02 source was 0.45 kg of dry ice placed in 443 ml insulated plastic thermos bottles with four 2 mm holes in the top. Tests were run on days when the temperature was between 15 and 25 °C. Models were set at 4 p.m. and left until the next morning when they were collected before fly activity began. Black flies were removed from each body zone of the models (see below), placed in mineral spirits and later counted and identified. When counting the black flies on the models, lines were drawn in the Tangletrap to delineate the sections. Black fly feeding sites on dogs. A male and female beagle dog of similiar size and weight were used (38 cm tall at the shoulder, 12.6 kg). The dogs were brown, black and white. They were mostly white on the undersides, black on the back, and mixed brown and white on the head. The dogs 56 were tethered at the study site for 30 min after which they were examined for fresh black fly bites. Bites were readi¬ ly identifiable by their bright red wheal. Body sections were divided as follows: anus/base of tail, belly (from behind the rib cage to genitals), chest (belly to the base of the neck and up to the mid-line to the flanks), back (down to mid-line of the flanks) and head and neck. Simi- liar observations also were made on horses. During the observations of the anti-black fly behavior of the dogs and of the black fly attacking behavior, the observer stood ca. 5 m from the dog and used 8 X 35 power binoculars to aid observations. To determine which black fly species were engorging on the dogs, the animals were placed in a vehicle with windows closed after a 10 min exposure period. En¬ gorged flies were collected from windows inside the vehicle as they left the dog. Three overhead net sweeps were taken at the same time to determine the species composition of the host seeking population that was attracted to humans. Landing sites on humans in 2 different body postures. The 2 positions were upright (standing) stance and quadruped stance (Chapter III). The subjects were two males of similiar size and height (ca. 1.8 m tall, 72 kg). While in the quadruped stance, the subject held his head up so his breath was not exhaled downward toward the stomach. Body 57 sections were divided as follows: waist up to the shoulder blades on the back; waist up to the armpits on the stomach; shoulders, front and back; head and neck; and arms. The dividing line for the stomach and back areas was the middle of the side of the rib cage. The experimental procedure was to have the subject assume a given posture and remain completely motionless. A second person collected as many of the black flies as possible which landed on each body region for a 5 min period. Flies were collected with a mouth aspirator and placed in vials pre-1abelled with each body region (Simmons and Edman 1978). All tests comparing the 2 stances were paired. There was no significant dif¬ ference between the number of flies captured on the 2 subjects. Thus, these data were pooled. Comparison of upright and horizontal 3-D models. All black models were baited with C02 and placed as paired tests 50 m apart around the edge of a field ca. 5 m from the forest. Comparison of perpendicular and flat 2-D models. Two- dimensional models were placed in perpendicular (long side horizontal to the ground) and horizontal (panel horizontal relative to the ground) positions near the edge of the forest as described above. Models were baited with C02 and arranged as paired tests. Model visibility. The effect of model visibility was ex¬ amined by placing a black horizontal model beneath a hem- 58 lock tree on the edge of a field. The branches on the hemlocks created a shield around the model so that it was nearly invisible from the edge of the field. The branches were removed from 1 m around the model. A second model was placed in the field 15 m adjacent to the hemlock. Tests were paired. Both models were baited with C02. Color intensity of models. Models painted the following neutral colors were tested: black, white, light gray (937. white: 7X black), and medium gray (507. white: 507. black). The entire model was painted including the ends and legs. The paints used were black enamel and titanium dioxide white. The spectral reflectance of the paints was measured with a Shimadudu Spectronic Spectrophotometer with a UV 200 integrated sphere attachment compared against a magnesium oxide standard. The spectral reflectance of each color is shown in Fig. 4. The white model was high in intensity, the 2 grays intermediate, and black low. Models were baited with C02 and placed in randomized positions 50 m apart around the perimeter of a field. Single versus bi--colored 3-dimensional models. Horizontal models were painted the following colors! all black, top green and bottom black, all white, top white and bottom black, top black and bottom white, top black and bottom magnesium oxide (high UV reflectance), all green, top green RELATIVE 61 and bottom black, and top black and bottom green. The color pattern included the plywood ends and legs. Upright models were painted as follows: all black, top green and bottom black, and top white and bottom black. Green was a mixture of green and yellow oxide latex paints designed to approximate background vegetation in hue and intensity. Magnesium oxide powder was mixed with Tangletrap, spread over a white titanium dioxide painted surface, covered with Saran Wrap which was then coated with diluted Tangletrap. Spectral reflectance of the colors is shown in Fig. 4. Models were baited with C02. Model positions in the field were randomized. Effects of carbon dioxide. In the first experiment, I compared the number of each black fly species captured on black horizontal models baited with and without C02. Models were placed in randomized positions along the edge of a field and separated from each other by at least 75 m. A second experiment was conducted to determine if C02 affected the landing sites selected by £. venustum on the host models. Prosimuliurn mixtum was not included in these experiments because few were captured when C02 was not present. Blue was chosen because it is the color known to be most attractive to S. venustum (Davies 1951, 1972, Browne and Bennett 1980). The spectral reflectance of the blue paint is shown in Fig. 4. Paired tests of the black 62 and blue models baited with or without C02 were run as described above for black models. Orientation to models with and without legs. The model without legs was the same as the horizontal 3-dimensional model with legs except that the legs were removed. It was held 0.75 m above ground by a 4 m long X 3 cm diam. pole skewered lengthwise through the center of the model and secured to a stake at each end of the pole. The model was stable in this position and did not move. Paired tests on C02 baited legless and legged models were run by randomiz¬ ing their positions along the edge of the field. All statistical tests followed procedures in Sokal and Rohlf (1969). Results Feeding on dogs. The distribution of black fly bites on the 5 body sections of the dogs was not random, as indi¬ cated by the significant Chi-square test for independence (Table 10). Eighty-six percent of the total bites on the 2 dogs were on the belly. The rest were on the anus/tail (11)1) and chest (351) regions (Table 10). On beagles, the belly and anus/tail regions are almost hairless whereas the back hair is longer, more dense and coarser. A number of anti-black fly behaviors were exhibited by 63 Table 10- Number of black fly bites on 2 dogs of the same breed and percentage of total bites on each body region. V. of total bites/body region** X + S.D. dog n bites/rep.♦ head chest stomach back anus/tai1 A 6 54.67 ± 31.42 a 0 3 86 0 11 B 6 6.83 ± 8.08 b 0 2 91 0 7 7. per body region 0 3 86 0 11 ♦Difference between means was significant (Student's t-test P< 0.05). ♦♦Difference among proportions captured per body region tested for independence by the Chi-square test. P< 0.05. 64 the dogs. These included paw scratching, head shaking, and vigorous rolling on the ground. Quick, jerky body move¬ ments were often followed by running, snapping at and often eating flying, crawling or engorging flies. The frequency of these behaviors was not quantified but dog B was highly defensive and dog A was not. There were significantly more bites on dog A (Table 10). Ninety-four percent of the engorged flies collected off the dogs were S. venustum and 67. were P. mixturn. At the time of the tests, P. mixtum comprised 677. of the host seeking population attracted to humans compared to 277 for S. venustum (Table 11). Simuliurn venustum appeared to prefer to land on the undersides and lower flanks of the dogs. Many S3. venustum that landed on the flanks were observed to crawl toward the underside. They also were observed to make touch and go landings along the lower flanks and underside of the dog. Large numbers of £. mixtum also landed on the dogs but most landed on the back regions and attempted to work their way through the coarse, dense hair to gain access to the skin. No P. mixtum were observed to crawl towards the undersides. Those that land¬ ed on the head were driven off by scratching or head shak¬ ing. Observation of horse anti-fly behavior. I observed, but did not quantify, defensive behaviors of horses. During Table 11. Numbers o-f P. mixturn and S. venustum engorging on dogs versus those attracted to humans at the same time. TJ r 0 Cf 3 E c nj N* •P -M H H .0 H .X *4- •ft •H r. TJ 0 it U QJ c 0 > Ql L Ql »B TJ w M c 4J 0 Ql C o 0 L o Ql c ai id Qi Z ai Q id c 0 U >B a ■ • • •iH ■P cni 4J •H •P E X 3 cni E > QJ C 3 id E 4J 3 4J r E] X 3 E > ai C 3 id 3 Ej 0 L ai • » 1 w O K) GO *** GO GO H o H w hO 'H (NK) n in in o O' W (N o K) W (N K) GO N-* in 0* H N o 00 s-* >0 H w H >0 O' N w >0 O' #*■<* OD N N K) O' NO w O' 65 66 the observation periods, small numbers of tabanids and mos¬ quitoes were present but black flies were the dominant pests. The following behaviors were noted: face and neck scratching on trees and fence posts, leg and underside scratching and kicking with hoofs, tail and mane swishing of the back, neck and undersides, trotting and running, and head bobbing. During one observation in July when S.. verecundum and S. jenninasi complex were particularly abun¬ dant I observed a horse, which had been constantly repeat¬ ing the behaviors listed above, violently roll over several times while kicking and slashing with his feet. Landing on humans. Five species of black flies were cap¬ tured landing on the test subjects of which 987. were P. mixtum and S. venustum. The other species captured were £. magnum. S. vittatum and S. jenninosi complex. Too few of these species were captured to draw conclusions about landing site preferences on humans. For both P. mixtum and S. venustum. there were no significant differences in the total number of flies captured when the subjects were in the upright vs. quadruped stance (Table 12). More S. venustum were captured in the quadruped stance. The re¬ verse was true for £*. mixtum. The Chi-square goodness of fit test of the frequency of each species captured within each body region was significant for both body postures 67 <•> > * ♦ * * Ql Ql Ql a III * ♦ ♦ ♦ L u i 0 L * * * * * Ql c Ql 13 3 * * * ♦ ♦ 2 IQ J 0 * 00 m O N u L 4- N >0 (h m N ■H 0) 1 X • • t 10 4- c a -* N cs >0 P ■H o H •O N >0 c c •H TJ 4 w "t fQ ai a a Qi £ E "tH • L Ql U P 10 L m 3 >0 <1 £ |Q £ |Q wo P TD * E >0 O' M (O Ql a ai H * O N M 00 L L X ow ni a TJ P P 0 ■o u 3 Ql 10 4- 0 V L L £ £L II TJ 3 c TJ Ql |Q P 10 ft Of OS £ 01 L 4J H Ql 3 Q E CN O o H Ql p Ql IQ a Hi L V) H 2 10 a £ U ft P Ql 4J Ql u Ql P a r 4- 3 10 |Q £ £ Ql Q) 10 Ql TJ OS ft ▼H (Table 12). This suggests landing was not random. The percentage of the total captured on the back decreased significantly from 30% to 1% for P. mixtum and from 38% to 1% for S. venustum when the body posture was changed from an upright to a quadruped stance (Table 12). Conversely, the percentage of the total flies captured on the stomach increased significantly from 16% to 79% for P. mixtum and from 22% to 83% for S. venustum when the body posture was changed from an upright to a quadruped stance (Table 12). More S. venustum and P. mixtum were captured on the head and arm regions when the subjects were in the upright stance compared to the quadruped stance but these differen¬ ces were not significant. Black 3-D models. Significantly more P. mixtum were cap¬ tured on the upright model (X = 368) than on the horizontal model (X = 249) (Table 13). More S. venustum were captured on the upright model (X = 160) than on the horizontal model (X = 98) but this difference was not significant (Table 13). The Chi-square goodness of fit test on the frequency of each species captured in each of the 4 circumferential and 3 longitudinal sections of the horizontal and upright models was significant, indicating that selection of land¬ ing sites was not random. On the 4 circumferential sec¬ tions of the upright models, the percentages of the total females captured ranged between 22-27% (Chi-square test, P< Ql tt L Oi 1 £ p * * ♦ 4c 4c 0 L c P 0 * ♦ * * 4c 4c 4- ft Ql 0 C 4c * 4c 4c 4c a C £ 0 ♦ 4c 4c 4c L Qi 0 •H CM n CN O TJ Qi TJ P X in K) 4) m 0) P C TJ H u Ul o H CN P P -*i at Ol 01 c K) U> Ql L TJ Ul 0 Oi -H 3 0 T« P L P E p Ui 43 10 4) 0* P 0 Q TJ u CN CN H c 4- n 4J 0 Qi Qi Qi u £ £ U) L L TJ ai N r CD 0- CN o Ol Ol Ql E •ih T3 • i ^ CN K) 2 4- P 3 L Qi CJ 4- 0 P a L c •H Qi It 3 3 0 u 43 O CN H TJ TJ P 3 P l-H UJ in m in in Oi C TJ a L >t C ai C ft * * 4c 4c 4c 3 0 > ft u * * 4e 4c 4: P > •*« * ♦ 4c 4c 4c aa CP ■ H 0 * 4c 4c 4c ft Qi ail Qi CM io 4) N N U TJ L p H X Ch H H CN 01 T3 c ft Ul 10 Ul 2 > C 0 E c H H Ql 0 TJ at N Ol 0 •H pH 0 ■r* 4- lH u. O' N in H U r-i £ E L P -1 K) (N H CN Qi 0 3 0 H u a 4- L P £ ft Ol 0 Qi X P m 1- N «* m 0 a -rH .X 0 CN CN £ L E U P • • U Qi TJ ft E Ll N O' «■ Ul ft £ Qi ■ H 4* 3 a CN CN c Ql E L Ol|£ 0 u 0 3 3 L •H 4- Z P at *4* •iH D H CN 10 O p 0 a • H 0 u *H CN N ro it ns u in *o Qi L • u o E Ul Ul Qi P ■ Qi •H Ul ft £ ft ft £ 0 0 o 4- X Ol * N IN in 43 > E 0) c ft H * H 4) 00 N TJ 3 P 0 V H ft a • • • t O C 1 -»■« Cl ft H E Qi 4) in 00 H £ 3D • p P ft Oi L CN o H 10 pH ^ L r* 0 c 4- \ H CN H H Qi «t >in 0 Qi P TJ £ P Ql o a 3 TD • Oi + +1 +1 -H P o c ■ 0 pH ai 3 Q L P P o L • ft £ P • 3 O 10 o O 4- ap > p •H cn p 4J io O' 00 O Ql £ V 0 CP a • ■ • • £ 3 Ol Ql Ql Oi s- c +i »t 00 00 N c P 1 ~ £ P L 0 0 u * 43 O' in 0 C P >t H 1 X CN K) H •*■4 ai C P Oi 3 Ul p C ft c L 0" O a Qi Z Oi CJi it 0 ns TJ P p •H Ol L C 3 1 ■p C pH Oi • £ * £ L 2 Qi Oi 0 CJ c ft Ol u N CP N CP U P £ 4- E 0 £ qj TJ c •H i-t •t-4 •H m QJ P 4- it 1 U u • 0 ft L L L L 0) £ *r4 •H L -< CN E p 0 a 0 a TJ 3XTJ £ P ai ft □ Ul £ 3 £ 3 £ 0 U c 0. •H U L 0 > Qi ft P O Qi Ql f-« U Oi u C £ E 4- UUP C £ Ol «P 3 c c c 01 P 4- • L «H E P P Qi ft ft L •H K) Of 2 3 Ul X L U U flj > c H 4- Ul P 3 Qi Oi H -H 4- £ D> E TJ QI X C P 4» 4- 4- 4- •l-» ai 3 Qi •i-i •H Ol 4* 'H -H •H Oi cn H U P u E > Oi •H c C Q U 4c X] L tH Ol Oi Q cr cn 4c C 4c ft •H ft a • • 0) 4c *H -H 4c ft 4c i- U £ Ul O-i coi 4c 4c 0 0 4c TJ 4c 70 0-05) and 21 - 30% (Chi-square test, P< 0.05 ) -for S. venustum. The closeness of the total females captured in each section to 25% suggests orientation to the models was nearly random, despite the significant Chi—square tests. Mind direction would have influenced upwind orientation but no directional measurements were taken. There were large differences in orientation of both P. mixturn and S. venustum to the circumferential sections of the horizontal models, as indicated by the highly signifi¬ cant Chi-square values for the percentage of the total captured per section (Table 13). Forty-four and 397. of the total P. mixtum captured were in the 2 flank sections, 11% were in the underside section and 7% in the top section. Seventy-three percent of the total S. venustum females captured on the horizontal model were in the underside section, 9% and 15% in the 2 flank sections and 3% in the top section. On the longitudinal axis of the horizontal model sig¬ nificantly more P. mixtum and S. venustum were captured on the end with the C02 canister (section EC, Table 13). Eighty—eight percent of the total P. mixtum captured on the upright model were in the top section. Fifty—one percent of the S. venustum captured on the upright model were in the top section vs. 22% and 26% in the middle and bottom 71 sections, respectively (Table 13). On the horizontal models, 56% of the P. mixtum were captured in the CD2 section vs. 187. and 267. in the middle and non-C02 end sections, respectively. Fifty-one percent of the S. venustum were captured in the C02 section versus 307. and 19% in the middle and non-C02 end sections, respectively. The difference between the percentages of P. mixtum and S. venustum captured in the section with C02 versus the end without C02 were significant as determined by a Mann- Whitney U-test Visual observations of near orientation flight beha¬ vior of Ej. venustum to horizontal models, dogs, horses and humans in a quadruped stance indicated that when they approach to within ca. 1 m of the host or model they fly downwards towards the ground and then up toward the belly/- chest region. On quadruped humans or horizontal models without Tangletrap, S. venustum foraged on the belly/chest region by making numerous touch and go landings every few centimeters. Many landed and crawled on the skin for sever¬ al seconds (Chapter II). In contrast, P. mixtum flew directly to the host or model. On humans, P. mixtum either immediately began to engorge or flew away. On the beagles, they landed on the flanks or backs and attempted to work their way through the hair to the skin but did not crawl to the undersides. Observation of the horizontal models 72 baited with C02 showed that P. mixturn tended to orient to the C02 source and then -flew in a circular pattern near the outline of the flanks. They did not orient directly to the underside like S. venustum. 2-D models. Significantly more P. mixtum and S. venustum were captured on 2-dimensional models placed in a perpendi¬ cular vs. a horizontal position (Table 14). On the perpendicular model, the landing sites chosen by P. mixtum and S> venustum were not random, as indicated by a significant Chi-square test (Table 15). Seventy-seven percent and 427. of the total P. mixtum and S. venustum. respectively, captured were on the 3 sections along the top edge, which was significantly higher than the percentages captured on the middle or bottom sections (Table 15). Fifty-nine and 857 of the P. mixtum and S. venustum. respectively, were captured on the underside of the hori¬ zontal model. This was significantly higher than the per— centage captured on the top side (Table 16). Models in the woods vs. field. Significantly more P. mixtum and S. venustum were captured on horizontal models placed in the field vs. in the woods (Table 17). The differences in the number of P. mixtum captured on the lower and upper right flank and top sections of the woods vs. field models were significant (Table 17). There were 73 Table 14. Number of P. mixtum and S. venustum captured on vertical and horizontal black panels baited with carbon dioxide. X ± S.D. females species treatment n captured/rep.* Jp. mixtum horizontal 12 36.58 ± 27-26a vertical 12 204.25 ± 113.38b s. venustum horizontal 12 3.42 ± 4.58a vertical 12 40.25 ± 63.47b ♦For each species, difference between treatments were tested for significance with the Wilcoxon's signed-rank test for paired observations. Numbers followed by different letters are significant at P< 0.05. 74 c 0 * at * L *0 o at 01 H 00 L as 2 3 P CN N pH u ID P 0 <* CM 10 at a P CM •H as X u u at a to U TJ CM s0 00 >0 9 ai I C pH CN H C 0) a <0 0 u c <8 at 0 c ID 18 H K) L E 0 3 n *□ 10 o IN 0 3 * c rH TD ▼H P P TJ C at ■ as •i-4 0 at o > a p E c 3 L -r4 os c 0 C 3 P E u 0 1-4 at P u 3 N p > a at •i-f H **H N N N «-4 u <8 tA H os L if? pH pH pH DO at u ^ 3 p 0 U Ol 0 • rH E 0 a cn ut at ■iH p a • w at c cn pH os at o ft as it at E U) L. pH c as a U 0 rH 0 rH P a (IS • E •«H 1-4 a TJ P as as at at p 0 TJ P p a p E o P 0 t s p 1-4 0 0 u 3 •H 0) E a p as > P X r-i at > at u X 0 as o * c •«-4 •H P -H * c at E TJ o ut pH 1-4 O' p H o c C .x OS N CM pH a -n • 0 at u p N pH pH ii at p 0.10 a as 0 L C L p n p CM 3 0 rH ITS X P U ns U P TJ E CM a p o c 3 o ID □ as io o a as P 00 •H U T3 O' u p P at X in X I C 10 M 10 X •H cn p •r4 as C Oi K> 0 at 2 18 c E ii 0 C 10 a P 0 rH c 0 p TJ C L E • as 1-4 as at at p 3 a p p p p u ■f-4 os c 0 L >• 0 •H L H u 0 pH in o a as at 3 N TJ 00 IO CM a at a a E H lH TJ pH 0 at cn •lH os L •rH L u ns in ID p 0 6 a c p pH 0 0 -C at c at L p a at c Cl w cn c u as C 0 L a 60 s0 o a at o c <1 in uo 10 E 0 a. U OJ CM as a u c os iH H 0 • JQ U L ■ in C O in H pH pH 0 P o os os * L • u u in at E Ut 0 a o at •H •iH *H rH 0 rH P 0 pH p P X TJ P as p p V a L l ns a TJ P p •p4 0 a «TS 01 Oi 0 •H 0 0 Q 0 ♦ H- > > p E a p * P ♦ 75 Table 16. Number of P. mixtum and S. venustum females captured on the top and bottom of horizontal black panels baited with carbon dioxide. 7. of total females cap • total no. species n captured top bottom X2 P. mixtum 12 453 41 59 15.28* S. venustum 12 41 15 85 25.06* *Chi-square test -for goodness of fit, P< 0.05. 