University of Ghana http://ugspace.ug.edu.gh BIOLOGY AND ECOLOGY OF THE PREDATORY MOSQUITO, CULEX (LUTZIA) TIGRIPES GRANDPRE AND CHARMOY (DIPTERA:CULICIDAE) IN SOUTH-EASTERN GHANA. by Maxwell Alexander Appawu A DISSERTATION Submitted to the Zoology Department in partial fulfilment of requirements for the degree of MASTER OF PHILOSOPHY University of Ghana 1990 University of Ghana http://ugspace.ug.edu.gh ii DECLARATION This study was undertaken by Maxwell Alexander Appawu at the Department of Zoology and the Noguchi Memorial Institute for Medical Research, University of Ghana under the supervision of Dr.S.Q.Quartey ~ M.A.Appawu ~ S.Q.Quartey (Author) (Supervisor) University of Ghana http://ugspace.ug.edu.gh iii DEDICATION To my parents, Mark and Mante, my wife, Martha, and children, Kofi Kyenkyehene, Mame Mante and Mamaa Ampofoa. University of Ghana http://ugspace.ug.edu.gh iv ACKNOWLEDGMENT I would like to extend my deepest appreciation to all the people who contributed in various ways to the completion of my M.Phil. programme. special thanks go to my major supervisor Dr. S.Q. Quartey whose friendship, direction and support made this work possible. I wish to express my appreciation to Dr. Yaa Ntiamoa-Baidoo a member of my supervisory committee for her contribution in forging the direction of my programme. Special thanks go to the Noguchi Memorial Institute for Medical Research for providing facilities and congenial atmosphere for doing the research work. My thanks go to Abdul Haruna for his assistance in the collection of data; Alfred Dodoo for assisting in the photographic work; George Mensah for his assistance in the preparation of this script and Mary Nukuteye for typing all the manuscripts. I am greatly indebted to the University of Ghana for their financial support that made this study possible. Lastly, but not the least, I wish to extend my warmest thanks to my wife, Martha, for her support throughout this programme and my children Kofi, Marne and Mamaa for enduring with me. University of Ghana http://ugspace.ug.edu.gh v ABSTRACT Culex (lutzia) tigripes Grandpre and charmoy is a larvivorous mosquito with all instars of the larvae feeding primarily on the immature stages of other mosquito species found in their habitats. They breed in a wide range of water bodies but seem to prefer those already containing larvae of other mosquito species.Y The fluctuations in the population of the larval instars and pupae were studied by weekly sampling throughout the year. It breeds thoughout the year and the population peaks either coincide with or follow that of the preys; with both fluctuating with the rainfall. The larval densities of ~. (~) tigripes were very small compared with those of other mosquito species; thus only 392-952 larvae of the predator were collected in the peak periods of May to July compared to 2786-8676 larvae of the prey mosquitoes. No significant correlation was noted between variations in the numbers of ~. (~) tigripes and the following physical and chemical properties of the breeding water: pH, Temperature, Chloride, Dissolved Oxygen and Total Alkalinity. Life-table studies showed the exist ,mce of high mortal i ties in the later stages of the predator. ~ Starting from egg rafts collected from the field and providing ~. ~uinguefasciatus larvae as the larval food and chicken as a source of blood, a colony of ~. (~) tigripes was started but poor insemination appeared to be the major obstacle to successful and permanent colonization. University of Ghana http://ugspace.ug.edu.gh vi Artificial insemination and copulation were also not successful. - The optimum larval developmental temperature was 30°C, and 32°C for the pupae. / Even though more prey larvae were consumed between 30-32°C than between 20-26°C, there was a reduction in weight of the final instar larvae and pupae at the higher temperatures. 'f Depending on temperature each predator consumed between 160-229 larvae of Q. guinguefasciatus during its entire larval development. with this rate of prey destruction, the predator can have big impact on the prey population despite the low proportion (1:7-9) of predator to prey./ When larvae of Q. (~) tigripes were reared on three non- living diets namely; Cerelac infant cereal, dog biscuit and milk casein, the developmental period of all instars was greatly p~olonged and only one larva, reared on milk casein developed into adult mosquito. The weights of the final instar larvae reared on non-living diets were significantly lower than those reared on larvae of ~. guinguefasciatus. Culex (~) tigripes has well developed mandibles and serrated mouthbrushes for effective predation. The effect of the following factors on prey capture were studied: mobility, size, posture, the density and the extent to which prey and predator occur simultaneously in the same habitat. Ae. aegypti which mov\'!s more frequently was more preyed upon than An. gambiae and Q. guinguefasciatus, and similarly, Q.guinguefasciatus was selected more than chironomid. The strong