76 l c * * o> * 01 c in L Ql 0 TJ Ql L c f if if £ 2 QJ -< 01 0 lii H in GO U •«H 2 it K) K) CN K) P U £ pH u if if if if 01 in u a Ql O' H ns O' CN •Pi c in w H CN H ai * u • 0 c C ai in •H • if if if If a c p in • C If in o oi H in 0 O in in o u c in 01 c ID in in •rH Ql —i p 01 0 o TJ TJ ns u P in h • ai h £ if p o E ai U > L X ai in ns > if 3 0 if ns ns if L TJ > V 4- in in N CN QJ 01 P H >1 in rH O L 0L a-o CN rH H in £ ai pH TJ c L £ m i« 0 0 in p O 0 C £ If u c P £ •«H if if ns If o 0 V 4- ai o P IN >0 <* TJ ai p E £ P TJ u CN CN 3 L in ai 2 TJ c 01 01 P L P ns 4- L in P Ql Ql C H Ql L L in u 0 3 if £ if 3 If ai ns £ -rH QJ P ai K) m O o E a C £ X QJ u H if 4- p p TJ a 4- 01 if . c lf L •|H > U Ql Ql Ql O L L TJ 2 L if £ if Ql ns L 4- 3 O cni *a o CD IN K) p P 4- > 4- CN CN p a rH 0) E TJ P c in If p p 3 01 QJ u in c C h U if £ if if ns ns 01 P Ql if L K) O' GO K) 2 in p u 13 rH rH »H E P -X 01 •rH 3 M 01 c •pi -X 4- £ If u c •H P P if if if X in If L Ql 01 c N O' GO TJ 1 a l o> o >0 N) r 01 TJ in i •Pi L Ql TJ in 3 C £ Ql ♦ . 01 pH * £ if £ P O' u c Ql if O' >0 if a h If Q) L Q.I "0 ns >0 GO IN 0 if in Ql •* if p in a O' • • CN u in ai e o ai ai • • H K) ■ in 4- L P pH L «0 rH O' in L o in Ql ai .x It X N CN K) Ql . E U c P • E TJ 0 £ C L C P ai ns Q Ql Ql +1 + •H 4- rH +1 ■H E O 01 O Ql • 4- L P 3 x £ X rH £ cn 3 CN H N) in u C O E O • p rH >0 O' o 01 U flj rH • 3 U p ■H 0 a • • • in rH C -H c p ns c <0 >0 rH >0 O' rH *H •pi Ql □ p lx u in <* H- 4- if 3 PH 3 L P c CN 0 0 P If 01 4- N O >• p > 4- c h- CN CN c P £ O £ 4* 0 H CN IN IN CN L 0 P •H •«H 01 01 0 rH 01 TJ ai£ P C U C U 01 a ai c 01 c If TJ •rH ai ns QJ if P H- "O 0 in TJ in TJ C Oh L 2 u 2 U >* E TJ H pH TJ U P H P rH £ Ql 01 o Ql o 01 in U C QJ 4- Ql 4* o 0 •rH Ql £ H £ -rH TJ L 0 c in 2 2 01 H o o 4* TJ C c 01 Q. P o> o> 2 •H £ L in h 01 H O U L p 0 ai in u in rH Ql 0 ns E 4- in u c . rH u 3 C L 01 L O in 0 P P 01 0 TJ ut L O 4- 0 X L 4- QJ 4- 3 01 01 4- in ns 0 01 C P 4- TJ Ql C 2 •H 4- TJ L Ql 4* 01 •H QJ 01 U > H TJ 01 01 •H P Q P £ 01 Ql q in * in E no 3 £ a 01 H P P cn ♦ * 01 3 1* * * p * p Z 77 no significant differences in the numbers of S. venustum captured on each body section of models in the woods vs. field (Table 17). The differences in the total of either species captured on each of the 3 longitudinal zones of the models in the 2 location treatments also were not signifi¬ cant (Table 17). Carbon dioxide. Significantly more P. mixtum and S. venustum were captured on black horizontal models when C02 was present vs. when it was absent (Table 18). This also was the case for £3. venustum when blue models were tested (Table 19). The only significant differences in the percentage of the total S. venustum captured on the circumference sec¬ tions in either the blue or black models baited with or without CD2 was on the underside (u) and lower right flank (lrf) sections. Fewer flies were captured on the u section and more on the lrf section of the black model with C02 vs. the model with no C02 (Table 19). Dn the longitudinal axis, a significantly higher percentage of the total S. venustum were captured on the C02 canister section when C02 was present compared to when it was absent on both the black and blue models (Table 19). This was not the case when C02 was present (Table 19). Too few P. mixtum were captured on the models when C02 was absent to draw conclu¬ sions about the differences in orientation compared to when 78 a: a p as u os u os L os as as 11 a (0 * a n a £ • d H CM N a L as a H o L C as c 4- as • • * O as a 0 L • m m 3* 4- L p • S o H CM 1 0J XJ o a 10 a H as c as +1 +1 +1 +1 P as L L C c p 3 • 3 N K> N N m O' c P • Q P H CO H H E •rt as a as a • ■ • ■ P (0 L as a . os o O' m 00 OS as u ^ cn u H 10 ai 0) 4- x L E 0 +i P C 3 0 XJ P IX C X ai Hi o as 3 c as u C 0 2 i—i ai a P •h a > L 01 3 nj a XJ u a 111 XJ p 2 cni-u Ql 0 L 2 »H ■o o 3 c a P o as p a *+- as E u in in 3 CM IN L O •P L us as X 0 L 4- a as •H E a C 3 V E CTZ CL •tH 3 •*« •, 2 C 0) p Q-l p CM CM as a c □ □ 01 L • 4- as 01 CM u CM U a O 01 P O P E O Q p 4- c c •H P U a U a o as as OS p p c a •H L l a as O •H O •H 01 as p as as L c 2 C 2 as p as 4- a oi P 2 01 > 4- E r-< P as lL -h 3 QJ as as Z T3 a os 0 at a us L o as as P • rH u 2 x> C CD as E c as as «-« p 3 as L u c ui P L os 0 as X as as as h* OS N •H •h a *H H »H U u c XI L as as L CD as o a Q a o -h h a ui &-I cn| * 01 4- 01 Table 19. Mean and percentage of the total S. venustum captured per body section on blue and black model hosts baited with and without carbon dioxide. M -P 4- 0) 4J x -P 0 H X pH *H 0 L o a * TJ E U QJ Ql >8 -P Ql 0 iff c tj u n a * 3 X CD <8 Q QJ +| QJ ,, £ t H • 81 4- -t-« 4- •P ft c O -P 4-1 ft u L U 3 £ QJ L QJ u c QJ QJ u * * * 0 c c a QJ u 0 c IS 4-1 -P 4-1 TJ \ D -j a: Ll D L QJ d £ cc L. at c * * D -I U. -I u. c u 18 a 3 u u L Qi L a Ql uJ • U □ CM H H >0 00 K) H H H O N O K) +i CD 3 aj C O 0 m H CM O 18 K) O >8 <8 >8 <8 <8 18 18 18 18 • • a 4J u O CM >0 ** K) 00 CM O* 00 a H H N N +1 3 QJ O CM H in CM K) CM 18 in a >8 18 >8 )8 >8 18 18 ■ s -X U O CM 00 K) H 9- CM H O' O O +1 O <8 u O c 0 O >0 K) GO CM O 18 CM i8 18 18 18 f8 18 >8 18 >8 t • .x £ ■P U O CN 0 H N >0 in H H K) a a m o h- a O i8 +1 O u O o a n0 CM CM H 00 a 18 18 18 18 18 • • CD 4-1 -P 4- TJ •H 4-1 ■ft * 4- -P •H OJ Ql QJ X 0 L QJ 8) U L a c 0 O 8) QJ u 0 c • Q 81 •H 4J 4- QJ 4- TJ a •P H ♦ OJ * 4-1 a -P C TJ 81 a a* -P Q| ■P -P 4J h- a QJ L 4- QJ 0 C L a o U QJ 18 81 ft *■< > QJ 4- 2 U QJ 18 C U C 2 3 i-i 0J c £ 4-1 QJ L 3 8) |H U O 18 x a O 3 ' L 81 QJ 0J C 2 C QJ at L 18 QJ £ C QJ 0 >8 L iH qj oj 18 TJ u U QJ 0 L 0 (8 L -P 2 C L • at w QJ H TJ -P 4- a o» c ft 0J U L 18 c 18 OJ 0 L L 8) I 0 Q a a qj4-i 4J 0J * ♦ a * u -P C •H C TJ U a TJ -H U 4J □» -P 18 4-* QJ 0 LrtU QJ c U -P 4J ID Ql L 4- 2 0J C 0J C 4-1 3 01 QJ 4-» E TJ QJ 4- L 0 81 L U 81 18 ft a qi C U 3 H L 4- Qj ft QJ 18 2 U H 0 3 81 QJ U 0 81 C QJ 2 C QJ C I L C Ql -O Qj .x (8 £ 4J QJ C (8 18 0 L oj a >8 18 u 0 L 2 81 18 01 u 0 x 0 C 8) L TJ ■ft L QJ 8) QJ L > >8 0 C 8) • a 3 £ QJ L 18 followed by a different letter are significantly different at P< 0.05. 79 80 it was present. Horizontal models with and without legs. Significantly more P. mixtum and S. venustum were captured on the models with legs compared to models without legs (Table 20). For all but 2 sections, there were no significant differences between the percentage of the total P. mixtum captured on any of the 6 circumferential or 3 longitudinal sections on the legless vs. the standard horizontal model (Table 20). The only circumferential section where there was a signifi¬ cant difference between the percentage of the total S. venustum captured was the undersides where fewer females (44%) were captured on the legless model vs. the model with legs (71%) (Table 20). On the longitudinal axis, there was a significant difference between the percentage of the total S.. venustum captured on the end without C02 (36%) on the model with legs compared to the model without legs (18%). More S. venustum were captured on the C02 section of the longitudinal axis on the model without legs (68%) compared to the model with legs (56%) but this difference was not significant (Table 20). Color intensity of host models. Significantly more P. mixtum were captured on the black models than on the gray or white models (Table 21). Significantly more P.. mixtum were captured on the dark gray compared to the light gray Table 20. Mean and percentage of the total P. mixtum and S. venustum captured on each of the 3 longitudinal and 6 circumferential sections of black horizontal models with and P H P •H P o» QJ a 0 3 X H 4- n-t P N QJ 4* L P XJ P a E oi QJ a 0 V H 0 0 3 •H P P * 0 C >\ ai U a U ft QJ IX •4J UlEO Q hL fl 4-L • QJ • as. •H P •H P * * * 0 C IA u c oi a 01 QJ u 0 c a a a d Qj Q| C (Q o a tu u U. u E li. •J LU H D J D a: Ll J CL U. p P P ♦ •H * u 3 L a £ c QJ at U QJ a a a QJ IN is H in H GO in IN IN a- M H IN K) K) IN (N a IN M >0 a a 4 a a a a 0-1 •f* P a a c o QJ O' E E X 3 • • K) K> is IN o IN K> H >0 «* N K> GO M M GO IN H >o is m M N >0 a a a a a £ a P +1 a a a a at Oi H GO CN »■« a* is o GO IN in 4 a- is <* * n is O' GO O a a a a a a a P +1 a p Ml a p a a c o 01 O' 3 E C 3 a > Qt K> m >0 P IN H o in H >0 n IS in p K) O n GO K> a a a a a p p a o> a QJ • » 4* P •H ■rl •H p p 0) p P 4- 0 c a a u ♦ c 0 a U a 0 L a L 0 a a X a a • 4- •H 4- •H ■ri P P p p a 4* p a P CL o> u C U 0 L a c a QJ p a a a QJ a P H P c a X L u > a a u a Q •H 3 4- H 4- r-< a l a h a QJ 0 L P L -H a L P a u 0 a l a 3 a a l 3 E P * P ♦ >> C L C .X a oi x a a p a i c 0 u c a * a a u L 0 a x p P 4- 0 c a a c a a • P H •h L p P P 4- p p a p p •H X 4- u P o a C P c 1 U P 0 a u a a o x a P -H a a a a l P 0 P L a -x L c p a a a 3 O' a p U 0 a c u a q a •H P 4- E 4- a Q. H qj o a -h QJ U u >» a a p L -H a u o c * a a C -H 3 0» E c a 4- * p * L C a a p a X P a c U X a a l a c L P i n • 3 a 0 L 4- c 1 a a a a r a P •H > a 0 c Z P a QJ L 3 E followed by a different letter are significantly different at P< 0.05. 81 82 and white models. There was no difference in the number of P. mixtum captured on the light gray vs. white model (Table 21). There were no significant differences in the percen¬ tages of the total P. mixturn captured between any of the circumference or longitudinal sections on any of the 4 colors tested when compared to the black model (Table 21). There was no significant difference among the means of the number of S. venustum captured on the black, gray and white models (Table 22). The percentage of the total S. venustum captured on the underside, top and upper left flank sections of the circumferential axis of the white model were significantly different than the percentages captured on the black and gray models (Table 22). There were no significant differences in the percentage of the total S. venustum captured on the 3 sections on the longi¬ tudinal axis of the models (Table 22). Bi-colored horizontal models. Significantly more P. mixtum and S. venustum were captured on the all black model compared to on models painted the 7 other colors tested (Tables 23 and 24). The fewest number of either species were captured on the blacksmgo (high UV reflectance on the underside) model. The next lowest numbers of either spe¬ cies captured was on the all white model and the black:white model. There were no significant differences in the number of either species captured on the black:white r iH Qj U C L 3 U 44 E |Q 3 L Table 21. Percentage of the total P. mixturn females captured on eac i h ferential and longitudinal section on horizontal model hosts painted colors with varying intensities. M a 4- L 44 T3 44 XJ -M TJ 4- 0 3 0 X It 0 E QJ iq u uj o 0) 0 m h 4J 44 * u n flJ c IX -3 cn E\ Q hqi +14-01 • ata • IQL 4- 44 ft 44 -1-1 c a 0 4J * * QI TJ in • U * L 3 E ai c u Qi L ai 0 at u 0 c 0 o u ID u 0 ai c 0 -J ferential and longitudinal section on horizontal models painted neutral c 0 varying intensites. M 4- P p H 0 0 ns 01 E os QJ in "C £ X3 P •H ♦ 01 G > in Ql U 0 C IX .3 tfl E\ Q HQJ +1 4-Ql • flUL ■ qja •H 4- c a 0 P QJ *0 P •H in • u L U 3 E QJ L p QJ c U Ql in 01 U 0 c 0 in c CJ LU in 01 u • u pH u 0 D a: Ll 0 L -J D cn * Ll * c u h- u L D Ll -J IL r 13 H X H *H 10 O O CN N 0s H <0 O H K) na u +1 N 10 >0 H H (8 in CN >8 >8 H h* K) JQ H na nj na It ■ it ■ na TJ H H CN CN <* m CN H H N E in H n 0) O' L +1 H na CN *H H H H CO na H CN 18 *0 na K) >8 10 J8 OS • 05 <8 ■ na • *8 rH P H H 10 H N 10 r) >0 o 00 H O' +i K> H L na > in 10 H H O na CN >8 CN N »8 10 H na >8 J8 • na na • na • na £ •H P H H H >0 o O H GO GO 10 H H 2 O' JO CN QJ +1 CN >0 Ch CN 10 n H CN in N na H N na (8 CN >0 01 >8 na >8 • >8 • cn p P 4- 4c TJ •H «p ■H 4- P •H ai QJ at X 0 L QJ in U L a 0 c 0 in QJ u 0 c in • Q P •H 4- in 4- -H 4c C 4c 0 n p X £ rH P -P QJ O' L C TJ -P £ QJ H C 4- U H TJ •H 4- Oj u in na 4- p rH Qj w 2 L QJ 3 H c X C 1 TJ qj in 3 E Oi QJ m L -n T3 H in U P Oi QJ a in 3 L • QJ 0 C OJ QJ L QJ c > u 0 0 L QJ E 0 QJ in 2 ai L QJ ■p rH c os a c •H *iH 4- H-H 4-0 4- h oicn 4c 4-0 4c Oh 4c LO'0 p > £ E P xu 4- rH0 P 0C -P •P 0 onu £ 0 P 0Ql rH 2 l cna 0 -Hc C 4-P u -H 0 UP 0 na£ P Ql TJ ^0 ■h 38 C H £ 3H 0 P3 a atp OJ -H l x.a P 0TJ H U LrH OJ 3QJ C 0 na *8o OlrH E qj Ia PPL p 2u 0 3Oi OJ P 0 H 0 P >8 U 0 na tj min a 4-3 na 0 3 £TJ L Ql OJ P C 0QJ QJ na <8 HL U 1E 0 rH01 QJ rH U -H 0 P4- C Ql4- OJ QJ n c qj naoj £ CL £ rH •H 0 C 0 u .X +J Oi L OJ P L >8 C 0 QJ •U P oi E c O' a 0 L U 0 c OS O' 0 0 • Z n 3 E QJ L in o o in followed by a different letter are significantly different at P< • 84 Table 23- Mean number and percentage of the total P. mixturn captured on the circumferential and longitudinal sections of variously colored horizontal ■o H 0 01 0 £ rH 4- p p 4- rH 4- H P TJ £ TJ p * 0 0 ns QJ E as QJ 0 Ql E as in u >« Qi a L OJ 3 0 > 0 QJ u 0 c lx cn Q •H 4- P rH P •H rH TJ •H fH •H P p •H + i u L U 3 E 01 L QJ c as u 0 Ql 0 C 0 3 0 c □ c as 0 u 0 QJ c 0 • • rH p H £ 4- £ P P rH £ 4* U □ CN TJ * •H TJ CN u O TS u 0 0 L c a L L QJ a 01 0 a as 0 0 E as OJ C E c c 0 C QJ 1 1 • • rH £ x. in ro O CN H CD O hO in sO K) Ch >0 CN H nO +1 CN CN os u • rH £ .X in K) <* o in 0- H GO CD >0 N H o in H m +1 * ♦ * ♦ * 01 U M E O' 0 • £ •H p in cfr CN o >0 O in rH Ch N H >0 K) N +1 4c 4c 4c 4c 4c * 2 QJ • £ •H P £ rH in rH H rH .x O O N H N K) O O <* in CN CN K) ro +1 4c 4c 4c 4c ♦ 4c 4c 4c 4c 4c 2 Qi • os u • rH p in £ .X £ •H O O CN rH rH H rH in C> O O CD CD +1 4c 4c 4c 4c 4c 4c 4c 4c M **Significantly different than the black model (P< 0-05) ■ 85 86 Oi • £ * TJ C .X ■P H ♦ C o c QJ c QJ 4c ♦ * •H IQ TJ 0 ♦ * ♦ * ■P L C 0 •»H CN * in m GO O m O U 1 0 E ■p □ GO GO 10 10 in 10 H u u U 10 TJ pH QJ Ql I ai IQ in in c L •P 0 3 C > c signed •P 0 TJ IQ * each se Q N 0 C * IQ IQ fH £ •»H CN H H N O' K) CN c in U L "x TJ CN CN H CN 0 ' 0 TJ 3 c E £ aj •P TJ 0 3 L •«H TJ OJ x •P TJ 3 cH C * ♦ * * * L 0 0 QJ •P c QJ 4c ♦ * * * 3 U 3 L a I N O "0 in H CN 00 £ rH C 0 IQ (N >0 in 10 NO h* a QJ rH u O |Q 3 > 0 U 8 u in u QJ a> £ m « pH o o oji*h «Q |Q f-t a in E •P Qi o pH 3 QJ IQ C TJ IQ 0 4- u £ * * QJ 0 V ■P •iH QJ * |Q 4c U E Ql 0 L pH U) E O 10 CN O' H >0 >0 * L ■P IQ IQ o O' K) CO N O' O' O' NO OJ -X > ■P ■P a u pH QJ 0 Hi ■P (Q OJ £ 4- •P -«H 0 QJ ^ TJ ■P 0 •P X) £ £ 0 QJ c ■P E 4- in £ QJ at 0 c -P L ♦ 4c TJ £ -X 0 QJ OS ♦ 4c C -P u QJ ■H 4* 4- £ O N 00 H O' H IQ IQ CJt+J 0 E >0 H CN 10 £ pH IQ u 3 a rH in £ ■P QJ X U o QJ C C in L ■P TJ H QJ QJ tH 0 IQ £ U pH u E O' •P L IQ * ♦ * * * * * |Q QJ C * * ♦ * ♦ * * £ C a O CN IO >0 o O U OJ • IQ TJ * >0 GO GO CD o *H >0 iq u in £ TJ 3 ■ • 8 ■ • a a a a Ql c c •P c •P a a 0* IO N O' GO IQ 0 |Q •fi . at <* H CN 10 C U -H £ O' CO L 0 -H -P C L C ■H +1 + 1 +1 + 1 +1 +l +1 4- IQ OJ QJ 0 ■H L TJ > L XJ pH QJ O O o o O O o O Ql C L QJ E ix a CN >0 CN >0 nO >0 L O' OJ 4- 3 TJ 8 • • a a 8 8 3 *n in 4- C C io CN in H NJ CN H •p in £ •(H |Q o CN CN a o TJ C IQ L IQ H U 0 TJ >< QJ IQ 4- QJ rH iH Z c in in m m in in in in in l £ ■P L TJ T4 C c .x QJ .x c QJ QJ iq |Q Hi u £ u Ql £ -p a U • L o IQ •»H IQ QJ e in •H QJ Ql H £ H L 3 Ql L 4- CN 4- E £ 2 £ O' C -P 0 •tH E 88 •a aa aa ■« 4- c QJ 3 L Jit .X Qi QJ .x c c x O' pH U o u u •p •P u QJ QJ U QJ QJ model vs. the all white model (Nilcoxon's signed—rank test). Signficantly more of either species were captured on the white:black model vs. the all white model (Wil- coxon's signed-rank test, P< 0.05). More P. mixtum and S. venustum were captured on the greensblack and blacksgreen models vs. the all green model but the differences were not significant (Wilcoxon's signed-rank test) (Tables 23 and 24). Fourteen percent of the P. mixtum captured on the blacksmgo model were on the mgo half (bottom), which was a significantly lower percentage than the 39% captured on the bottom half of the all black model (Table 23). The percen¬ tage of total P. mixtum captured on the top or bottom half of all white vs. all black models were not different. On the white:black model 70% of the P. mixtum captured were on the bottom half, which was significantly higher than the percentage captured on the same zone of the black model (Table 23). On the black:white model 10% of the P. mixtum captured were on the bottom half, which was significantly lower than the same half of the black model (Table 23). The differences in the percentage of the total P.. mixtum cap¬ tured on the bottom half of the white vs. white:black and black:white models were significant (Wilcoxon's signed-rank test, P< 0.05). Differences between the percentages of the 88 total P. mixturn captured on the top and bottom halves of the all green model compared to the corresponding zones on the black model were not significant. Dn the greensblack model, 77% of the total P. mixturn captured were on the bottom half, which was significantly higher than the per¬ centage captured on the same zone of the black model. On the blacks green model 23% of the total P. mixtum captured were on the bottom half, which was significantly lower than the 39% captured on the same half of the black model (Table 23). The differences in the percentage of P. mixtum captured on the bottom half of the green model was signifi¬ cantly different than the percentage captured on the bottom half of the greensblack and blacksgreen models (Wilcoxon's signed-rank test, P< 0.05). The differences in the percentage of the total P. mixturn captured in each of the 3 sections of the longitud¬ inal axis of the all green, greensblack, blacksgreen, blacksmgo and whitesblack models vs. the black model were not significant. The total captured in the C02 section of all white models (47%) was significantly lower than in the all black model (62%) (Table 23). On the blackswhite model 81% of the flies were captured in the C02 section. This was significantly higher than the number captured in the corresponding sections of the black (Table 23) and white models (Wilcoxon's signed-rank test, P< 0.05). There were 89 no differences in the percentage of the total P. mixtum captured in the longitudinal sections of the green vs. greensblack and black:green models (Wilcoxon's signed—rank test). There were significant differences in number of S. venustum captured on the circumferential axis for the blacksmgo and blacksgreen models vs. the black model (Table 24). Fewer flies were captured in the bottom half of the black:mgo (33%) and black:green models (64%) vs. the black model (90%). Significantly fewer flies were captured on the bottom half of the black: green model (317.) vs. the green model (96%) (Wilcoxon's signed-rank test, P< 0.05). The percentage of the total S. venustum captured in the C02 section of the longitudinal axis on the black:mgo, white, white:black, blackswhite and black:green models was significantly lower than on the same section of the black model (Table 24). The difference in the percentage of the total S. venustum captured in the C02-end of the green model (61%) was significantly higher than the percentage captured in the same end of the black:green model (48%) (Wilcoxon's signed-rank test, P< 0.05). Mixed color upright model. Significantly fewer P. mixtum were captured on the green:black and white:black upright models compared to the all black upright model (Table 25). Table 25. Mean number of P. mixtum and S. venustum females captured on upright models painted all black, top half green and bottom half black, and top half white and bottom half black. lx IX ifH TJ cn Q U CLiP P 4- * H cn Q U H cnl 5 H +! E 0 QJ X 3 E QJ E 4 0) 0 at tj q % >1 Oi q * E m 01 ai 0 ■ it « 3 ■ it H TJ \ u 0 0 L L a 0) L a 01 a L a ai ■ £ H .X in n o* CM H o in o* k) CM o K) >0 m O O N +i * ♦ ■H 4 u * * • ■ • • n m CM 'O o N s0 H H o m m 00 * * +i O' C * ♦ L QJ QJ 4 u • ■ £ •H P in o cm o o in fO O' >0 O CM CM O * * +1 2 QJ 4 u * * • • • H £ £ > pH TJ4 .X 4- TJ P H £ £ pH C -x n r-4 4 TJ TJ 0 •H c p * £ 0 -P 0 £ *h ! ■X C * n tj TJ Q) H «P *4- 4- QJ Qi H LUO aj 4£ QJ C0 C P01 •• L 4 P U 4--H C 4 Oi 2 03 QJ C 4 3 U E E L 0 QJ U 0 a 2 3 at l 1 at QJ T3 U £ 0 > E 4 i-* L h Oi U 2 0 aj o> 4 L U 4 E 0 P 01 0 0 L •H •H 4- TJ £ .X 0 OI c QJ L c U QJ 0 QJ X c ai •H a 4 L QI 0 C 0 ■ cn •H •H 4- 4- P £ P £ £ .X cl o m £ TJ v o ♦ * Qi c U 4 c QJ L QJ C 4 C QJ 4 u E 0 QJ • 90 91 There was no significant difference in the number captured on the green:black vs. whitesblack model (Wilcoxon's signed-rank test). On the black upright model 977. of the P. mixtum were captured in the top 2 sections. On the green: black and white: black model only 41 and 277., respec¬ tively, of the total P. mixtum captured were in the top half (Table 26) Fifty-nine and 73% of the total £. mixtum captured on the green:black and white:black models were on the 2 bottom sections compared to only 3% on the same sections of the all black model (Table 26). Significantly more S. venustum were captured on the i black upright model than on the green:black model (Table 25). There was no significant difference between the num¬ ber captured on the white:black versus black model (Table 25). Forty-four and 49 percent of the total S. venustum captured on the black model were in top sections 1 and 2, respectively. These percentages were significantly higher than the 13 and 7% captured in top sections 1 and 2, re¬ spectively, of the the green:black model. There were no significant differences in the percentages of S. venustum captured on the 4 sections of the white:black model com¬ pared to the all black model (Table 26). Discussion My results for P. mixtum and S. venustum show that Table 26. Percentage of the total P. mixtum and S. venustum females cap tured on 4 sections of upright models painted various colors. N* »+■ •p H- H ■P ■P TI \ •P •»h 0 0 nj £ 4c Oi ns ai in u it a 3 L QJ m oi U 0 C •H ■P 4c 4c £ 3 £ X 3 E XI rH rH L £ U 0 Oi 0 H- cn i- cn cn IN a CD 0) CM H- cn a H •X o CN in CN •t It u O' in CN a rH .X CN CN H O in H H 4c 4c 4c 10 O' L at c •• 4c 4c 4c 4c N 01 it U 4c 4c K) 4c 4c 4c K) N 4c 4c 4c 4c 4c 4c 4c 4c 4c 4c 4c 4c a -M rH •H a x H H n N >0 >0 r) <* 4c 4c 4c <* K) 2 01 •9 4c 4c H no u 4c 4c CN 4c 4c H cn -P •P H- TJ •rl -P •rH 4c *+- ■P •rH 01 Oi X Oi 0 L QJ in U L a 0 c 0 tn c 01 U 0 in ■ •H Q 01 S- U 4- CJiO a cin •rH x;c a tjoi -P rHc s- oa 4c 4J-P 4c a ■p Oins a rHu rH -P 26 •P LTJ s- oia rH III> •n -p •P 0 a ccn •P c 01 L L Ol in hc •rH ItX 01 Qjo> c in u ••a mao H- -P c o 01 u a h oj o L -Pc u atin Q) n c a-rH oi in.x no ns CJi U-P oi .xa O £ O QJ ns 0)oi 01 Ol3 U S-u in tjh moo a lx 3 inc L -ft' 01 O' as- o> 01 -HC LUO) in c oi ucr U 0)ITS 0 O' c ns o in TJ ft ■p •H *H It TJ U ‘ft us 2 -P «B T3 C 01 rH >p 01 s- •rH TJ a ft ■p 0 L a ns L 01 0 in 01 L > ns 0 c in • cn •H •tH S- •rH rH •p a •rH s- s- rH 4c 4c 4c -p •p a a X ■p a a pH V CL o o m o c u ns c > L 01 01 c ns c Ol ns u £ 0 01 • 92 93 both visual and olfactory cues are involved during the long- and middle-range phases of orientation to a host. In contrast, Bradbury and Bennett (1974b) hypothesized that long-range orientation was stimulated by specific host odors and middle-range orientation by C02. They believed that visual cues might be important during the final por¬ tion of the middle-range phase of orientation but gave no indication of the actual cues involved. Smith (1966) sug¬ gested that vision was involved during mid-range orienta¬ tion but C02 was the most important stimulus. I found that near orientation was solely visual, which agrees with pre¬ vious findings (Wenk and Schlorer 1963, Golini 1970, Bradbury and Bennett 1974b). Since near orientation is the only distinct phase of host seeking known at this time I feel it is best to divide orientation into 2 phases: long range (e.g. directional orientation from the perching site to the near vicinity, i.e. ca. 1 meter from the host) and close-range (final approach and selection of landing site on the host). Bradbury and Bennett (1974b) hypothesized that initial host orientation (=long-range) was stimulated by specific host-odors. Two cases of specific host odors for black flies have been reported. Bennett and Fallis (1971) ob¬ served that S. eurvadminiculurn goes on host-seeking flights 94 along lake shores in search of its host, the common loon. Bennett and Fallis (1972) found that C02 alone is a poor attractant for S. euryadminiculurn but in combination with extracts of the loon uropygial gland it is highly attrac¬ tive. In comparison to mammalophi1ic species, perhaps bird feeding species rely more on olfaction during host seeking due to the smaller size and often hidden habitats of their hosts. Thompson (1976a,b) found that the savanna form of S. damnosum complex relies principally on vision when host seeking while the forest form relies more on specific human odors. Presumably, the different host-seeking strategies of these 2 sibling species is an adaptation on the one hand, to the open, dry savanna and on the other, to the humid, dense forest habitat. Davies (1972) did not observe any increase in landing rates of S. venustum on blue cloth soaked with natural or artificial human perspiration. More studies are needed to determine the role of specific odors during host seeking. I found that C02 was an important attractant, most likely for stimulation and initial long-range orientation, for host-seeking P. mixturn and S. venustum. This agrees with Bradbury and Bennett's (1974b) model. Significantly more females of both species were captured on models baited with C02 compared to models without it (Tables 19 and 20). Carbon dioxide has been found to be an important attractant 95 for several black fly species (Fallis et al. 1967, Bradbury and Bennett 1974a, Raybould and Fallis 1975, Browne and Bennett 1980). Golini and Davies (1971) showed that S. yenustum orients upwind toward a C02 source. Mochadisky (1956) hypothesized that black flies do not make host—searching flights but instead perch in resting sites and wait for hosts to come within detectable range. Service (1977) agreed with this hypothesis, based on the fact that catches near the edge of a forest are high ini¬ tially but drop off quickly. I have made similiar observa¬ tions with P. mixtum and S. venustum. I also have data from truck trapping, resting site and bite count collec¬ tions, as well as observations of black flies flying from resting sites in trees which all lend support to Mochadisky's hypothesis (Simmons unpubl. data). Bird feed¬ ing species such as S. eurvadminiculum (Fallis and Bennett 1971) may be an exception to this strategy. Three ornitho- philic species have been captured throughout the night in light traps (Williams 1962, Raastad and Mehl 1972). Vision is important in other diurnal blood sucking insects such as tabanids and tse-tse flies that have more of a wait-and- seek host-location strategy (Thorsteinson et al. 1964, Brady 1972, Allan 1984). I conducted 3 experiments which demonstrate that vis- 96 ual cues are important during long-range orientation of P. mixturn and S. venustum. The first involved capture of p. mixtum and S. venustum on models without C02 (Tables 18 and 19). Although significantly more females of both spe¬ cies were captured on models with C02, these data suggest vision is important during the initial phases of orienta¬ tion. I cannot rule out the possibility that host-seeking flies that happened to fly close to the models were the ones captured. Wenk and Schlorer (1963) and Fallis et al. (1967) also captured black flies on 3-dimensional models without C02. The second experiment was placement of C02 baited models in an obscured (under hemlock tree) versus a visible (field adjacent to the hemlock) location. Signifi¬ cantly more P. mixturn and S. venustum were captured on the visible model (Table 17). In the third experiment signifi¬ cantly more females of both species were captured on a C02 baited 2-dimensional panel placed perpendicular to the ground versus horizontal to the ground (Table 15). Data from these 3 experiments suggest that long-range orienta¬ tion involves both visual and olfactory cues. In dense cover such as woods, olfaction probably is more important than vision in stimulating host-searching flight. If vision were important only during the final phases of orientation, as suggested by Bradbury and Bennett (1974b), the same number of flies should have been captured on the 97 visibly obscured models as the models in the field since they were clearly visible during the final meter of near orientation. Based on results of experiments aimed at determining the distance from a downwind C02 source from which black flies would orient to traps, Bradbury and Bennett (1974b) hypothesized that near orientation is purely visual and occurs between 0 - 1.8 m from the host. I observed no significant differences in the landing sites chosen by S. vensutum on the circumference of horizontal models baited with or without C02, suggesting that near orientation is visual (Table 19). The data from my experiments with horizontal models in obscured (woods) versus visible (field) locations support the hypothesis that near orienta¬ tion occurs during the final meter or two from the host. The models under the hemlock were not visible to the flies until they were ca. 1 meter away. Therefore, initial orientation to the models had to have been in reponse to the C02. Had the final meter of orientation also been olfactory, a concentration of flies near the C02 canister would be expected. However, there were no differences between the landing sites chosen by P. mixtum or £. venustum on models in the field versus the woods (Table 17). These data also support the contention that near 98 orientation is primarily visual since orientation appeared to have switched from an olfactory to a visual response once within the final meter- Visual observation of black flies orienting to hosts or host models also suggested that orientation to the initial landing site occurs within 1 — 2 m of the host. Prosimuliurn mixturn appeared to make a more direct flight to the host than S,. venustum- Si mul i urn venustum almost always flew to the undersides of horizontal models. Once in contact with the host, selection of the feeding site may be olfactory (Chapter II). My results agree with previous findings that P. mixtum and S. venustum orient to the edges of 2-dimensional models (Peschken and Thorsteinson 1965, Fallis et al. 1967, Bradbury and Bennett 1974a, Browne and Bennett 1980, Table 14) and to regions on 3 dimensional models that correspond to feeding sites on natural hosts (Wenk and Schlorer 1963, Bennett et al. 1972, Tables 11 and 12). However, my study is the first to demonstrate these phenomena with sta¬ tistically significant results. Browne and Bennett (1980) hypothesized that biting flies orient to points of optical convergence on the outline of the object because they associate these areas with the regions of a host's body where they are most likely to obtain a blood meal. On most mammals, the undersides and ears are the areas where hair density is lowest and capillary density is highest. I 99 observed large numbers of P. mi xtum landing on the backs of the beagles (where hair is black) but they were not able to blood feed due to the coarseness and density of the hair. In contrast, S_. venustum oriented to the undersides where hair density is low and feeding was successful (Table 12). The failure of P. mixtum to orient to the undersides of the dogs may be due to their natural tendency to land on flanks. I observed greater catches on the "flank" sections of the horizontal models (Table 13). They also may have been avoiding the white hair on the undersides of the beagles. Prosimuliurn mixtum avoided the white sections on bi-colored models more than S. venustum (Tables 23, 24 and 25). The underside region apparently is the preferred feeding site of S. venustum on natural hosts. I also observed S. venustum orienting to the undersides of horses and subsequently collected engorged individuals from this region. Significantly more S. venustum were captured on the underside section of horizontal models (Tables 13, 16 and 19) and humans in the quadruped stance (Table 12). Anderson and DeFoliart (1961) reported that in Wisconsin, S. venustum fed on the chest, neck, inner margins of the legs near the body, and jaws of horses. Rempel and Arnason (1947) believed S. articum preferred to feed on older cattle because they have shorter hair than younger ones. 100 They also observed that sheared sheep were more severely bitten than unsheared sheep. Breyev (1950) reported that S. tuberosum and Simuliurn pusi11 urn Fries oriented to and fed on the undersides of reindeer. Peterson (1959) ob¬ served S. arcticum feeding on the less hairy body regions of livestock. Fredeen (1961) capitalized on the underside orientation of S. arcticum to develop a silhouette trap to capture host-seeking females. Davies (1957) reported that S. ornatum landed almost exclusively on the undersides of cattle. The few that first landed on the back, flanks or legs left and then immediately flew to the undersides. Wenk and Schlorer (1963) found that S. ornatum oriented to flat ventral surfaces and Boopthora ervthrocephalurn DeGeer oriented to protruding parts of a horse silhouette. Boopthora ervthrocephalurn feeds in the ears of horses and cattle (Wenk and Schlorer 1963). Simuliurn damnosum orient to the leg regions of humans which is their preferred feeding site (Duke and Beesley 1958). Lacey and Charlwood (1980) reported orientation to different human body regions by 3 species of black flies from Brazil. Waage (1979, 1980) hypothesized that host defensive behavior is an important selective force on biting flies. Perhaps black fly orientation to ears and undersides of hosts has been selected in response to host defensive behavior. The anti-black fly behaviors of dogs and horses 101 could result in high mortality to black -flies, particulary given their long engorgment time (Davies and Peterson 1956, Fallis 1964). The safest places for a black fly to blood feed would be the ears and undersides, since these areas are difficult for the host to defend. Probably the most difficult body region for a host to defend would be the back. However, the density of hair on the back would inhibit feeding, as I observed on the beagles. Simulium venustum make touch and go landings on the undersides of the host. This may be to search for olfactory cues from conspecifics that have already located a suitable feeding site (Chapter II). Such a strategy would minimize the risks of foraging for feeding sites on a defensive host. Bradbury and Bennett (1974b) hypothesized that close- range orientation is in response to host color, shape and movement. I did not examine the role of movement in close- range orientation of P. mixtum and S. venustum but I found color and shape to be important. The role of host movement during near orientation of black flies remains unclear. Wenk and Schlorer (1963) showed that S. ervthroceohalum and S. ornatum were attracted to the movement of artificial models. In contrast, Peschken and Thorsteinson (1965) captured significantly fewer black flies in Manitoba Fly Traps with free hanging vs. stationary targets. 102 The role of color and shape is better known. Bradbury and Bennett (1974a,b) and Browne and Bennett (1980) stated that black flies orient to areas on 2-dimensional targets where there is maximum contrast with the background. Or¬ ientation of P. mixturn and S. venustum to the long axis of the horizontal black cylinder model was to the end sections and on the vertical models it was to the top sections (Tables 13, 17 and 19) where there was maximum contrast with the background. Fallis et al. (1967) obtained simi- liar results with S. venustum in Canada. On the horizon¬ tal models, there was a bias towards the C02 end but more flies were captured on the the non-C02 end than on the middle section (Table 13). On the vertical models, orien¬ tation to the top was most likely due to this being the area of maximum contrast with the background and not be¬ cause of the location of the C02 canister. When I changed the color of the upper half of the models to green or white more P. mixtum and S. venustum landed on the edges of the dividing line between the colors than in the top section (Table 26). Landing patterns of P. mixtum and §. venustum on the circumference of horizontal and vertical models suggest they can perceive form. On vertical models and on humans in an upright stance, landing by P. mixtum and S. venustum was essentially random but on horizontal models and on 103 humans in the quadruped stance orientation, was non-random (Table 13). Legs and shaded undersides are regions of contrast on horizontal models which are not present on upright models. I believe orientation to vertical models was random because there were no visual cues for the fe¬ males to orient to due to lack of contrast within the sphere of the model. On the horizontal models, flies re¬ sponded to visual stimuli produced by legs and shading of the undersides. Bradbury and Bennett (1974b) and Browne and Bennett (1980) showed that P. mixtum and S. venustum can discriminate between shapes on 2-dimensional objects. Evidence that legs may be a visual cue to host—seeking P. mi xtum and S.. venustum is derived from the fact that fewer females were captured on models without legs despite the presence of C02 (Table 20). The absence of legs on the model had no effect on the landing sites chosen by P. mixtum. which landed more on the flank zone (Table 20). This result was not surprising since I observed that once E« mixtum arrived near the models, they made direct flights to the flank zones from above the plane of vision of the underside. Thus, legs may not be an important cue during the near orientation phase of P. mixtum. However, the fact that significantly fewer females were captured on the leg¬ less model suggests they are used as a cue during the long- 104 range phase. There was a significant decrease in the percentage of the total S3. venustum captured on the under— side section of the model without legs compared to on the model with legs (Table 20). When near a quadruped host S. venustum flies underneath it and then orients up to the undersides. Perhaps models without legs disrupt this or¬ ientation in some way, possibly because of the lack of visual cues or a change in the way light is reflected from the underside when legs were removed. Legs may also be a cue for £>. venustum during long-range orientation since significantly fewer females were captured on models without legs compared to models with legs. Davies (1972) hypothesized that the underside orienta¬ tion behavior of black flies was due to their preference for the lower intensity of light reflected from the under¬ belly than from other regions of the host's body. This hypothesis was based on his findings that more host—seeking S. venustum landed on cloth squares with a low intensity of reflected light compared to cloths with a high intensity of reflected light (Davies 1951, 1972). I contend that this response is to visual stimuli produced by shading, legs and contrast with the background and not a preference for lower light intensity. I captured significantly more S. venustum on black compared to white models but there was no differences between black and gray models (Table 20). The 105 underside orientation of S. venustum appeared to be dis— rupted on the higher intensity models since a lower percen¬ tage of the total was captured on the underside sections of the 2 gray and white models that on black models. The white model may have been so high in intensity that S. venustum could not determine body regions due to lack of contrast within the sphere of the model's body. Thus, there was a more random orientation (Table 20). There was essentially a linear correlation between the number of P. mixtum captured and the intensity of reflected light, the greatest number being captured on the black model and the least on the white model (Table 21). These data show that P. mixtum can detect different intensities of reflected light and is attracted to lower intensities. However, there was no change in the landing sites selected by P. mixtum on the circumference sections of the neutral colored models. This may be due to the fact that P. mixtum flies directly to the flank zones which would have still been an area of high contrast with the background even with high intensity pigments. In the 2-dimensional model experiment, I found that P. mixtum and g. venustum oriented to the top edges of the model when it was perpendicular (Table 13) but when the panel was horizontal to the ground, they oriented to the underside (Table 16). This probably is due to the 106 ■fact that when the panel is perpendicalar, there is no shading so flies orient to the edge which is the area of maximam contrast- When the panel was flat, it was much darker on the underside so this was the area of maximum contrast, and hence, mixturn and S. venustum oriented to the underside. Significantly more P. mixtum and S. venustum were cap¬ tured on solid black compared to other colors tested in the bi-colored experiment (Tables 23 and 24). The fewest of either species was captured on the models with UV reflec¬ ting magnesium oxide pigment; the next lowest number was on the white model. These results are si miliar to Davies (1972) findings with S. venustum. More P. mixtum were captured on the black half whether it was on the top or bottom side of the green and black, and black and white models (Table 23). This suggests that flies are most at¬ tracted to dark areas. There was no change in orientation of P. mixtum to the leaf green mimic model. This suggests there was still sufficient contrast within the model's sphere for P. mixtum to detect shape. Similiar results were obtained on the colored upright models (Table 26). Orientation of S. venustum to upright models was less affected by white on the top half than when the top half was green (Table 26). The underside orientation of S. 107 venustum was less affected by color than for mixtum. although models with a lighter underside captured signifi¬ cantly fewer females than ones with dark undersides (Table 24). The only colors causing a shift in the landing sites of S. venustum were magnesium oxide or green when they were on the bottom of the model (Table 24). Animals with white undersides may gain some protection from hordes of black flies; many wild quadrupeds have white undersides. Anderson and DeFoliart (1961) observed fewer S. venustum feeding on white horses than dark brown horses. Host seeking black flies appear to be most attracted to colors in the blue range (Davies 1951, 1972, Bennett et al. 1972, Bradbury and Bennett 1974b, Browne and Bennett 1980, Walsh 1980). My experiment with blue and black models with and without C02 support these findings. Six times as many S. venustum were captured on the black model with C02 compared to the black model without it. In con¬ trast, only 4 times as many S. venustum females were captured on the blue model with C02 vs. without it. This suggests that blue is more attractive than black. Electro- retinogram studies on the spectral sensitivity of S. vittatum showed they were sensitive to wavelengths between 300 — 560 nm with peak sensitivity at 470 nm (blue) (G. Bernard and S.R. Bennett pers. comm.). Owens and Allan 108 (pers. comm.) -found that cattle reflect enough blue light to give them a hazy blue outline to insects with sensitivi¬ ty in the UV and blue range. Perhaps attraction to blue is related to the spectral reflectance of natural animal hair. My data agree with Davies (1972) finding that S. venustum avoids UV light when host seeking. This may be related to the fact that during twilight, when S. venustum oviposits (Chapter VII), water reflects UV light (Silbergleid 1979). Thus, UV light may serve as an attrac- tant to gravid females. Gravid S. damnosum are attracted to aluminum sheets which reflect UV light (Bellec 1976). Conclusions 1. Black flies were most successful feeding on the under— sides of dogs where hair density is lowest. Simulium venustum successfully fed on dogs whereas P. mixtum did not. This difference in feeding success probably was re¬ lated to the orientation and crawling behavior of S. venustum and the defensive behavior of the dogs. 2. Simuliurn venustum and £. mixtum oriented to the sto¬ mach when humans were in a quadruped stance. When humans were in a normal upright stance black fly orientation to the upper torso was more random. 3. Simuliurn venustum oriented to the undersides of the 109 circumferential sections of black horizontal models. Prosimulium mixturn oriented to the flanks of these models. Both species oriented randomly to the top sections of the upright cylinder models. Data from my experiments with 2— dimensional models, models with or without legs, different colored models, models of different intensity and models baited with or without C02 suggest that orientation to the underside of quadruped hosts is in response to the area of high contrast within the body area of the model. 4. Vision is important during the initial (long-range) and final (close-range) phases of orientation. Carbon dioxide was a more important long-range attractant than a close range orientation stimulus. Close-range orientation was found to be mainly visual though I cannot rule out the po¬ tential use of specific host odors or stimuli from engorg¬ ing conspecifics ("invitation effect"). 5. Model color was important in attracting both S. venustum and P. mixtum. Both species were most attracted to dark colored models of low intensity, though this was more pronounced for P. mixtum. Simuliurn venustum were more attracted to non-C02 baited blue models than non-C02 baited black models. This may be related to how they perceive light reflected from natural hosts. / 6. Both species were repelled by UV reflecting pigments. 110 This may be related to the fact that gravid females orient to water to oviposit at a time of day when it reflects UV light. CHAPTER V CHRONOLOGICAL AGE AT BLOOD FEEDING AND FOLLICLE GROWTH IN FEMALE SIMULIUM VENUSTUM Introduction Information concerning the length of a gonotrophic cycle is important for determining survival and vector potential of field populations of biting flies (Mullens and Rutz 1984, Birley et al. 1984, Tyndale-Biscoe 1984). Simuliurn venustum s.l. is a vector of Dirofilaria ursi Vamaguti (Addison 1980) in Canada, a suspected vector of Dirofilaria sp. in Massachusetts (Simmons unpubl. data), and a pest of man and domestic animals (Fallis 1964). Simuliurn venustum s.l. reportedly seek blood when their ovarian follicles are in stage I, I—II or II (Davies 1963, Magnarelli and Cupp 1977). The time post-emergence required for follicles to develop to these stages and the stage when host seeking commences has not been reported for this species. Similarly, the time from blood-feeding to fol¬ licle maturation and from oviposition to the re-initiation of feeding has not been reported for S. venustum. Ill 112 I studied the blood -feeding behavior and population dynamics of the CC cytotype of the S. venustum complex in Franklin Co., MA. In this chapter I report results of experiments to determine the following: 1) age post-emer— gence when blood feeding occurs in the laboratory, 2) follicular growth before and after blood feeding, 3) folli¬ cular stage of females that bite humans in nature and 4) the time after oviposition when blood-feeding is re-ini¬ tiated. Materials and Methods Source and procedures for holding flies for laboratory studies. Simuliurn venustum pupae were collected from the Saw Mill River below the outlet of Lake Wyola (Franklin Co., MA) during mid-May 1981-1983. CC is the only cytotype of the S_. venustum complex that occurs at this site (Appen¬ dix I). Pupae were placed on wet cotton in petri dishes housed in emergence cages. Newly emerged adults were re¬ moved from cages at 12-hr intervals. Cohorts of females of known ages (+ 12 hrs) were held in separate cages (Simmons and Edman 1978) at 21 - 22 °C, 16:8 hr light:dark cycle and 85-90% RH. Females were provided 10% sucrose ad lib. (Simmons and Edman 1981) except for 12 hr prior to feeding assays. 113 Assays to quantify blood feeding. Females were assayed for blood feeding 12, 24, 48, 64, 96, 120, 144 and 168 hr post emergence. Assays were conducted by placing 20 cohort females in a 28 ml clear plastic cup (Simmons and Edman 1981) held against a human bicep for two 10-min periods, 30 min apart. Assays were done in the morning before 1000 hr or in the evening between 1600 and 1800 hr. These are the periods of maximum feeding of S. venustum in nature (Chapter IV). Field studies. The follicle stage of biting nulliparous females in nature was determined by collecting flies as they began to bite the collector and holding them on ice until returning to the laboratory. Collections were made on several days during May 1981 - 1983 near Lake Wyola. To determine when host seeking and blood feeding began after oviposition, I sampled both biting and flying females hour— ly between 0600 and 2100 hr on 21 May 1981. Biting counts were 5 min. Flying samples consisted of 3 successive overhead net sweeps. Females were categorized as recently oviposited (=fully stretched follicular intima) or as having oviposited more than 12 hr previously (=partial1y or fully dilated follicular intima). I had previously deter¬ mined these 2 categories by measuring the rate of shrinkage of the intima of Simuliurn decorum at 22 °C following ovipo¬ sition in the laboratory (Simmons unpubl. data). Simulium 114 venustum usually will not oviposit in the laboratory. Follicle stage, measurement and parity determination. Parity was determined using the follicular relic method (Detinova 1962). Terminal follicle stage was determined following Detinova's (1962) description of Mer's (1936) modification of Christopher*s (1911) follicle classifica¬ tion scheme. Follicle length was determined by removing an ovary from a female, placing it on a slide in buffered physiological saline, gently placing a coverslip over it, and measuring the length of 10 terminal follicles at 100X. Statistical testing followed procedures of Sokal and Rohlf (1969) and Steele and Torrie (1960). Results Follicle growth. Table 27 is a summary of follicle size and stage of non-blood fed S. venustum between the ages of 0-12 and 168 hr. Females 0 - 12 hr old had stage N fol¬ licles. Stage I follicles were present in 24 and 48 hr females. Follicles of 72 and 96 hr females were stage I- II. Forty-nine and 51% of the 120 hr females had stage I- II and stage II follicles, respectively. Females that were 144 and 168 hr old all had stage II follicles. Although unquantified, there appeared to be more yolk in follicles of 168 hr females than those 144 hr old. The only signifi- 115 4- * a 13 XI U TJ TJ TJ C 1 1 4J 0 qj a rH rB 05 TJ H O o 10 in N 00 QJ E 0 m □ TJ 05 N 0* pH 00 O 10 2 0 4* ■p O' in 0 QJ • ■ a • a a • 4J U c c 05 pH 4- O O pH 10 O 10 in QI QJ •H X3 *h 10 K) a L u o> QJ »—1 QJ Qi c N fH 4- 0) ft m a QJ 4- 4- •P Qj -H o> -H •H U -P < TJ TJ X) O o o O o o O o C rH •P 0 C O CN CN CN nj O' O' 10 QJ 3 in > 0 in <* 10 CN pH L E rH rH QJ TJ •P xj 4* in ns C 4- ' si IB •P •H rH U m * Q 3 * •H QJ * QJ NJ 4- QJ pH QJ b-i in •H O' u O' ►H ►H ►—« 1 c ns ••H 05 »-H •—t • C cr pH •P z KH ►H 1 1 0» t-H ►h »-h 1 •H ns 0 TJ -P pH rH 01 C rH L 0 E QJ 0 01 c E TJ L TJ 4- •P 0 3 05 XJ U U TJ TJ TJ TJ 0 3 -P L W 4- -P ►H QJ r • in in in N 10 CN O' in cn ►H rH u Q £ in CO 00 in 10 H H pH c 1 ■ -P ro >. ►H QJ TJ cn O’ o d o o pH CN pH H L XJ E C C 4J QI ns 05 +i QJ +1 +i +1 +l +1 +1 +1 +l Qi CJi in pH C U IB SI IX pH >0 N n pH N 00 O' C •P OJ -P ■ o o O in CN N NJ 00 •h nj in a 2 pH in u ■p 0 pH d 00 o o pH pH pH pH U h TJ L 0 pH pH pH pH pH L 4- ns > CJi 4- ns h a a c ID C O' -P TJ H 0 -H C QJ u in QJ 2 •«H U 0 pH C CO CN N CN pH NJ o o TJ L L rH pH CN pH pH pH pH pH pH pH QJ 0 OJ rH 0 E 4- a 0 Ll L 4- 0 TJ a • #•4 4- QJ •P L • •P E tt L -P in 3 0 3 L qj in QI 0 c • 0 O ffl in >• • in • C QJ QJ l CN N O' CN NJ cant differences in follicle lengths was between 12 and 24 hr females and 24 and 48 hr females (Table 27). Blood feeding. No females between 0 and 12 hr old blood fed. Percentage blood feeding by 24, 48 and 72 hr females was 0.7, 1.9 3.1% and 11, respectively (Table 27). The percentage of 120, 144 and 160 hr females that blood-fed was 33, 34 and 45, respectively. The increases in feed¬ ing between 72 and 96, and 96 and 120 hr females were significant (Table 27). Mean follicle length among biting nulliparous females collected in nature was 11.4 urn. This was not significant¬ ly different from the follicle length of 120 hr old females that bit humans in the laboratory (Table 28). Four percent of the females collected biting humans in nature had stage I follicles whereas between 0.7 and 1.9% of those that had stage I follicles (24 & 48 hr) bit humans in the laboratory. Fifty-eight percent and 38% of the biting females in nature had stage I—II and stage II follicles, respectively (Table 28). A group of 120 hr old females that bit humans in the laboratory were 45% stage I—II and 55% stage II (Table 28). Biting or flying parous females with stretched intimas (females which had recently oviposited) were not captured before 1400 hrs (Table 29). There was no significant dif¬ ference between the stretched intima to dilated intimata 117 Table 28. Follicle size and stage of S. venustum feeding on humans in the laboratory and nulliparous females insert- ing their mouthparts into a human in the field. follicle stage age and source of X + S.D. (7. of total) females assayed n foil, length (urn)* I I-II II 120 hr females - 38 11.76 ± 1.17 0 45 55 in laboratory wild,nul1iparous 41 11.40 ± 1.23* 4 58 38 females - in field ♦Differences between mean follicle lengths were not signi- ficantly different as determined by Student' s t-test. Table 29. Condition of follicular intima of parous S. venustum captured while biting or flying near a human between 0600 - 2000 £ L •a •P •H u o c c E H •P 4- H XI •H •H 4- D> E Ql c Ql ns ID C o» E n ID Hi Hi X3 0 •P •P 0 00t'0OO 'ONOOO'OHNW^in'ONOOOO ooooooooooooooo ooooooooooooooo i Io I I0I0I0^-«00*H| o o K) CD>0O o K) ^O a • 00 o 00 o I 118 119 ratio of biting vs. flying females (X = 1.92). An over— head net sweep sample of S. venustum taken between 1700 and 1800 hr on 14 May 1982 near Lake Wyola outlet contained 22 parous females (n= 44) all of which had non-dilated, stretched intimas and stage I primary follicles. Two fe¬ males still had live sperm packed near the oviducts, apparently the unused residuum from oviposition. Egg development. Forty-two percent of S. venustum females held at 22 °C developed mature oocytes within 96 hr after blood feeding. Stage V oocytes were present in all fe¬ males 120 hr post-feeding (Table 30). I found no evi¬ dence of autogeny in Jj». venustum. No females from 4 dif¬ ferent populations held in the laboratory for 5-15 days at 22 °C and provided a 10% sucrose solution developed follicles beyond stage II (Table 31). Discussion My laboratory feeding experiments showed that blood¬ feeding by S. venustum increased with age and appeared to be associated with follicle growth. Few fe¬ males fed when their follicles were stage I or early stage I-II (Table 27). Maximum feeding was by females 168 hr old, which was the oldest age category assayed. A total of 967. of field collected, blood-seeking, nulliparous females 120 Table 30. Ovarian development in wild S. venustum held at various intervals at 22 dC. ■follicle stage (% o-f total dissected) hrs post ood—f eedi ng n II III IV V 12 12 100 0 0 0 24 15 0 100 0 0 48 15 0 33 67 0 72 14 0 0 100 0 • 96 19 0 0 58 42 120 20 0 0 0 100 Table 31- Ovarian development of non-blood-fed S. venustum held in the laboratory 5+ days on 107. sucrose. rH »H H- rH in u TJ r s- >< pH C rH TJ •H pH 4-1 + 4J L 0 Ql c ns 0 Ql TJ E ns "o in as > in ai in c ai 0 pH 2 0) U Ifl 0 QJ had -follicles in stage I — 11 or II (Table 28). This sup¬ ports my laboratory data which showed that -feeding rarely occured when follicles are in stage I. This suggests that the minimum age of host—seeking females in nature is 24 hr and most females probably are greater than 3 days old. This would most likely vary with field temperatures. Sutcliff and Mclver (1979) and Smith and Friend (1982) have suggested that host orientation may be necessary be¬ fore biting behavior is released in S. venustum s.1. It is logical then to question whether bypassing the host orient¬ ation phase of the blood feeding sequence (Bradbury and Bennett 1974b) influences laboratory feeding. I found that 10 - 207. of the S. venustum that landed on a human in nature began to blood feed (Chapter III). This is similiar to Davies (1952) results for S. venusutm s.l. in Canada. The closely related species. Simuliurn verecundum. rarely were attracted to humans at my field sites (Appendix II) but 617. blood fed when assayed in the laboratory (Simmons unpub. data). Mating also may be important in stimulating host¬ seeking and/or biting behavior. This has been shown for several mosquito species (Klowden 1983). I did not examine spermathecae for sperm, but since S. venustum CC rarely mate in the laboratory, it is doubtful that any of the females which I assayed were mated. In contrast, in nature 123 nearly all host seeking females are inseminated (Simmons unpub. data). Magnarelli and Cupp (1977) found no differ¬ ence in the follicular development of inseminated vs. unin¬ seminated host seeking S. venustum s.l.. My data on the ovarian stage of host seeking S. venustum are not in complete agreement with the findings of Davies (1963) or Magnarelli and Cupp (1977). Davies (1963) reported that follicles do not develop beyond stage I unless a blood meal is obtained. Magnarelli and Cupp (1977) found that in 3 different New York populations of S. venustum s.l., 287. of the biting females had stage I, 587. had stage I — 11 and 147. had stage II follicles. This is a higher percentage of females biting with stage I and stage I-II follicles than I observed. There are several possible explanations for this difference. First, it may have been due to local population variation. I have observed that the host-seeking population of S.. venustum at one site may be comprised primarily of females with stage II follicles while another population 3 or 4 km away (on the same sample day) may be comprised of females with primarily stage I-II follicles (Simmons unpubl. data). Perhaps this variation is related to larval developmental rates and nutritional differences among local populations, as suggested by Davies (1963). A second possible explanation is that Magnarelli 124 and Cupp worked with a different cytospecies which may blood feed at an earlier age than the CC population I studied. A third possibility lies in the interpretation of follicular stages which is a somewhat subjective categori¬ zation. I found that the most difficult stage to identify is late stage I and early stage I — 11. Davies (1957) observed that follicle length of Simuliurn ornatum Meigan, an anautogenous species, increased from 65 urn at emergence and reached a maximum length of 91 urn on day 5, after which there was a slight decrease in size. McMahon (1968) reported that peak feeding of S. ornatum occurred in females 4-5 days old. My results on feeding and follicle growth of S. venustum exhibited trends simi liar to both studies although I observed the highest feeding rate among the oldest females tested (Table 27). Mokry (1980b) observed peak feeding of the autogenous species, Simuliurn vittatum Zett., 24 hr post- emergence but noted a decline beyond this age. He sug- gested this was due to the fact that if a blood meal i s obtained before the female was 24 hr ol d the number of oocytes matured was enhanced but after this age blood did not increase fecundity. Studies with anautogenous Simuliurn damnsoum s.l. Theobald (Disney 1970) and Simuliurn squamosum (Vajime and Dunbar) (Simmons in ms.) have shown that peak feeding occurs soon after emergence, generally 125 within 24 hr. This may be an adaptation to the tropical environment of these species. My data clearly show that S. venustum is anauto- genous since no females developed follicles beyond stage II unless blood was ingested (Table 31). These results agree with the previous findings of Davies (1963) and Magnarelli and Cupp (1977) for S.. venustum 5.1.. Maturation of the egg batch by S. venustum after blood feeding took 4-5 days at 22 °C. Yang and Davies (1968) reported S. venustum s.l. completed blood digestion 120 - 160 hrs after ingestion. They did not report on follicular development. Blood-feeding behavior of S. venustum is re-initiated immediately following oviposition, even though the terminal follicle is still in stage I. This was demonstrated by the fact that more parous females with stretched than with dilated intimas were captured in the afternoon and evening than in the morning (Table 29). Oviposition by S. venustum occurs between 1400 and 1900 hrs on clear days (Chapter VII). Davies (1963) noted an increase in the parity rate of S. venustum s.l. during the evening. Si miliar observa¬ tions have been made for other black fly species (Lewis 1956, Davies 1955, 1957, Duke 1968, Disney 1972, Lacey and Charlwood 1980). I also have found that parous S.. venustum 126 are more likely to bite a human than nulliparous females (Chapter III). Results of this study suggest that the first gono- trophic cycle for S. venustum at 22 °C requires 8 - 10 days; 4 - 5 days for onset of feeding behavior and 4 - 5 days for egg maturation. Survival studies with this spe¬ cies in the laboratory have shown that 357. of non-blood-fed o females are alive 9 days after emergence when held at 22 C and 57 when held at 25 °C (Chapter VI). At these develop¬ ment rates, a bi-parous female would be between 12 and 14 days old. Survival up to 12 days was 4 percent and at 14 days <17. (Chapter VI). These data support Magnarelli and Cupp's (1977) obsevation that only about 17. of the S. venustum collected in New York are bi-parous. Conclusions 1. Simuliurn venustum females emerge with stage N follicles. At 22 °C follicles remain in stage I until 48 hr old and in stage I - II until 120 hr old. Stage II follicles are first present in 120 hr old females. 2. Blood feeding first occurs among females 24 hr old (<17) but does not reach a high level until they are 120 hr old (337.). Maximum feeding (457.) was observed among the old¬ est group assayed (168 hr). 3. Females that blood fed on humans in nature have the same 127 -Follicle stage and length as 120 hr -females blood -feeding on humans in the laboratory. This suggests blood feeding does not occur in most wild females until they are 5 days old. 4. Maturation of oocytes after blood feeding requires a minimum of 4 days at 22 °C. 5. Females began to host seek immediately after oviposi— tion. 6. Average minimum length of the first gonotrophic cycle is estimated to be 8 days and 4 days for subsequent cycles. 7. Simuliurn venustum females are anautogenous. CHAPTER VI VARIATION IN FITNESS OF TWO POPULATIONS OF SIMULIUM VENUSTUM FROM THE SAW MILL RIVER Introduction Body size, particularly in relation to fecundity, is an important component of fitness in a wide range of insect species (Labeyrie 1978). Little is known about various fitness components of natural black fly populations, al¬ though a correlation between size, generally winglength, and fecundity has been demonstrated for several species (Chutter 1970, Cheke et al. 1982, Colbo 1982). Colbo (1982) observed significant differences between wing lengths of female Simulium venustum s.l. from different populations both within years and between years. Labora¬ tory studies have shown that adult size is affected by larval rearing density, food quantity and temperature (Colbo and Porter 1979, 1981, Mokry 1980b, Simmons and Edman 1981). Mokry (1980b) made the interesting observa¬ tion that among laboratory reared S. vittatum. smaller females bit humans less readily than larger ones, despite the fact that this species is autogenous. 128 129 Lake outlets are particularly important in the popula¬ tion dynamics of black flies since the largest populations of many species are concentrated there (Carlsson et al. 1977, Back and Harper 1979, Colbo and Wotton 1981). I found this to be the case at the outlet of Lake Wyola on the Saw Mill River (Simmons unpubl. data). I also observed that mature larvae of 4 species were significantly larger when collected from the outlet of Lake Wyola than 1 km downstream. Since S. venustum is a major nuisance to humans in western Massachusetts during May and early June, I questioned whether the size of §_. venustum larvae af¬ fected adult fitness and human biting rates. The objec¬ tives of this study were to compare the size (subcostal length) of adult S3. venustum produced from the Saw Mill River - Lake Wyola outlet and 1 km downstream and to deter— mine its relationship to fecundity, longevity, host seek¬ ing, and blood feeding. The CC cytospecies was the only member of the S. venustum complex identified in a survey of 21 breeding sites in Franklin Co. MA (Appendix I). Materials and Methods Study sites and source of flies for laboratory studies. Populations of S. venustum from 2 sites in the Saw Mill River (Appendix I) were studied. Site 1 (SI) extended from 130 just below Lake Wyola outlet to about 100 m downstream. Site 2 300 m upstream of a point in the river where black fly population densities decrease sharply and anchor/surface ice forms in winter allowing year round study. Collections of wild biting/hovering S. venusutm were made near the edge of a small clearing in a forest ca. 400 m from SI and 0.5 km from S2. Simulium venustum pupae were collected from natural substrates, placed on wet cotton in petri dishes that were housed with 107. sucrose in emergence cages, and held at 22 0 C and 90% RH (Simmons and Edman 1981). Larval and adult size measurements. Total body length of mature, last instar larvae that had been stored in 95% ET0H was determined with an ocular micrometer in a binocular microscope at 12.5X. Larvae were positioned on their side and the distance measured between the anterior tip of the head capsule and the anal hooks (Ross and Merritt 1978). Subcostal length commonly is used as an indicator of adult black fly size (Edman and Simmons 1985). Subcostal lengths were measured (Simmons and Edman 1981) with the same ocular micrometer at 25X. Sub-samples of non-blood- fed, 5-day-old females were dissected to determine the number of stage I-II and stage II oocytes (Detinova 1962). Laboratory blood feeding assays. Females were assayed by 131 placing 20 individuals of the same age in a 28 ml clear plastic cup (Simmons and Edman 1981) held on a human bicep for two 10-min periods, 30 min apart- Sugar was removed from the cage 12 hr before feeding assays. Assays were done in the morning before 1000 hr or in the late afternoon between 1600 - 1800 hr- These are the periods of maximum feeding of S. venustum in nature (Chapter III). Females were assayed for feeding each day during the age of maximum feeding (4-7 days old, Chapter V)- All test females were dissected for the presence of mermithid parasites since they were observed to inhibit blood feeding (Chapter VII). Field studies to compare the size of biting and non-biting females. Biting females were captured by a shirtless col¬ lector. At each specific site, he remained motionless for 1 min and then collected females for a 5 min period as they inserted their mouthparts into the skin. At the end of the 5-min test period, several overhead net sweeps were made to collect non-feeding females hovering nearby. Collections were done between 1600 and 1800 hr. One test/hr was done at the site. Subcostal lengths and parity (Detinova 1962) of captured females were determined. In 1983, females attracted to a collector located in a forest clearing between SI and S2 were captured by sweep net. The number 132 of stage I — II and stage II oocytes (Detinova 1962) was counted, and subcosta length was measured for comparison with females from SI and S2. Larval rearing experiment. The rearing system used was modified from a Dayno Aqua-Lab model 103 (Fig. 5). Mater was drawn from the bottom of the main reservoir (74 cm long X 44 cm wide X 35 cm deep, 115 liter capacity) and pumped using a 47.5 1/min pump (model AC-3C-MD, March Mfg., Inc., Glenview, IL) into an external plexiglas reservoir (20 cm long X 45 cm wide X 60 cm deep, 54 liter capacity). Mater flowed from the top of the external reservoir onto a plexiglas trough (90 cm long X 30 cm wide, sloped 5 , with 12 pegs to hold strands of polyester fiber for larval attachment (Simmons and Edman 1982)) and back to the main reservoir. The external tank had a removable plexiglas plate 0.25 cm below the surface of the water to force water to flow evenly onto the trough. Mater from the field site where larvae were to be collected for rearing experiments was placed in the system 1 week prior to colonization with larvae. This likely established in the system some of the natural microorganisms that detoxify ammonia wastes (Spotte 1979). The system was refilled with water from the field site the day larvae were added. A removable screen cage that fit over the system was used to capture adults after emergence. 133 Figure 5. Diagram of modified Dayno Aqua—Lab. er, external reservoir; pi, plexiglas plate just below the water surface in the external reservoir; tr, trough for holding larvae; mr, main reservoir; gr + cc; gravel and cooling coils on bottom of main reservoir; st; stand on which the main reservoir sits and it houses the compressor (not shown) for cooling the water; et, exit tube leading from bottom of main reservoir to pump; pu, pump; rt, return tube leading from pump into external reservoir 134 135 Six hundred, 3rd instar S.venustum larvae from SI and 300 from S2 were reared in separate systems. Larvae from SI were fed a "low" diet consisting of 1 g of ground TETRA (Simmons and Edman 1981) suspended in 1 liter of field water once every 3 days. S2 larvae were fed a "high" diet consisting of 1 g of food daily. Water temperature in each system was 14 °C, the lowest temperature that was possible with the cooling system. Longevity of females. Daily cohorts of females were held in 483 ml unwaxed ice cream containers with 10% sucrose provided ad lib. At the end of each daylight period be- ginning when they were 24 hr old (day 1), dead females were removed and counted. In 1981, females from SI and S2 were held at 25-27 ° C and 907. RH. In 1983 survival of females from wild SI pupae and females resulting from SI larvae reared on low diet was determined at 22 °C and 90% RH. i Subcostal length of 1983 SI females from field pupae was regressed against day of death by least squares linear regression to determine if there was a correlation between size and longevity. Results Subcostal length and fecundity. There were significant differences in body length of larvae as well as subcostal 136 length of adult females and males from SI vs. S2 in 1981 and 1983. The smallest individuals originated from S2. The differences in the size of larvae and adults within sites between the 2 years also were significant (Table 32). SI individuals were smaller in 1983 than in 1981 but the reverse was true for individuals from S2 (Table 33). Females from SI had an average of 202 and 207 stage I- II follicles in 1981 and 1983, respectively. Females from S2 had 165 and 175 SI-II and SII follicles in 19B1 and 1983, respectively. Follicle number was significantly different between sites for each year. However, there was no significant difference in the number of follicles within sites between years, despite significant within site size differences in subcostal length (Table 33). The num¬ ber of oocytes matured by SI females after blood feeding (S2 females did not feed) was 183 in 1981 and 141 in 1983. This difference was significant, as was the difference beteween their subcostal lengths (Table 33). The same human host was used each year. Regression of subcostal length of non—blood-fed females (1981 and 1983, SI and S2 populations combined) against oocyte number indicated no correlation between female size and potential fecundity (Table 34). Regression of the number of oocytes matured by blood fed females against subcostal length indicates a significant correlation between number of oocytes matured Table 32. Body length of larvae and subcostal length of adult Simulium venustum from SI and S2. •H n TJ N a» >] a 0 TJ H 4- TJ 3 it E it ai it 3 QJ it It > It ai L |X u |x u IX TJ 13 -M 03 it Q X3 4-> 03 it +i a 03 a Q CP +i a U) C 3 CP V a c 3 CP I £ I r • H • H • -P • C +> 4-» H £L xj ai o ai pH -P •iH 0 a it L c >t 0 a o a 3 in CM «* NO 00 00 in o O' N NO 00 H m O +i 00 CM O' N o +i to CM It >0 CM CM 00 H in O H H Ch 00 it + 1 03 it a • • • • a in 00 X3 m K) m H H o 00 N N CO 13 H CM 10 +» H O' m XI h- O' CM +1 K) 00 CM >0 H nO O 10 in O H O' 00 +1 a a a a a a to m N and size (Table 34). Forty-four percent of SI females blood fed on humans in the laboratory compared to 0.8% of S2 females (Table 35). A significantly higher percentage of wild SI females took blood than SI females which were reared under low diet conditions in the laboratory (58% vs. 41% feeding, respectively) (Table 36). However, a significantly higher percentage of SI females fed than S2 females reared from wild pupae. Blood feeding by females reared from field collected S2 larvae under high diet conditions was greater than S2 females from wild pupae (7.0 vs. 1.0%) but this difference was not significant and sample size was small (Table 36). Subcostal length for females that fed in laboratory assays was significantly longer than for females which did not feed (Table 37). This was also true for nulliparous and parous females that bit a human collector in nature during a 5-min test when compared to the females captured hovering overhead at the end of the test period (Table 38). The subcostal length for wild nulliparous females captured while host seeking was longer than for females reared from S2-collected pupae but this difference was not significant. There also was no significant difference between subcostal lengths of females reared from SI—col— 141 Table 35. Percentage blood feeding of 96+ hr old S.venustum females on humans in the laboratory. population year n 7. blood fed SI 1981 232 36.6 1982 28 36.7 1983 137 58.4 total 397 44. 1 S2 1981 80 1.3 1982 60 0 1983 100 1.0 total 240 0.8 CD! 0 Table 36- Subcostal length, fecundity and percentage blood feeding of • L venustum females reared in the the laboratory from field collected pupae pH •H P K) ■p aj L L > (It c in as L ■ 4- T3 i-t P P TJ H pH a 13 P TJ pH •o 4- -i-t 13 > 01 u 3 C as as C as c O' in 3 u 0 in 01 c o> as 0 0 * as > in n1 as in in 1 1 XP Q (S3 co Q X +i m +1 c 0 • • ■ ■ • P TJ H pH P a ♦ P * ♦ a TJ * * 4- •H L as pH •rl > E as 3 4- 0 0 u as 01 c O' u in as in 3 0 c as C c 0 O' L >* 00 K3 o *H K> >0 GO CN 10 N in 0" IN 10 +1 >0 O 00 +1 H K) h* in •H pH TJ H as (S3 as a 3 a as as 2 ■ t 1 • • • (SI 0* CN o a H H 00 00 00 \n 0* a W li*3 00 CM 00 *H 10 N a +i ▼H o 00 pH TD a H (S3 L L as a as a OS os ■ • I • ■ ■ •H pH w as 0 2 a a *• 00 o 10 in 10 00 N pH >0 +1 O H O o •fH pH CN cn l U as a 3 a as 2 1 • • ■ in o a o CN E ■C P ID 0 ID p ♦ * * ♦ * QJ L • L L D> * * * * 4c H Qj P QJ • c O' o Is* o K) p P P P a 0) CM N m N m 3 •«H ID El •H • H • • a a a 0 X3 OJ 3 X) tn m m >0 N 1 -P P 1 H E C 1 in c ■H ns 3 +1 +1 +i +l +1 ns 0 P 3 0 p W H c C c IX ID O' H m 10 0 QJ 0 O' >0 O' O' > ID > u ■ a a a a 3 ■ • JD N N >0 O' H ID P • 3 O' GO N N GO QJ > C cnt ID QJ TJ a> -J ID 3 c L P f4 C H O' CN O 10 P • QJ CO P O' 'O <* CN ns u P ~ •H H H H « ■ri 13 TJ a m 1 QJ o c P *-« 4- • 0 U 0 o c QJ P V a »-« ns ID Q. TJ p H a c CT> o in CN in 0 ID p p ITS • C >0 O' 10 U C tn ns a QJ • a a a a 0 c CP ID ■ H K> in N CO Qj ri QJ P c L tn ns p r-4 C •rt QJ H E +1 +i +1 +1 +t > ns P P + 1 ns 3 L TJ r-l U W ■ri *H ■p CN CN O K) cfi C ns ’ ei-p o> ■P £ C iB iB £ IB Qi -n • CN N M 0 (fl 3 L 0 fO 0 0 £ U □ ns at c * • t ■ c c U QJ in >0 CN K) QJ Hi ■ Ql >0 10 L > in u a > QJ at ■ ■p «H ■H 4- 0i • 0 u *fH m ■ flS 4J 0 CN CN O 4- QJ mi £ in •H 0 Ch 0 TJ >S QJ QJ • • ■ £ £ 4- L lx H >0 • 0 U *4- 0 H o N H ^ L 0 0 U 4- CN (N H ■P 4J 4- 0 c in •H >B QJ TJ L m z C C iB •P« C 01 -P U in CT 0 C 0 £ O' in in c p |g u 3 ■ • • s->* m L Qj ih 4- Ql E £ in m in IB pH 4J 0 iB 3 3 QJ a QJ * £ pi 1 in ph in -P TJ 4- 0 O QJ -• o > Q1 H in 4- 3 4- IB C QJ 4* QJ TJ Qj -rt L -pi ^ r-H 4- 0 TJ I ►h lil C CN O' iB IB 01 TJ pH cn CJi c £ Qi i—« ib C > 5 iB QJ ■p TJ •pi £ 0) •P £ in c -P • z U) o «B •p* CN £ TJ I OJ u £ cn \ QJ -M O L £ \ a> 2 C •P QJ 3 0 CJ>TJ c 0 Hi 2 cn C C •pt pH TJ hh £ i-« ns L • 3 *—i Qj 0 13 QJ 0 -P IB • L c ^ c > ai oj 4- 0 I a 0 4- ns cn 0 0 Ql Ql rH L 2 3 TJ i 1 £ o IB -P £ IB H •P a H £ •P u iB QJ £ OJ 0 •H rH CM 3 v 0 QJ H L 2 £ a 2 0 0 z a * pH 145 lected pupae vs. field captured host-seeking females (Table 38). The number of oocytes present in wild hovering/biting females was not significantly different from the number present in SI females, but it was significantly greater than the number present in S2 females (Table 38). Larval rearing experiment. Adults reared in the labora¬ tory from field collected 3rd instar larvae and under a low diet feeding regime were significantly smaller and matured significantly fewer oocytes following blood feeding than adults reared from wild SI pupae (Table 36). Subcostal length for males from SI larvae also was signficantly shorter than for males emerging from wild SI pupae (X = 73.46 ± 5.08 vs. X = 85.35 ± 4.58, P< 0.001). There was no significant difference in the size of females originating from field collected 3rd instar S2 larvae reared under high diet conditions vs. females reared from wild S2 pupae. Too few females from either lab-reared 3rd instar or wild pupae blood fed to draw conclusions on fecundity. Longevity. Maximum longevity of SI females at 22 and 25 C was 14 and 10 days, respectively (Fig. 6). Data from 1981 show that the mortality curve for SI females was not as steep as the curve for S2 females (Fig. 6). SI females reared from 3rd instar larvae on a low diet in the labora¬ tory had a steeper mortality curve compared to females from SI pupae. Maximum survival of SI females reared from • IS MOtaq UI>1 | aqis ‘zs S^axqno ex%M a>je-| woxaq qsnf aq|S ‘|g ’satui6aj buijeaj pue suoi^exndod snoueA uioj^ saxeuiaj. uinqsnuaA •§ ±o x*AiAjna Axi«P aBequaDjad ’9 a-inBij 91rl 147 148 larvae on a low diet was 10 days compared to 14 days for females from wild SI pupae (Fig. 6). Meadian days survived was lowest for smallest (subcostal length = 76 urn) S. venustum females which resulted from SI pupae and highest for the largest females (subcostal length = 100 urn) (3 vs. 9 days. Table 39). There were no significant differences among the mean days survived of the 7 size classes of females (Table 39). Least squares linear regression of subcostal length against day of death of females from SI pupae indicated no significant correlation between size and survival, although there is a positive slope to the line generated ( Y = 2.04 + 5.53X, r2 = 0.01, P< 0.14). Discussion I found that small size affected fitness of S. venustum 3 ways: 1) decreased fecundity, 2) decreased blood feeding, and 3) decreased longevity. Several pre¬ vious studies have demonstrated that size of individual black flies varies considerably in nature (Chutter 1970, Neveau 1973, Cheke and Harris 1980, Colbo 1982), suggesting that this phenomenon is not unique to §,. venustum. Ad¬ ditional ways size may affect fitness include flight capa¬ bility, mating success, egg size and subsequent 1st instar fitness, and susceptibility to parasites. 149 Table 39. Mean longevity at 22 °C of different size classes of S. venustum reared from pupae collected from SI and S2. subcostal X ± S.D. median days range length (urn) n days survived* survived (days) 76 4 4.00 + 2.00 3 3-7 80 23 5.61 + 2.93 5 1 - 13 84 56 7.47 + 3.34 7 1-14 88 102 7. 12 + 4. 15 7 1-14 92 87 6.62 + 3.63 7 1 - 14 96 47 7.53 ± 3.69 8 1 - 13 100 25 7.61 4- 3.75 9 1 - 13 *No significant differences among the means (ANOVA) 150 Linear relationships of subcostal length with fecun¬ dity have been reported for autogenous (Chutter 1970, Colbo and Porter 1979, Simmons and Edman 1981, Colbo 1982) and blood-fed, anautogenous species (Cheke £l_. 1982). There was no linear relationship between subcostal length and number of resting stage oocytes in anautogenous S. venustum, despite the fact that smaller flies had fewer oocytes than larger flies (Tables 33, 34, and 35). However, there was a significant correlation between subcostal length and number of mature oocytes among blood-fed fe¬ males (Tables 34 and 35). This suggests that large females took more blood than small females, which resulted in more oocytes maturing. This would explain why SI females from 1983 that were significantly smaller but had the same number of resting stage oocytes as 1981 SI females, matured significantly fewer oocytes after blood feeding (Table 33). Klowden (1979) and Simmons and Edman (1982) showed that blood meal size, precisely controlled by enema "feeding", was positively correlated with the number of oocytes ma¬ tured. Males from S2 were smaller than those from SI (Table 32). It is difficult to speculate on the importance of this since nothing is known of the mating behavior of S. venustum. I also have observed that in colonized Simulium decorum Walker, small males produced from high density 151 rearings attempted significantly -fewer copulations and inseminated significantly fewer females than large males produced from low density rearings. Preliminary studies also suggested that small male S. decorum produce fewer sperm than large males (Simmons unpubl. data). In free cage mating experiments, S. decorum females that were inseminated were significantly larger and hence more fecund (Simmons and Edman 1981) than females which were not insem¬ inated (Simmons and Edman in prep.). Longevity studies indicated that females from S2 and females from SI larvae reared under low diet conditions and warmer temperatures had higher mortality rates than females from SI pupae (Fig. 6). Small SI females survived an average of 4 days compared to 7.6 days for large females (Table 39). This is important for reproductive success since it takes a minimum of 4 days before a female will host seek and 4 more days for egg maturation (Chapter V). At this rate only about 65% of the females from either population would be alive after 4 days but more wild SI females are alive after 8 days than females from popu¬ lations with reduced winglength (Fig. 6). Davies (1953) reported similiar survival data for S. venustum in Canada. I can only hypothesize why smaller females were less frequent human biters than larger females. One possible 152 explanation is that small -females are stressed physiologi¬ cally, possibly a hormone imbalance, as a result of larval stress. Larval nutrition has been found to affect juvenile hormone titers and subsequent pre-vitellogenic follicle development in Aedes aeavpti (Feinsod and Spielman 1980). A second possible explanation is that populations of S. venustum from SI are "super" flies due to an ideal larval habitat created by its proximity to Lake Wyola and thus are more aggressive human biters. The natural habitat of j3. venustum in Massachusetts probably is beaver ponds. I surveyed 8 beaver pond sites with CC populations and none had larvae as large as SI. Biting assays from 2 of these sites indicated a lower percentage of the females blood-fed and they were significantly less fecund than females from SI. However, neither trait was a low as for S2 females (Simmons unpubl. data). My data (Tables 37 and 38) on field biting of small vs. large females are not conclusive, but they suggest that SI is a major source of human biting §. venustum near my study sites. Other sites such as S2 or beaver ponds may produce large populations of females that annoy humans but do not bite. The percentage of S. venustum that land and actually bite humans in the vicini¬ ty of my study sites is between 10 and 20% (Chapter III). Mark-recapture studies must be done to confirm these obser¬ vations and more extensive studies on the cytospecies and 153 population genetics of 5. venustum should be conducted to confirm that there are not major genetic differences in the 2 populations. Mokry (1980b) also observed that S. vittatum which fed in the laboratory were larger than those which did not feed. He suggested it may be due to physical incapability of small females to feed. I do not know why £. venustum from SI are larger than those from S2. Vannote and Sweeney (1980) hypothesized that aquatic insects which develop under sub-optimumal conditions mature with a smaller final body size which in turn decreases their fitness. Perhaps SI is a more suit¬ able habitat than S2 for development of S. venustum. Colbo (1982) suggested that size of wild black flies is the combined effects of temperature, competition, and food quality and availability. Field studies have shown that temperature controls developmental rates of early spring species (Merritt et al. 1982) and size and fecundity of au¬ togenous species in summer (Chutter 1970). I suspect the fitness differences between SI and S2 are not temperature effects since the average daily water temperature at SI and S2 during the larval developmental period are nearly iden¬ tical (Simmons and Edman in prep). Field studies demonstrating differences in nutritional value of seston between sites or the effects of seston 154 removal on fitness have not been reported. Merritt et al. (1982) observed an increase in the organic content of seston during the time of maximum larval growth (last 2 instars) of P. mixtum/fuscum. It is not known how this influences growth. Laboratory studies with several black fly species have shown that food quantity and larval densi¬ ty does affect adult size and fecundity (Colbo and Porter 1979, Mokry 1980a, Simmons and Edman 1981, Colbo 1982) and developmental rate (Colbo and Porter 1979). My population dynamic studies in the Saw Mill River (Simmons and Edman in prep.) agree with previous studies that the highest densities, including those of S. venustum, are at lake outlets (Wallace and Merritt 1980, Colbo and Wotton 1981). Carlsson et al. (1977) speculated that quan¬ tity and quality of food at lake outlets are responsible for high densities found there. They observed that seston counts were as high at downstream sites where population densities were lower. They hypothesized that food may not be of as high a quality downstream or within the optimum size range for ingestion and assimilation. Results from my rearing experiments with S. venusutm are not conclusive since larvae were brought in from the field where water temperature was 7 °C and reared at 14 °C. Despite the elevated temperatures (field temperatures also rose to about 15 °C by the time pupation occurred), the develop- 155 mental rate of SI larvae was not increased compared to the ■field populations. This may have been due to decreased food supply, which Colbo and Porter (1979) found greatly extended development time of S_. verecundum. S2 larvae fed a high diet completed development before SI larvae and several days sooner than most S2 larvae in the field. Wild SI populations develop more synchronously than the S2 populations (Simmons and Edman in prep.). Carlsson et al. (1977) reported that large quantities of food, produced during the winter by decomposition of detritus, within the <2 urn size range that are preferred by black fly larvae, is flushed from lakes following ice out. This may be due to spring overturn of the lake (Reid 1962). Maximum larval growth of £. venustum from both SI and S2 occurs after ice out of Lake Wyola when water tem¬ peratures rise steadily (Simmons and Edman in prep.). I did not measure seston in the Saw Mill River but perhaps there is an increase in the quality and quantity of food released from Lake Wyola at this time which results in better nutrition of SI individuals due to their close proximity to the lake. S2 larvae may not receive as high a quality food. Kurtack (1979) observed that the organic content of seston ingested by black flies in New York streams was highest during May. Reduction in quantity or 156 quality of seston before it reaches S2 could occur as a result of settling, consumption by competing black fly species or other invertebrates, or some other factor. Naiman (1983) showed that seston undergoes considerable biological and physical change as it is transported down¬ stream. He suggested that the general trend for popula¬ tions of aquatic insects to decrease in density with dis¬ tance from lake outlets may be due to a decrease in seston quality rather than quantity. Reison (1974) found that black fly larvae remove significant amounts of seston from the water column. Interestingly, winter black fly spe¬ cies that inhabit SI (P. fuscum, Prosimuliurn magnum Dyar and Shannon, St. mutata and §. vittatum) complete develop¬ ment by the time S. venustum begins its maximum growth phase and summer species (§. verecundum ACD, S. decorum, and second generation S. vittatum) are just hatching. Theoretically, then, competition for food and space is reduced during the most critical developmental period for S. venustum at SI. The opposite is true at S2. There are very dense populations of mixtum and P. magnum at S2, neither of which complete development before late April or mid-May. During its maximum growth period at S2, S. venustum must co-exist with species that also are under— going maximum growth (Simmons and Edman in prep.). I obser ved that mature P. fuscum■ £. maonum and S£. mutata larvae 157 from SI also were significantly larger than individuals from S2. Whether black fly larvae compete for food resources has not been shown in natural populations (Merritt et al. 1982). This will be difficult to document. The size of particles and type of food that larvae ingest are known (Wotton 1977, Kurtak 1978, 1979, Merritt et al. 1978, 1982), but it is not known what food sources are most efficiently assimilated. Competition for space (Ross and Merritt 1978, Colbo and Harding 1980) also may be an impor tant component of black fly population dynamics which should be further examined. Conclusions 1. Larvae and adults of both sexes of S. venustum were significantly larger from the Lake Wyola outlet as compared to populations from 1 km downstream in the Saw Mill River. 2. Larger females had a significantly higher potential fecundity than smaller females. Larger females matured more eggs after blood feeding than smaller females. 3. Females from Lake Wyola outlet population fed on humans in laboratory assays but females from the population 1 km downstream did not. 4. Females that bit humans in the field were larger than 158 those that hovered overhead. Females that bit humans in laboratory assays were significantly larger than females which did not bite. 5. Females that bit humans in nature were larger and were •wore fecund than females from the downstream population. This suggests that Lake Wyola Outlet may be the major source of human biting S. venustum in the vicinity of the 1 ake. 6. Longevity of S. venustum females was greater among the lake outlet population compared to the downstream popula¬ tion and from females reared from outlet larvae under stressed laboratory conditions. 7. Preliminary laboratory rearing experiments showed that temperature and food quantity affect size and blood feeding of S. venustum. The reason for the small size of down¬ stream individuals may be related to quality of seston in the river. CHAPTER VII EFFECTS OF MERMITHID PARASITISM ON BLOOD FEEDING AND REPRODUCTION OF SIMULIUM VENUSTUM Introduction Neomesomermi s f 1 umenal i s Welch, -first described from specimens reared from larvae of S. venustum s.I. (Welch 1962), originally was believed to parasitize only immature black flies. Mokry and Finney (1977), however, reported 12.57. parasitism of adult female EL. venustum reared from pupae collected in Newfoundland. Colbo and Porter (1980), also in Newfoundland, reported 657. parasitism of adult S. venustum s.I. by mermithids presumed to be mostly N. f1umenalis. Bruder and Crans (1979) found N. f1umanalis infections in The effect of mermithid parasitism on adult black flies has received little attention. Mokry and Finney (1980) reported that mermithid parasitized S. venustum were not attracted to humans. However, since at least 4 cyto- species of §.. venustum were known to breed at their study sites these authors did not rule out the possibility that hj. f 1 umenal is parasitized a cytospecies which did not at- 159 160 tack humans. Col bo and Porter (1980) found that N. fi.^gnal is completed development in S. venustum females, which are anautogenous (Davies 1963, Chapter V), even when they had not had a blood meal. Others have reported that mermithid parasitized black flies will blood feed (Anderson and Shemanchuck 1977, Shipitsina 1963, Mondet et. al.. 1976). Col bo (1982) noted that mermithid parasitized S. venustum females were significantly smaller than non—parasitized females at one site but not at another. Neomesomermis f1umenalis is a factor in the population dynamics of S. venustum in the Saw Mill River because of its high parasitism rates (Simmons unpubl. data). The purpose of this study was to determine how N. f1umenalis affected the following fitness components of female S. venustum; (1) potential fecundity, (2) ovarian develop¬ ment, (3) size, (4) survival, (5) host-seeking and biting behavior, and (6) oviposition behavior. Materials and Methods Mermithid identification. Fully developed juvenile mermi— thids were dissected from 5+ day-old S. venustum and pre¬ served in TAF solution until identification. Mermithids were identified by Dr. D. Molloy (New York State Science Museum). Laboratory blood feeding tests. Simuliurn venustum pupae 161 were col1ected from the Saw Mill River just bel ow the outlet of Lake Wyola (Franklin Co., MA) in early to mi d- May. Pupae were placed on damp cotton in a petri di sh which was then housed in a cage. Each dai 1 y cohort of emerged -females was removed -from the cage at the end of the daylight period, transferred to 3.8 liter, unwaxed paper ice cream cartons, provided 107. sucrose solution ad l_ib (Simmons and Edman 1982) and held under a 16:8 hr light: dark cycle at 21-22 °C and 907. RH. Non-parasitized S. venustum begin to blood feed in the laboratory when 96 hr old (Chapter V). However, to deter¬ mine whether parasitized females feed at an earlier or later age all females were tested for blood feeding at 24 hr post emergence on each following day until 6 days old. Females were tested for blood feeding by placing up to 20 individuals in a 28 ml clear plastic cup (Simmons and Edman 1981) which was held on a human bicep for 10 min. After 30 min the feeding procedure was repeated again. Engorged females were held for 4 days post-feeding and then dissect¬ ed to determine if they were parasitized and the stage of egg development. Females that had not blood fed by day 6 were dissected to determine the following: (1) number of mermithids present (2) stage and length of 10 primary follicles (ovary placed on a slide in 0.857. saline with a cover slip and individual oocytes measured with an ocular 162 micrometer at 100X), (3) total number of follicles per ovary, and (4) length of the subcostal vein on the right wing (Simmons and Edman 1981). Female survival. Daily cohorts of female S. venustum that emerged from pupae collected 9-15 May 1982 were held in 483 ml unwaxed paper ice cream cartons and held under the environmental conditions described above with a maximum of 50 individuals per carton. At the end of each daylight period beginning when they were 24 hr old (day 1), dead females were removed from the carton, counted and dissected to determine if they were infected with mermithids. Field studies to determine if mermithid parasitized S. venustum host seek. Female S. venustum were sampled hourly from 0600 - 2000 hr at the Lake Wyola outlet in the following 3 ways: (1) females that landed on the skin of a shirtless male during a 5-min period were aspirated, (2) three sweeps were conducted overhead with an aerial net, and (3) three net sweeps were conducted over the ovi- position site. Sampling was done once a week during the last 2 weeks of May and the first week of June 1981. During 1982 and 1983, net sweep samples were taken at the same site on several occasions during May. All S. venustum females captured were dissected and examined for parity, ovarian stage and mermithid infection. Parity determina¬ tion and ovarian staging followed methods described in 163 Chapter V. Parasitized females captured in the net sweep samples taken over the oviposition site were examined for evidence of blood feeding by examination of the mid- and hind—guts at 100X for digested blood and expansion of the tracheoles attached to the gut. Statistical tests were conducted using procedures out¬ lined in Sokal and Rohlf (1969) and Steel and Torrie (1960). Results General observations. The majority of mermithids examined clearly resembled N. flumenalis but hereafter will be re¬ ferred to the as "mermithids" since the species present in all parasitized females were determined and the taxonomy of the group is unclear (Gordon 1984). The overall percentage parasitism of female S. venustum reared from pupae collect¬ ed at Lake Wyola outlet was 31% (n= 300 females dissected) in 1982 and 38.9% (n= 275) in 1983. Although unquantified, it appeared that parasitism increased near late May, the end of the S. venustum season at the breeding site. The number of mermithids per parasitized S. venustum ranged from 1 to 5. A random sample of 100 five-day-old females that emerged from pupae collected on 12 May 1983 were dissected. Fity-six percent had no mermithids, 8% had 1 12% had 2, 16% had 3, 7% had 4 and 1% had 5. 164 Blood feeding and oviposition. In laboratory tests, only 4-47- of parasitized females (n= 105 tested) fed on humans and these ingested only trace amounts of blood. One of the 5 parasitized females that took blood developed ovaries to stage III. The other 4 showed no evidence of ovarian development. In contrast, 68.67. of the non-parasitized females took full blood meals and all those that survived 5 days developed mature eggs (Table 40). Mermithid parasitized S. venustum appear to rarely host seek in nature. Only 0.27. of the nulliparous females (n= 345) captured coming to a human in 1981 were parasi¬ tized (Table 41). Zero (n= 528) and 27. (n= 50) of the females dissected in 1982 and 1983, either captured after landing or while hovering near humans, respectively, were parasitized. None of 290 host-seeking parous, females captured in 1981 were parasitized (Table 41). One (2.57) of the parasitized g. venustum captured while flying over the oviposition site had a small amount of blood in the gut (Table 41). Mermithids emerged from an undetermined number of non-blood fed females held in the laboratory for 5 days. This established that a blood meal is not necessary for the mermithids to complete development in S. venustum. None of the gravid female S. venustum captured hover— ing over the oviposition site were parasitized with mer mithds. However, 917 of the nulliparous females captured 165 Table 40. Laboratory blood feeding on humans by mermithid parasitized and non-parasitized S. venustum reared from wild pupae. parasitized females non-parasitized females no. V. blood no. 7. blood year tested fed ( )* tested fed ( )* 1982 77 1.3 (0) 18 55.6 (100) 1983 105 6.7 (0) 151 70.2 (100) total 182 4.4 (0) 169 68.6 (100)** ♦Number in parentheses is the percentage of the blood fed females that developed mature oocytes 5 days post feeding. **Chi-square test to compare blood feeding by parasitized and non-parasitized females. X2 = 155.49, P<0.001. . 166 Table 41. Effect of mermithid parasitism on ovarian de¬ velopment and blood feeding behavior of Simuliurn venustum. V. of female S. venustum with mermithids (total number dissected) collection type nul1iparous parous gravid over lake outlet* 40 (44) 0 (8) 0 (168) overhead net sweep 0 (292) 0 (199) — landed on human 2 (53) 0 (91) — ♦One female contained a small amount of undigested blood. All other parasitized, nulliparous females showed no evi¬ dence of having blood fed. 167 Mere infected with at least 1 fully developed juvenile nematode (Table 41). Many S. venustum when captured over the oviposition site had mermithids emerging from their anus and in some cases mermithids emerged from captured females soon after they Mere placed in vials. Parasitized §L- venustum Mere captured at the oviposition site from afternoon until dark. Peak oviposition by non-parasitized gravid females Mas at 1700 hr. The peak for nulliparous parasitized females Mas at 1900 hr (Fig. 7). The dif¬ ference between the observed frequency of parasitized fe¬ males captured at the oviposition site during each hourly sample Mas significantly different from the expected (based on the frequency of oviposition by non-parasitized females X2 = 62.30, P<0.05). Female subcostal length and ovarian development. There Mas no significant difference between the subcostal length of non-parasitized and parasitized females, including those Mith Mith multiple mermithids (Table 42). The mean number of oocytes (= potential fecundity) in females parasitized Mith mermithids Mas significantly lower than in non-parasi¬ tized individuals (Table 42). The mean number of oocytes per female dropped as the number of mermithids increased. The differences in oocyte number between females with no mermithids vs. 1 mermithid was significant but the differ¬ ences between females with 1 vs. 2, 2 vs. 3 and 3 vs. 4 168 Figure 7. Percentage of the total non-parasitized gravid and mermithid parasitized female S. venustum captured in hourly net-sweep samples over an oviposition site at the outlet of Lake Wyola. The difference between the observed capture of parasitized females per hour was significantly different than the expected (based on capture of gravid, non-parasitized females) (Chi-square = 62.30, P < 0.05). 169 170 mermithids were not significant (Table 42). The mean follicle length of non-parasitized 5+ day old females was 12.46 urn. This was significantly larger than mean follicle lengths in females parasitized by 1 or more mermithids (Table 42). Follicles in females with multiple infections were shorter than in females with 1 mermithid but the differences were not significant (Table 42). Survival. There were no apparent differences in the sur— vival of parasitized and non-parasitized females (Table 43). Maximum survival was 13 days for parasitized females and 14 days for non-parasitized females (Table 43). The 50/C survival time for both groups was ca. 6 days. Discussion I observed that host seeking and blood-feeding be¬ haviors of S. venustum females parasitized with mermithids were almost completely inhibited. Approximately 987. of the parasitized females failed to host seek or blood feed (Tables 40 and 41). Parasitism of S. venustum females by 1 or more mermithids essentially resulted in sterilization (Gordon 1984) since their oocytes did not develop normally, even in rare instances when the female had taken blood. The fact that only 1 of 40 (2.5%) nulliparous, mermithid parasitized females that were captured while flying over the oviposition site had any visible traces of blood in Table 42. Wing length, potential -fecundity and follicle size of S. venus females non—parasitized and mermithid parasitized. £ ‘HH L 01 C E4- 0 -Hai ID TD01 I s. . 4Je |X 4J . ® lx H H Q £ 4- E cn cr cn 4- Q E cn qj Q O' X) E +1 OH +1 01 £ os 0 - u m +1 3 u ~ 01 OS 01 w 0 * • -p a • ns > QJ • -P ■ C H * H £ * 3 0) c QJ > 0 0 u c IO m IN >0 CN GO N CN IO >0 O' CN o -H CN in O' H tH +1 111 o tt +i ns ns a a a a a a >0 CN K) n m K) O' >0 X) 10 (N H o 10 GO NO GO ■H CN O' O' IO in O +1 GO ’H CNK) +1 ns a a a a a a H n N N IO O' 10 >0 o H N IO H >0 IO *H m CN o o >0 H O' o +1 CN +i u ns a a a a a a in H GO X) K) N <3 IO >0 m O' CN m H CN •H IO n o O' GO o +i H o >0 +1 u ns a a a a a a N N 13 K) N o CN CN o in TJ H H O O o ■H <* in o GO >0 +1 Is* O' +1 ns a a a a a a o o o IO O O o o ■H GO 00 o o o o +1 H O' >0 o o in +l a a a a a a .p z co T3 xs ■P •H H 4- ai -»h C 4- •P > TJ ns -P TJ c 01 Z E TJ ns fH C 4* •P E■ 3 01 xi zo > oi •rH 4- 4- U -P ns ai C L u qj ns c a oi qj CT-P L 4- i 0 4- 1 * QJ QJ 01 XI C 01 E 4- QJ 0 IB rH C r-H 01 0 z n QJ L 01 qj 3in 0J qj QJ L QJ c u OJ •P H i-i 4- •fH -P T3 C u ns C >- L cr •p -P Qj H 0l L 2 QJ 01 3 ui ai QJ . 0_ o •p +j V 4- QJ 01 c ns L ■ 171 172 Table 43. Percent daily survival of mermithid parasitized and non-parasitized S. venustum females. Vm of total females surviving Age (days) parasitized 1 91.20 96.30 2 82.60 83.30 3 74.70 76.20 4 66.30 64.00 5 62.00 62. 10 6 54.70 58. 10 7 46.50 49. 10 8 33.70 36. 10 9 26.70 29.60 10 17.40 22.0 11 14.00 15.90 12 6.98 9.03 13 3.50 6.90 14 0.00 2.2 15 — 0.00 *N is the total number of females that were used to start the experiment. 173 their digestive tracts (Table 41) confirms that: (1) in nature, parasitized females generally do not blood feed, (2) mermithids can complete development in an anautogenous host that has not blood fed, and (3) parasitized females return to the stream and "oviposit" as if they were gra¬ vid- Mermithids were able to complete development in 4-5 days in non-blood-fed S. venustum reared in the laboratory from field collected pupae. Mokry and Finney (1977) and Col bo and Porter (1980) reported si miliar results for mer— mithid parasitized S. venustum. Anderson and Shemanchuck (1977) reported that infection of Simuliurn arcticum Mailoch with Isomerm:is sp. did not disrupt host seeking, blood feeding or mating. Mondet et aK (1976) did not determine whether parasitism of Simuliurn damnsoum s.l. Theo. by Isomermis 1airdi inhibited host seeking but they observed that parasitized females would blood feed in the labora¬ tory. Perhaps Isomermis spp. affect their host differ¬ ently than do Neomesomermis. spp.. Sterilization of female black flies when parasitized by mermithids has been reported for a number of species, including S. venustum (Peterson 1960, Shipitsina 1963, Phelps and DeFoliart 1964, Le Berre 1966, Hunter and Moorhouse 1976, Mondet et al.. 1976, Anderson and Shemanchuck 1977, Mokry and Finney 1977). Not all black fly species are sterilized by mermithids. Hunter and 174 Moorhouse (1976) found mature eggs in some parasitized Austrosimuliurn bancrofti (Tayl.). I have collected a few Steaopterna mutata Mailoch and P. fuscum females with both mature eggs and mermithids, although the numbers of eggs in parasitized individuals were reduced (Simmons unpubl. data). Both these species are autogenous for the first gonotrophic cycle. The cause of sterilization of female black flies by mermithids is not known. Gordon (1984) suggested it could be due to depletion of host fat body, disruption of the endocrine system, or both. In the case of S. venustum. the facts alone that parasitized females do not host seek or blood feed account for sterilization since S. venustum always require blood for egg maturation (Chapter V). Moreover, my results indicate that even if parasitized females ingest blood, oocytes do not mature. Almost nothing is known about the physiological con¬ trol of host seeking behavior in black flies (Cupp 1981). Therefore, it is difficult to even speculate on how this behavior might be suppressed by the mermithids in S. venustum. However, two general possibilities can be sug¬ gested. The first involves disruption of post-emergence growth of the ovaries. I found that in non—parasitized S. venustum, mean follicle length at emergence was 6 urn. Blood feeding in the laboratory or host seeking in the 175 field did not occur until the follicles were 12 um long, which requires 4 days at 21-22 cC (Chapter V). In parasi¬ tized females, the follicles decrease in size after emer— gence (Table 42). When parasitized females were 5 days old, their follicles were significantly smaller than the folli¬ cles of non-parasitized females that sought hosts in the field or blood fed in the laboratory (Table 42). It is possible that these behaviors are suppressed in parasitized females because the oocytes never reach resting stage. The reason for this may be due to depletion of fat body by the mermithids, a common observation of mermithid parasi¬ tized black flies, including S. venustum. (Finney 1981). The roles of the fat body and ovaries in host-seeking and blood feeding behavior of black flies have not been deter¬ mined. In mosquitoes, a humeral factor that triggers biting or host-seeking is released when the oocytes reach the resting stage (Klowden 1983). The fat body is the site of vitellogenin synthesis in mosquitoes (Hagedorn et al.. 1973). The fact that S. venustum females parasitized with one mermithid had larger follicles than females with more than one mermithid (Table 42) suggests the endocrine or other organ system responsible for controlling post-emer— gence follicle growth to the resting stage is affected by the parasite load. The second possible explanation involves growth of the 176 mermithid. As mentioned above, non-parasitized S. venustum ■females generally blood feed when 4-5 days old. The mermi- thids in parasitized S. venustum were ready to emerge 5 days post-emergence. Mermithid growth caused the fly's abdomen to become distended as if blood engorged or gravid. In mosquitoes, distention of the abdomen by a blood meal above a certain volume triggers stretch receptors that cause inhibition of further host seeking (Klowden 1983). When the blood meal has been digested and distention of the abdomen drops below the critical threshold, a humoral fac¬ tor associated with the developing oocytes continues inhi¬ bition of host-seeking (Klowden 1983). If a similiar sys¬ tem of distention inhibition exists in S. venustum. it is conceivable that distention of the abdomen by mermithids mocks distention caused by a blood meal. Alternatively, the mermithids may produce a factor that mimics the normal humoral inhibition system during egg maturation, as has been shown in mosquitoes (Klowden 1983). Female S. venustum. parasitized with fully developed juvenile mermithids, flew over the oviposition site during the same time period as gravid females (Fig. 7). My truck trap studies of S. venustum adults also have shown that females with mature 3rd stage juveniles have the same flight periodicity as gravid individuals (Simmons unpubl. data). Black fly mermithids. including f lumenal is. re- 177 quire water for mating and completion of their life cycle. My data suggest that the parasite promotes mock oviposi- tion behavior by the female. Mock oviposition of mermi- thids has been reported previously for S. venustum (Colbo and Porter 1980) as well as several other black fly species (Rubstov 1971, Wenk 1976, Molloy 1981). Parasitism of a black fly by more than one mermithid is common. In general the proportion of male to female mermithids in the host increases as the number of parasites increases (Gordon 1984). Enzenwa and Carter (1975) found that if there was one N. f1umenalis in an S. venustum larva 10.9% were males. When there were 4 mermithids per larva they were all males. Colbo (1982) observed a maximum of 6 mermithids per adult female £. venustum s.l.. The percen¬ tage parasitism with 1, 2, 3 and 4 mermithids that Colbo (1982) reported was similar to what I observed. The maxi¬ mum mermithid load I observed was 5. Ezenwa and Carter (1975) observed that maximum parasitization of S. venustum larvae by late stage f1umenalis was 4. I did not sex all the mermithids that I dissected from §. venustum but I did note that females were rarely involved in multiple infec¬ tions. Female mermithids are much larger than males. One female mermithid requires as much space in the haemoceol of an S. venustum female as 3 or 4 males. This suggests that during their parasitic phase of development mermithids 178 regulate their numbers in a sex dependent fashion. Aside from reproduction, mermithids did not appear to affect the fitness characteristics of S. venustum which I examined. I observed no significant differences between the subcostal length of parasitized and non-parasitized S. venustum (Table 42). Colbo (1982) observed significant differences in winglengths of female S. venustum that were parasitized in one population but not another. I also ob¬ served no difference in survival between parasitized and non-parasitized S.venustum. Mondet et al. (1976) observed a significant decrease in the survival of mermithid parasi¬ tized S. damnosum s.l.. In the laboratory, the mermithids require about 5 days to complete their parasitic phase of development (my study and Colbo and Porter 1980). Survival of both parasitized and non-parasitized S. venustum was 62% after 5 days (Table 43). Females generally die following oviposition of their mermithids (Colbo and Porter 1980). These observations suggest that the effects of the mermi¬ thids are concentrated on the host's reproductive system. Such a strategy would optimize the mermithids chance of being returned to water where it can complete its life cycle (Colbo 1982). The N. f1umenalis / S. venustum host/parasite re¬ lationship should receive furthur study from the physio¬ logical/behavioral and biological control points of view. 179 This mermithid is an ideal biological control agent in that parasitism of a female with only one mermithid results in sterilization and inhibition of biting behavior. Conclusions 1- Parasitism of S. venustum by N. f1umenalis blocked host seeking and blood feeding behavior. 2. Oocytes of parasitized females did not develop to the resting stage. Parasitized females that partially blood fed did not undergo normal ovarian development. 3. Parasitized females returned to the stream to "ovi¬ posit" mermithids at the same time of the day as non— parasitized, gravid females. 4. The number of mermithids per infected host ranged from 1 to 5. One mermithid was able to disrupt oocyte develop¬ ment and blood feeding. 5. Parasitism was not correlated with female size or survival. Literature Cited Addison, E. M. 1980. Transmission of Dirofilaria ursi Yamaguti, 1941 (Nematoda: Onchocercidae) of black bears (Ursus americanus) by black flies (Simuliidae). Can. J. Zool. 58: 1913-1922. Ahmadi, A. and G. McClelland. Invitation effect on the host-finding behavior of Aedes sierrensis. Physiol. Entomol. (in press) Alekseev, A.S., S. Rasnityn and L. Vitlin. 1977. On group behavior of females of blood-sucking mosquitoes. Communication. I. Discovery of the "effect'of invitation. Med. Parazitol. Parasitar. Biol. 46: 23. (in Russian) Allan, S.A. 1984. Studies on vision and visual attraction of the salt marsh horse f1v.Tabanus niarovittatus Macquart. PhD Dissertation, University of Massachusetts, Amherst, Massachusetts. Anderson, J.R. and G.R. DeFoliart. 1961. Feeding behavior and host preferences of some black flies (Diptera: Simuliidae) in Wisconsin. Ann. Entomol. Soc. Am. 54: 716-729. Anderson, J.R. and J.A. Shemanchuk. 1977. Parity and mermithid parasitism of Simuliurn arcticum caught attacking cattle and flying over an Alberta River. Proc. First Inter—Regional Conf. on N. American Blackfies. pp. 143—145. Mimeograph Document, University of New Hampshire. Anonymous. 1979. Proceedings of a workshop on livestock pest management to assess national research and extension needs for integrated pest management of insects, ticks and mites affecting livestock and poultry. Kansas State Univ., Manhatten. 322pp. Back, C. and P.P. Harper. 1979. Succession saisonniere emergence voltinisme et repeartition de mouches noires des Laurentides (Diptera: Simuliidae). Can. J. Zool. 57: 627-639. 180 181 Bellee, C. 1976. Captures d'adultes de Simuliurn damnosum Theobald, 1903 (Diptera: Simuliidae) a 1'aide de plaques d'aluminium, en Afrique de l'Ouest. Cah. ORSTOM ser. Ent. med. Parasit. 14s 209-217. Bennett, G.F. and A.M Fallis. 1971. Flight range, longevity and habitat preference of female Simulium eurvadminiculurn Davies (Dipteras Simuliidae). Can. J. Zool. 49s 1203-1207. Bennett, G. F., A.M. Fallis and A.G. Campbell. 1972. The response of Simulium eurvadminiculum Davies (Dipteras Simuliidae) to some olfactory and visual stimuli. Can. J. Zool. 50s 793-800. Birley, M.H., Y. Braverman and K. Frish. 1984. survival and blood—feeding rate of some Culicoides species (Dipteras Ceratopogonidae) in Israel. Environ. Entomol. 13s 424-429. Bradbury, W.C. and G.F. Bennett. 1974a. Behavior of adult Simuliidae (Diptera). I. Response to color and shape. Can. J. Zool. 52s 251-259. Bradbury, W.C. and G.F. Bennett. 1974b. Behavior of adult Simuliidae (Diptera). II. Vision and olfaction in near—orientation and landing. Can. J. Zool. 52s 1355-1364. Brady, J. 1972. The visual responsiveness of the tsetse fly, G1ossina morsitans Westw.(G1ossinidae) to moving objectss the effects of hunger, sex, host odour and stimulus characteristics. Bull. Entomol. Res. 62s 257-279. Breyev, K.A. 1950. The behavior of blood-sucking Diptera and Bot-flies when attacking reindeer and the responsive reactions of the reindeer. I. The beha¬ vior of blood-sucking Diptera and Bot-flies when attacking reindeer. Parazit. Sb. 12s 167-238. (in Russian) 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 Brunswick, Canada. J. Med. Entomol. 17s 58—62. 182 Browne, S.M and G.F. Bennett. 1981. Response of mosquitoes (Diptera: Culicidae) to visual stimuli. J. Med. Entomol. 18: 505-521. Bruder, K.W. and W.J. Crans. 1979. The blackflies (Simuliidae: Diptera) of the Stony Brook Watershed of New Jersey, with emphasis on parasitism by mermithid nematodes (Mermithidae: Nematoda). Bull. No 851, N.J. Agric. Expt. Stat. State Univ., NJ. Carlsson, M., L. Nilsson, Bj. Svensson, S. Ulfstrand and R.S. Wotton. 1977. Lacustrine seston and other factors influencing the blackflies (Diptera: Simuliidae) inhabiting lake outlets in Swedish Lapland. Oikos 29: 229-238. Cheke, R.A. and J.R.W. Harris. 1980. Seasonal size variation in females of the Simuliurn damnosum complex in the Ivory Coast. Tropenmed. Parasit. 31: 381-385. Cheke, R.A., R. Garms and M. Kerner. 1982. The fecundity of Simuliurn damnosum s.l. in Northern Togo and infections with Onchocerca spp. Ann. Trop. Med. Parasit. 76: 561-568.” Christophers, S.R. 1911. The development of the egg follicle in anophelines. Paludism 2: 73-89. Chutter, F.M. 1970. A preliminary study of factors influencing the number of oocytes present in newly emerged blackflies (Diptera: Simuliidae) in Ontario. Can. J. Zool. 48: 1389-1400. Colbo, M.H. 1982. Size and fecundity of adult Simuliidae (Diptera) as a function of stream habitat, year and parasitism. Can. J. Zool. 60: 2507-2513. Colbo, M.H. and G.N. Porter. 1979. Effects of the food supply on the life history of Simuliidae (Diptera). Can. J. Zool. 57: 301-306. Colbo, M.H. and G.N. Porter. 1980. Distribution and specificity of Mermithidae (Nematoda) infecting Simuliidae (Diptera) in Newfoundland. Can. J. Zool. 58: 1483-1490. 183 Colbo, M.H. and G.N. Porter. 1981. The interaciton of rearing temperature and food supply on the life history of two species of Simuliidae (Diptera). Can. J. Zool. 59: 158-163. Colbo, M.H. and R.S. Wotton. 1981. The bionomics of preimaginal blackflies. p. 209-226. In: Laird, M. (ed.), Blackflies: the future for biological methods in integrated control. Academic Press, London. Crosskey, R.W. 1973. Simuliidae. p. 109-153. In: Smith, K.G.V. (ed.), Insects and other arthropods of medical importance. British Museum (Natural History), London. Crosskey, R.W. 1981. Geographical distribution of Simuliidae. p. 57-70. In: Laird, M., Blackflies: the future for biological methods in integrated control. Academic Press, N.Y. Cupp, E.W. 1981. Blackfly physiology. p. 199-208. In: Laird, M., Blackflies: the future for biological methods in integrated control. Academic Press, London. Cupp, E.W. and A.E. Gordon. 1982. Notes on the systematics, distribution, and bionomics of black flies (Diptera: Simuliidae) in the northeastern United States. Search: Agriculture. Ithaca, N.Y.: Cornell Univ. Agr. Exp. Sta. No. 25. 76pp. Davies, D.M. 1951. Some observations of the number of black flies (Diptera: Simuliidae) landing on colored cloths. Can. J. Zool. 29: 65-70. Davies, D.M. 1952. The population and activity of adult female black flies in the vicinity of a stream in Algonquin Park, Ontario. Can. J. Zool. 30:287-321. Davies, D.M. 1953. Longevity of black flies in captivity. Can. J. Zool. 31: 304-312. Davies, D.M. 1972. The landing of blood seeking female black-flies (Simuliidae: Diptera) on coloured materials. Proc. Entomol. Soc. Ont. 102: 124-155. 184 Davies, D.M. and B.V. Peterson- 1956- Observations on the mating, feeding, ovarian development, and oviposition of adult black flies (Simuliidae: Diptera). Can. J. Zool. 34s 615-655. Davies, L. 1957a- A study of the black fly Simulium ornatum Mg. (Dipteras Simuliidae) with particular reference to its activity on grazing cattle. Bull. Entomol. Res. 48s 407-424. Davies, L. 1957b. A study of the age of females of Simuliurn ornatum Mg. (Diptera) attracted to cattle. Bull. Entomol. Res. 48s 535-552. Davies, L. 1963. Seasonal and diurnal changes in the age composition of adult Simuliurn venustum Say (Diptera) populations near Ottawa. Can. Entomol. 95s 654-657. Davies, L., A.E.R. Downe, B. Weitz and C.B.Wi11iams. 1962. Studies on black flies (Diptera, Simuliidae) taken in a light trap in Scotland. II. Blood-meal identification by precipitin tests. Trans. R. Entomol. Soc. Lond. 114s 21-27. Davies, L and D.M. Roberts. 1973. A net and a catch- segregating apparatus mounted in a motor vehicle for field studies on flight activity of Simuliidae and other insects. Bull. Entomol. Res. 63s 103-112. Detinova, T.S. 1962. Age-grouping methods in Diptera of medical importance. Ann. Rev. Entomol. 13s 427-450. Disney, R.H. 1970. The timing of the first blood meal in Simuliurn damnosum Theobald. Ann. Trop. Med. Parasit. 64s 123-128. Disney, R.H. 1972. Observations on chicken-biting black flies in Cameroon with a discussion of parous rates of Simuliurn damnosum. Ann. Trop. Med. Parasit. 66s 149-158. Disney, R.H.L. and P.F.L. Boreham. 1969. Blood gorged resting blackflies in Cameroon and evidence of zoophily in Simulium damnosum. Trans. R. Soc. Trop. Med. Hyg. 63s 286-287. 185 Downe, A.E.R. and P.E. Morrison- 1957. Identification of blood meals of blackflies (Diptera: Simuliidae) attacking farm animals. Mosq. News 17: 37-40. Duke, B.O.L. 1968. Studies on factors influencing the transmission of onchocerciasis. IV. The biting cycles, infective biting density and transmission potential of "forest" Simuliurn damnosum. Ann. Trop. Med. Parasit. 62: 95-106. Duke, B.O. and W.N. Beesely. 1958. The vertical distribution of Simulium damnosum bites on the human body. Ann. Trop. Med. Parasitol. 52: 274-281. Edman, J.D., L.A. Webber and H.W. Kale II. 1972. Host¬ feeding patterns of Florida mosquitoes. II. Culiseta J. Med. Entomol. 9: 429-491. Edman, J.D. and K.R. Simmons. 1985. Rearing and colonizaiton of black flies (Diptera: Simuliidae). J. Med. Entomol. 22: 1-17. Edman, J.D., J.F. Day and E.D. Walker. 1985. Vector host interplay: factors affecting disease transmission. In: Lounibos, L.P., J.R. Rey and J.H. Frank (eds.), Proc. Mosquito Ecology Workshop. Am. Entomological Institute, Ann Arbor (in press) Enzewa, A.O. and N.E. Carter. 1975. Influence of multiple infections on sex ratios of mermithid parasites of blackflies. Environ. Entomol. 4: 142-144. Fallis, A.M. 1964. Feeding and related behavior of female Simuliidae (Diptera). Expl. Parasit. 15: 439-470. Fallis, A.M. and G.F. Bennett. 1958. Transmission of Leucocvtozoon bonasae Clarke to ruffed grouse (Bonasa umbel 1 us L. ) by the black flies Simul.i.um latipes Mg. and Simulium aureum Fries. Can. J. Zool. 36: 533-539. Fallis, A.M., Bennett, G.F., G. Griggs and T. Allen. 1967. Collecting Simuliurn venustum females in fan traps and on silhouettes with the aid of carbon dioxide. Can. J. Zool. 45: 1011-1017. Fallis, A.M. and J.N. Raybould. 1975. Response of two African simuliids to silhouettes and carbon dioxide. J. Med. Entomol. 12: 349-351. 186 Fal11s, A.M. and S.M. Smith. 1964. Ether extracts from birds and carbon dioxide as attractants for some ornithophilic simuliids. Can. J. Zool. 42s 723-730. Feinsod, F.M. and A. Spielman. 1980. Nutrient—mediated juvenile hormone secretion in mosquitoes. J. Insect Physiol. 26s 113-117. Finney, J.R. 1981. Potential of mermithids of control and in vitro culture. p. 325-334. Ins Laird, M. (ed.), Blackfliess the future for biological methods integrated control. Academic Press, London. Fredeen, F.J.H. 1956. Black flies (Dipteras Simuliidae) of the agricultural areas of Manitoba, Saskatchewan and Alberta. Proc. Tenth Internat. Cong. Entomol. Montreal 3s 819-823. Fredeen, F.J.H. 1969. Outbreaks of the black fly Simulium arcticum Mailoch in Alberta. Quaest. Entomol. 5s 341-372. Fredeen, 1961. A trap for studying the attacking behavior of black flies, Simuliurn arcticum Mall. Can. Entomol. 93s 73-78. Golini, V. 1970. Obserations on some factors in the host¬ seeking behavior of simuliids (Diptera) in Ontario and Norway. M.Sc. Thesis, McMaster Univ. Hamilton, Ontario. Golini, V. and D.M. Davies. 1971. Upwind orientation of female Simuliurn venustum Say (Diptera) in Algonquin Park, Ontario. Proc. Entomol. Soc. Ont. 101s 49-54. Gordon, A.E. and E.W. Cupp. 1980. The limnological factors associated with cytotypes of the Simulium venustum/verecundum complex (Dipteras Simuliidae) in New York state. Can. J. Zool. 58s 973-981. Gordon, R. 1984. Nematode parasites of black flies, p. 821-848. Ins Nickle, W.R. (ed.), Plant and insect nematodes. Marcel Dekkar, N.Y. Hagedorn, H.H., A.M. Fallon and H. Laufer. 1973. Vitellogenin synthesis by the fat body in Aedes aeqypti; evidence for transcriptional control. Develop. Biol. 31s 285-294. 187 Harding, J. and M.H. Colbo. 1981. Competition for attachment sites between larvae of Simuliidae (Diptera). Can. Entomol. 113: 761-763. Hawkins, F. and M. Worms. 1961. Transmission of filarioid nematodes. Ann. Rev. Entomol. 6: 413-432. Hocking, G. 1964. Aspects of insect vision. Can. Entomol. 96: 320-334. Holbrook, F.R. 1967. The black flies (Diptera: Simuliidae) of western Massachusetts. Ph.D. Dissertation, Univ. Massachusetts, Amherst, MA. Hunter, D.M. and D.E. Moorehouse. 1976. Sexual mosaics and mermithid parasitism in the Austrosimuliurn bancrofti (Tayl.)(Diptera: Simuliidae). Bull. Entomol. Res. 65: 549-557. Jamback, H. 1973. Recent developments in control of blackflies. Ann. Rev. Entomol. 18. 281-304. Jamback, H. 1976. Simuliidae (Blackflies) and their control. IV. WHO/VBC/76.653. Mimeogr. Doc. Jamback, H. 1981. The origins of black fly control programmes. p. 71-74. In. Laird, M., Blackflies: the future for biological methods in integrated control. Academic Press, London. Klowden, M.J. and A.O. Lea. 1979. Oocyte maturation in the black fly, Simuliurn underhiHi Stone and Snoddy, resulting from blood enemas. Can. J. Zool. 57: 1344- 1347. Klowden, M.J. 1983. The physiological control of mosquito host-seeking behavior. p.93-113. In. Harris, K.E. (ed.), Current topics in vector research. Volumne 1, Prager, NY. Kurtak, D.C. 1978. Efficiency of filter-feeding of blackfly larvae (Diptera: Simuliidae). Can. J. Zool. 56: 1608-1623. Kurtak, D.C. 1979. Food of black fly larvae (Diptera: Simuliidae): seasonal changes in gut contents and suspended material at several sites in a single watershed. Quest. Entomol. 15: 357-374. 188 Labeyrie, V. 1978. The significance of the environment in the control of insect fecundity. Ann. Rev. Entomol. 23s 69-89. Lacey, L.A. and J.D. Charlwood. 1980. On the biting activities of some anthropophi1ic Amazonian Simuliidae (Diptera). Bull. Entomol. Res. 70s 495- 509. Le Berre, R. 1966. Contribution a 1'etude biologique et ecologique de Simul iurn damnosum Theobald, 1903 (Dipteras Simuliidae). Mem. 0RST0M 17s 1-204. Lewis, D.J. 1956. Biting times of parous and nulliparous Simuliurn damnosum. Nature, Lond. 178: 98-99. Lok, J.B. 1981. Surrogate and natural vectors of Onchocerca spp. Ph.D. Dissertation, Cornell Univ., Ithaca, N.Y. Lowther, J.K. and D.M. Wood. 1964. Specificity of a black fly, Simuliurn euryadminiculurn Davies, towards its host, the common loon. Can. Entomol. 96s 911-913. Magnarelli, L.A. and E.W. Cupp. 1977. Physiological age of Simuliurn tuberosum and Simuliurn venustum (Dipteras Simuliidae) in New York state, U.S.A. J. Med. Entomol. 13s 621-624. May, B., L.S. Bauer, R.L. Vadas and J. Granett. 1977. Biochemical genetic variation in the family Simuliidaes electrophoretic identification of the human biter in the isomorphic Simuliurn jenninasi group. Ann. Entomol. Soc. Am. 70s 637-640. McMahon, J.P. 1968. Artificial feeding of Simuliurn vectors of onchocerciasis. Bull. Wld. Hlth. Org. 37s 415- 430. Mer, G.G. 1936. Experimental study on the development of the ovary in Anopheles elutus Edw. (Dipt. Culic.). Bull. Entomol. Res. 27s 351-359. Meredith, S.E.0. and H. Townson. 1982. Enzymes for species identification in the Simuliurn damnosum complex from West Africa. Tropenmed. Parasit. 32s 123-129. 189 Merritt, R-W. and H.D. Newson. 1978. Ecology and management of arthropod populations in recreational lands. p. 125-162. Ins Frankie, G.W. and C.S. Koehler (eds.), Perspecives in urban entomology. Academic Press, NY. Merritt, R. W., M.M. Mortland, E.F. Gersabeck and D.H. Ross. 1978. X—ray diffraction of particles ingested by filter-feeding animals. Entomologia Exp. et Appl. 24s 27-34. Merritt, R.W., D.H. Ross and G.J. Larson. 1982. Influence of stream temperature and seston on the growth and production of overwintering larval black flies (Dipteras Simuliidae). Ecology 63s 1322-1331. Mokry, J.E. 1980a. Laboratory studies on blood-feeding of blackflies (Dipteras Simuliidae) I. Factors affecting the feeding rate. Tropenmed. Parasit. 31s 367-373. Mokry, J.E. 1980b. Laboratory studies on blood-feeding of blackflies (Dipteras Simuliidae). 2. Factors affecting the fecundity rate. Tropenmed. Parasit. 31s 374-380. Mokry, J.E. and J.R. Finney. 1977. Notes on mermithid parasitism of Newfoundland blackflies with the first record of Neomesomermis f1umenalis from adult hosts. Can. J. Zool. 55s 1370-1372. Molloy, D.P. 1981. Mermithid parasitism of black flies. J. Nemat. 13s 250-256. Monchadskiy, A.S. 1956. Bloodsucking Diptera on the territory of the USSR, and some features of their attacks on man. Ent. Obozr. 35s 547-559. Mondet, B., B. Pendriez and J. Bernadou. 1976. Etude de parasitisme des Simulies (Diptera) par des Mermithidae (Nematoda) en Afrique de 1'Ouest. I. Observations preliminaires sur un cours d'eau temporaire de Savane. Cah. 0RST0M Ser. Entomol. Med. Parasitol. 14s 141-149. Mullens, B.A. and D.A. Rutz. Age structure and survivorship of Culicoides variapennis (Dipteras Ceratopogonidae) in central New York state, U.S.A. J. Med. Entomol. 21s 194-203. 190 Naiman, R.J. 1983- The influence of stream size on the food quality of seston. Can. J- Zool- 61: 1995-2010. Nasci, R.S. 1981. A lightweight battery-powered aspirator for collecting resting mosquitoes in the field. Mosq. News 41: 808-811. Neveu, A. 1973. Variations biometriques saisonnieres chez les adultes de quelques especes de Simuliidae (Diptera: Nematocera) d'une ruisseau des Pyrenees- Atlantiques, le Lessuraga. Ann. Hydrobiol. 4: 51-75. Newson, H.D. 1977. Arthropod problems in recreation areas. Ann. Rev. Entomol. 22: 333-353. Peterson, B.V. 1959. Observations on mating, feeding and oviposition of some Utah species of black flies (Diptera: Simuliidae). Can. Entomol. 91:188-190. Peterson, B.V. 1960. Notes on some natural enemies of Utah black flies (Diptera: Simuliidae). Can. Entomol. 91: 147-155. Peschken, D.P. and A. J. Thorsteinson. 1965. Visual orientation of black flies (Simuliidae: Diptera) to color, shape and movement of targets. Entomol. Exp. Appl. 8: 282-288. Phelps, R.J. and G.R. DeFoliart. 1964. Nematode parasitism of Simuliidae. Res. Bull. Agric. Exp. Stn. Uni v. Wis. 245: 1-78. Raastad, J.E. and R. Mehl. 1972. Night activity of black flies (Diptera: Simul iidae) in Norway. Norsk Ent. Tidsskr. 19: 173. Reid, G.K. 1961. Ecology of inland waters and estuaries. Van Nostrand, NY. Rothfels, K.H. 1956. Black flies: siblings, sex and species groupings. J. Hered. 47: 113-122. Rothfels, K.H. 1979. Cytotaxonomy of black flies (Simuliidae). Ann. Rev. Entomol. 24: 507-539. 191 Rothfels, K.H. 1981. Cytotaxonomy: principles and their application to some northern species-complexes in Si mul i urn. p. 19-30. Ins Laird, li. Blackflies: the future for biological methods in integrated control. Academic Press, London. Rothfels, K.H. and D.M. Freeman. 1977. The salivary gland chromosomes of seven species of Prosimuliurn in the mixtum (IIIL-1) group. Can. J. Zool. 55s 482-507. Rothfels, K.H. and D.M. Freeman. 1983. A new species of Prosimuliurn (Dipteras Simuliidae)s an interchange as a primary reproductive isolating mechanism? Can. J. Zool. 61s 2612-2617. Rothfels, K.H., R. Ferraday and A. Kaneps. 1978. A cytological description of sibling species of Simuliurn venustum and S. verecundum with standard maps for the subgenus Simuliurn Davies (Diptera). Can. J. Zool. 56s 1110-1128. Rubtsov, I.A. 1971. Mermithids from blackflies of Western Europe. Scr. Fac. Sci. Nat. Univ. Purkyn. Brun. (Biol.) 2s 97-132. Schlein, Y., B. Yuval and A. Warburg. 1984. Aggregation pheromone released from the palps of feeding Phiebotomus oapatasi■ J. Insect Physiol. 30s 153- Schreck, C.E., N. Smith, K. Posey and D. Smith. 1980. Observations on the biting behavior of Prosimuliurn and Simuliurn venustum. Mosq. News 40s 113-115. Service, M.W. 1977. Methods for sampling adult Simuliidae, with special reference to the Simuliurn damnosum complex. Trop. Pest Bull. No. 5, C.O.P.R. 48 pp. Shemanchuck, J.A. 1977. Effects of adulticides, repellents and attractants on populations and behavior of black flies around cattle. p. 138-145. Ins Proc. first inter—regional conf. on North American Blackflies. Mimeograph Report, University of New Hamphsire. Shipitsina, N.K. 1963. Parasitic infection of simuliids (Diptera) and its effect on ovarian function. Zool. Zh. 42s 291-294. Silberglied, R. E. 1979. Communication in the ultraviolet. Ann. Rev. Ecol. Syst. 10s 373—398. 192 Simmons, K.R. and J.D. Edman. 1978. Successful mating, oviposition and complete generation rearing of the multivoltine black fly Simulium decorum (Dipteras Simuliidae) in the laboratory. Can. J. Zool. 56s 1223-1225. Simmons,K.R. and J.D. Edman. 1981. Sustained colonization of the black fly Simuliurn decorum Walker (Dipteras Simuliidae). Can. J. Zool. 59s 1-7. Simmons,K.R. and J.D.Edman. 1982. Laboratory colonization of the human onchocerciasis vector Simulium damnosum complex (Dipteras Simuliidae), using an enclosed, gravity-trough rearing system. J. Med. Entomol. 19s 117-126. Simmons, K.R. and R.D. Sjogren. 1984. Black fly (Dipteras Simuliidae) problems and their control strategies in Minnesota. Bull. Soc. Vector. Ecol. 9s 21—22. Smith, S.M. 1966. Observations on some mechanisms of host finding and host selection in the Simuliidae and Tabanidae (Diptera). M.Sc. Thesis, McMaster Univ., Hamilton, Ontario. Smith, J.J.B. and W.G. Friend. 1982. Feeding behavior in response to blood fractions and chemical phagostimulants in the black—fly Simuliurn venustum. Physiol. Entomol. 7s 219-226. Snyder, T.P. 1982. Electrophoretic characterization of black flies in the Simulium damnosum and verecundum species complexes (Dipteras Simuliidae). Can. Entomol. 114s 503-507. Snyder, T.P. and M.C. Linton. 1983. Electrophoretic and morphological separation of Prosimulium fuscum and P. mixutm larvae (Dipteras Simuliidae). Can. Entomol. 115s 81-87. Sokal, R.R. and F.J. Rohlf. 1969. Biometry. Freeman and Co. San Francisco. 766. pp. Spotte, S. 1979. Fish and invertebrate cultures water management in closed systems. 2nd ed. Wiley Interscience, New York. 179 pp. 193 Steel, R.G.D. and J.H. Torrie. 1960- Principles and procedures of statistics. McSraw-Hill, New York. 481 pp. Steelman, C.D. 1976. Effects of external and internal arthropod parasites on domestic livestock production. Ann. Rev. Entomol. 21s 155-178. Stokes, J.H. 1914. A clinical, pathological and experimental study of the lesions produced by the "black fly" Simuliurn venustum. J. Cut. Diseases. 32s 751-769 and 830-856. Stone, A. and H. Jamback. 1955. The black flies of New York State (Dipteras Simuliidae). N.Y. State Mus. Bull. 349s 1-144. Sutcliff, J.F. and S.B. Mclver. 1975. Artificial feeding of simuliids (Simuliurn venustum)s factors associated with probing and gorging. Experentia 31s 694-695. Sutcliff, J.F. and S.B. Mclver. 1979. Experiments on biting and gorging behavior in the black fly. Simuliurn venustum. Physiol. Entomol. 4s 393-400. Tempelis, C.H. and M.F. Lofy. 1963. A modified precipitin method for identification of mosquito blood—meals. Amer. J. Trop. Med. Hyg. 12s 825-831. Thompson, B.H. 1976a. Studies on the attraction of Simuliurn damnosum s.l. (Dipteras Simuliidae) to its hosts. I. The relative importance of sight, exhaled breath and smell. Tropenmed. Parasit. 27s 455—473. Thompson, B.H. 1976b. Studies on the attraction of Simuliurn damnosum s.l. (Dipteras Simuliidae) to its hosts. II. The nature of substances on the human skin responsible for attractant olfactory stimuli. Tropenmed. Parasit. 27s 83—90. Thorsteinson, A.J., G.K. Braken and W. Henec. 1965. The orientation of horse flies and deer flies (Tabanidaes Diptera). III. The use of traps in the study of orientation of tabanids in the field. Entomol. Exp. Appl. 8s 189-192. Tyndale-Biscoe, M. 1984. Age-grading methods in adult insectss a review. Bull. Entomol. Res. 74s 341—377. 194 Vannote, R.L. and B.W. Sweeney. 1980. Geographical analysis of thermal equilibria: a conceptual model for evaluating the effect of natural and modified thermal regimes on aquatic insect communities. Am. Natur. 115: 667-695. Waage, J. 1979. The evolution of insect/vertebrate associations. Biol. J. Linn. Soc. 12: 187-224. Waage, J. 1980. Curse of the vampire: The evolution of blood-sucking insects. Antenna 4: 112. Walsh, J.F. 1980. Sticky trap studies on Si mul i urn damnosum s.l. in Northern Ghana. Tropenmed. Parasit. 31: 479-486. Walsh, J.F., J.B. Davies and B. Cliff. 1981. World Health Organization Onchocerciasis Control Programme in the Volta River Basin. p. 85-104. In: Laird, M., Blackflies: the future of biological methods for integrated control. Academic Press, N.Y. Washino, R.K. and C.H. Tempelis. 1983. Mosquito host bloodmeal identification: methodology and data analysis. Ann. Rev. Entomol. 28: 179-201. Welch, H.E. 1962. New species of Gastromermis. Isomermis and Mesomermis (Nematoda: Mermithidae) from black fly larvae. Ann. Entomol. Soc. Am. 55: 535-542. Wenk, P. 1965. Uber die biologie blutsaugender Simuliiden (Diptera). II. shwarmverhalten, geschlechterfindung und kopulation. Z. morph, okol. tiere. 55: 671-713. Wenk, P. 1976. Koevolution von Ubertrager und Parasit bei Simuliiden und Nematoden. Z. Angew. Entomol. 82: 38-44. Wenk, P. 1981. Bionomics of adult black flies. pp. 259- 284. In: Laird, M. Blackflies: the future for biological methods in integrated control. Acedemic Press, London. Wenk, P. and J.N. Raybould. 1972. Mating, blood feeding and oviposition of Simuliurn damnosum Theobald in the laboratory. Bull. Wld. Hlth. Org. 47: 627—634. Wenk, P. and G. Schlorer. 1963. Wirtsorientierung und Kopulation bei blutsaugenden Simuliiden (Diptera). Z. Tropenmed. Parasitol. 14: 177-191. 195 Williams, C.B. 1962. Studies on black flies (Diptera: Simuliidae) taken in a light trap in Scotland. III. The relation of night activity and abundance to weather conditions. Trans. R. Entomol. Soc. Lond. 114: 28—47. Williams, C.B. 1964. Nocturnal activity of black flies (Simuliidae). Nature, Lond. 201: 105. Wolfe, L.S. and D.G. Peterson. 1960. Diurnal behavior and biting habits of black flies (Diptera: Simuliidae) in the forests of Quebec. Can. J. Zool. 38: 489-497. Wotton, R.S. 1977. The size of particles ingested by moorland stream blackfly larvae (Simuliidae). Oikos. 29: 332-335. Wright, R.E. and G.R. DeFoliart. 1970. Some hosts fed upon by ceratopongonids and simuliids. J. Med. Entomol. 7: 600. Yang, Y.J. and D.M. Davies. 1968. Digestion, emphasizing trypsin activity, in adult simuliids (Diptera) fed blood, blood-sucrose mixtures and sucrose. J. Insect Physiol. 14: 205—222. APPENDIX I CYTOSPECIES IDENTIFICATION OF SIMULIUM VENUSTUM FROM FRANKLIN COUNTY Introduction Simulium venustum is a complex of species in North America composed of 10 known siblings (Rothfels 1979, 1981, Rothfels et al- 1981). Larvae of these complexes can be identified to species by differences in sex chromosomes, fixed autosomal inversions, and a number of floating inver sions on the polytene chromosomes of larval salivary glands (Rothfels 1956). There are no reliable means, including isozymes electrophoresis, for identifying adults within the complex (Ferraday pers. comm.). The purpose of this study was to determine which cytospecies of the §. venusfcyfQ complex are present in the eastern part of Franklin Co. Materials and Methods Simulium venustum larvae were collected weekly be¬ ginning in early April through June at 21 sites during 1980 and 1981. Larvae were placed in plastic bags, packed in ice and returned to the laboratory. Mature §. venustum 196 197 with white respiratory histoblasts were placed in cold Carnoy's solution (3 parts 95% ETOH: 1 part glacial acetic acid). The larvae were processed by Dr. K. Rothfels and associates, Dept, of Botany, Univ. Toronto. Results Most larvae scored in 1980 were not conclusively iden¬ tified because their chromosome preparations were not clear enough for verifying inversions on the IIIL arm. The reason for this most likely was due to poor preservation technique (R. Ferraday pers. comm.). Of the 78 female and 21 male larvae examined, all had the IISCC and IILCC inversions. Larvae with these inversions are either CC or CC1 but without knowing the inversions on the IIIL arm, it was not possible to separate them. Five female larvae collected 3 - VI - 80 from Tyler Brook were positively identified as CC. Fifty-seven larvae scored from 1981 collections were all identified as CC. Table 44 is a list of the collection sites, sample dates and number of larvae scored from each. 198 Table 44- Collection sites in Franklin Co- where S. venustum CC were positively identified in 1981 samples. no. S. venustum CC larvae identified site/date of sample males females Dud1eyvilie Brook 21 - IV - 81 1 5 Wickett Pond 5 - V - 81 0 8 Sibley Swamp no.1 17 - V - 81 0 4 Sibley Swamp no. 2 18 — V — 81 1 5 Wendell St. Forest beaver pond 17 - V - 81 1 2 West Branch Swift River 30 - V - 81 0 1 Saw Mill River - lake outlet 17 - IV - 81 3 5 21 - IV - 81 2 7 Saw Mill River - 1 km below lake 17 - IV - 81 1 4 5 — V — 81 3 4 totals 12 45 APPENDIX I I BLACK FLY SPECIES THAT ANNOY AND BITE HUMANS IN FRANKLIN CO. MASSACHUSETTS Introduction There are 31 black fly species known from Massachu¬ setts; 29 of these have been collected from Franklin Co. (Simmons unpubl. data). Preliminary studies by Holbrook (1967) suggested that P. mixtum/fuscum and S. venustum were the only human biters in Franklin Co. The objective of this study was to determine which spring species are at¬ tracted to and annoy humans, and which species bite. Materials and Methods Preliminary sampling was done in 1980 to determine the relative importance of P. fpixtum/fuscutn and g. ygnustum as human pests in eastern Franklin Co. Sampling procedures were the same as those described in Chapter III . Five sample sites (listed in Table 45) were chosen based on their proximity to black fly breeding sites. Sampling was done as often as possible from late April — June on days of high fly activity. Collections were done at the edge of fields bordering wooded areas since this is where the 199 200 largest host-seeking populations generally are encountered (see Chapter IV). A more detailed experiment was conducted in 1982 to quantify which species are attracted to humans and land versus those which are attracted to a human but do not land. The sampling procedure involved having a shirtless male capture as many landing flies as possible with a mouth aspirator (Simmons and Edman 1978) for a 5-min period. At the end of the sample period, 3 overhead net sweep samples were taken. Sampling was conducted between 1600 and 1800 hrs on 5 separate days during mid- to late—May 1982 at the Lake Wyola field site described in Chapter III. Results Simuliurn venustum were captured biting humans on only 107. of the days on which they were captured in overhead sweeps. In contrast, £. mixtum/fuscufD bit on 867. of all capture days (Table 45). Eight species were attracted to humans at the Lake Wyola site; g. ygnustum and P. Ktufn/miwere the most abundant (Table 46). A greater proportion of S. venustum was collected landing on humans relative to the number collected in the net sweep samples compared to P. mixturn/fuscum (Table 46). Large numbers of St. mutata were collected in the net samples (8.87.) but none were captured on the skin of the collector (Table 46). Table 45. Number of times §. venustum and P. mixtum/fuscum were captured biting humans in a 5-min test period on days after they were collected in TJ x: H in in E a at at a as c in 2 at at ns at at L o > \ H- 4J CL •H El E 3 in U 3 E x 3 3 E o in E 3 L •H 4-+j . c s- at>-* XI U 4J ato +» s-a 4J * xj cat •H 4- 4- QlpH n u +j ato 4J in* •H XI COJ +> a at el o in OlH c -P-M at oj nl u asin o ^at 3 atc c ui+J c 0 2 at el C 4j o rph in o in u Ein SrH c at h-l c at4J 0 2 • u . raj • u . rat pH * f-< * 0) ITJ * at C *p« 2 0 c at u 0 3 n e at at l ns * at in 3 X] c at u o c r4 2 o ns at at l oj at •H 4J in at H 3 -M -J 0 X. 4J ns at ph > 0 ns 3 \n oo Table 46. Black fly species captured in 5-min landing collections and overhead net sweeps near Lake Wyola, May 1982. total no. caught in total no. caught in 5-min landing coll. overhead net sweeps species (7. of total) (7. of total) p. mixtum/ fuscum 125 (46.3) 789 (57.9) p. maanum 2 (0.07) 12 (0.08) St . mutata 0 (0) 121 (8.8) s. venustum 141 (52.9) 387 (28.4) s. verecundum 0 (0) 6 (0.4) s. decorum 1 (0.4) 3 (0.2) s. jenninqsi* 1 (0.4) 39 (2.9) s. vittatum 0 (0) 5 (0.4) ♦Potentially includes S. fibrinf1atum. S. penobscotensis, S. nvssa and S. jenninasi. 203 Table 47. Species of Franklin Co. black flies collected biting humans at least once and species collected in over— head net sweeps but never collected biting humans in 161 samples, 1980 - 1983. species captured species captured by biting humans on at overhead net sweeps least one occasion but never biting Prosimul ium magnum St eg op tern.* mutat* Prosimulium mixtum/fuscum Simulium aureum Prosimuliurn fontanum Simulium croxtoni Syme and Davies Simulium vernum Simulium decorum Simulium jenninqsi complex* Simulium parnassum Mailoch Simuliurn venustum Simulium verecundum Simulium vittatum Simulium tuberosum ♦Potentially includes S. fibrinf 1 atum, S. oenobscotensis. S. nvssa. and S. jenninasi. 204 A total of 161 biting collections were made between April and July from 1980 and 1983 at various sites in Franklin Co. (Table 47). Females of 10 species bit on at least 1 occasion but £. mixtum/fuscum and venustum were the only 2 species that could be considered regular biters (Table 47). Four species collected in overhead net sweeps were never captured biting. Discussion Results from this study confirm Holbrook's (1967) finding that P. mixtum/fuscum and §. venustum are the major biters of humans in Franklin Co. They also support obser¬ vations (Chapter III, Holbrook 1967, Schreck et al. 1980) that P. mixtum/fuscum is a more aggressive biter of humans than §. venustum. The fact that S. venustum readily landed and crawled on humans (Chapter III, Table 46) makes it an annoying pest despite the fact that most do not bite. Ten to 207. of S. venustum that land on a non-defensive, shirtless human bite (Chapter III). However, the actual engorgement rate is probably much lower. After landing, S. venustum walk around for ca. 3 sec. before initiating feeding (Chapter III). This increases their chance of being detected, prevented from feeding, or being killed. Nearly all P. mixtum/fuscum that land on exposed skin of humans begin feeding immediately (Simmons unpubl. data).