integument of the pupae together with their large University of Ghana http://ugspace.ug.edu.gh vii sizes, spherical shape, posture in the water and ability to move quickly afforded them a better chance of escaping predation by ~. (L) tigripes. J The effect of prey stage, predator stage and prey density on the predation rate was investigated using (~. guinguefasciatus) as prey. X It was shown that the rate of predation increased with increase in the size of the predator and the density of the prey but decreased with increase in prey size. The functional response of the predator to changing prey densities followed Hollings type II model. The handling time of the predatory larvae on preys decreased as the length of time in which they were deprived of food was increased but the daily prey consumption was not affected. ~ Cannibalism occurred in all larval stages of the predator. The rate was higher among the early instars; was lower in the presence of mosquito prey and increased with crowding. University of Ghana http://ugspace.ug.edu.gh viii TABLE OF CONTENTS Page xii List of Tables xv List of Figures Plates xviii List of List of appendices xix Chapter 1 : General Introduction and Literature review 1 1.1 scourge of mosquitoes 1 1.2 Literature review 12 1.3 Description of the study area 18 Chapter 2 : Bionomics of~. (lutzia) tigripes 23 2.1 Introduction 23 2.2 Materials and methods 24 2.2.1 Distribution of ~. (L) tigripes and association with other mosquitoes. 25 2.2.2 Characteristics of breeding places of ~. (L) tigripes 27 2.2.3 Seasonal population dynamics of ~. (L) tigripes 27 2.2.4 Life budget studies 31 2.2.5 Physical and chemical properties of breeding places of ~. (L) tigripes 34 2.3. Results and discussion 35 2.3.1 Distribution and occurrence of ~. (L) tigripes in the breeding places. 35 2.3.2 Characteristics of breeding places of ~. (L) University of Ghana http://ugspace.ug.edu.gh ix 43 tigripes 2.3.3 Seasonal population dynamics 45 2.3.4 Life budget studies 50 2.3.5 Physical and chemical properties of breeding places of Q. (L) tigripes 56 Chapter 3 : Developmental biology of Q. (L) tigripes 63 3.1 Introduction 63 3.2 Materials and methods 64 3.2.1 Development under laboratory conditions 64 3.2.2 Effect of different constant temperature on development 65 3.2.2.1 Ability of larvae and pupae to withstand high temperatures 66 3.2.3 Development on non-living diets 67 3.3 Colonization in the laboratory 70 3.3.1 Rearing of the immature stages of Q. (L) tigripes 70 3.3.2 Feeding of the adults of Q. (L) tigripes 72 3.3.3 Artificial insemination 73 3.3.4 Longevity of adults in the laboratory 74 3.4 Results and discussion 75 3.4.1 Development under laboratory conditions 75 3.4.2 Effect of temperature on the developmental duration of the immature stages of Q. (L) tigripes 78 3.4.3 Effect of temperature on predation rate of Q.(L) tigripes 82 3.4.4 Effect of temperature on the size of the University of Ghana http://ugspace.ug.edu.gh x 84 immature stages of g. (b) tigripes 3.4.5 colonization of g.(b) tigripes in the 90 laboratory 3.4.5.1 Rearing of the immature stages of g. (b) tigripes 92 3.4.5.2 Artificial insemination 94 3.4.6 Development of non-living diets 95 Chapter 4 : Predatory behaviour 101 4.1 Introduction 101 4.2 Materials and methods 102 4.2.1 Prey capture and feeding habits of g. (b) tigripes 102 4.2.2 Effect of predator stage, prey stage and prey density on the predation rate of g. (b) tigripes 103 4.2.3 Effect of water volume on predation rate of g. (b) tigripes 104 4.2.4 Feeding preference of g. (b) tigripes 105 4.2.4.1 spontaneous and induced movements of g. (b) -tigripes and some mosquito preys 105 4.2.4.2 Feeding preference by prey species 106 4.2.4.3 Feeding preference by mosquito prey stages 107 4.2.5 Wasteful killing 108 4.2.6 Functional response of g.(b) tigripes to prey density 111 4.2.6.1 Effect of handling time on predatory activity 113 4.2.7 Cannibalism 116 4.2.7.1 Effect of predator stages on cannibalism 116 4.2.7.2 Effect of crowding on cannibalism 116 University of Ghana http://ugspace.ug.edu.gh xi 4.2.7.3 Effect of presence of prey on cannibalism 117 4.3 Results and discussion 118 4.3.1 Prey capture feeding habits of £. (~) tigripes 118 4.3.2. Effect of predator stage, prey stage and prey density on predation rate of £. (~) tigripes 125 4.3.3 Effect of water volume on predation rate of £.(~) tigripes 133 4.3.4 Effect of food deprivation on predation rate 135 4.3.5 Feeding preference 140 4.3.5.1 Feeding preference by prey species 140 4.3.5.2 Spontaneous and induced movements of mosquito larvae 143 4.3.5.3 Feeding preference by prey stages 147 4.3.6 Wasteful killing 154 4.3.7 Functional response of predator to prey density 157 4.3.7.1 Effect of handling time on predatory activity of £.(~) tigripes 159 4.3.8 Cannibalism 166 Chapter 5 : General discussion and conclusion 172 References 180 Appendix Tables 199 University of Ghana http://ugspace.ug.edu.gh xii List of Tables Page Occurrence and distribution of~. (1) tigripes in 36 breeding places 2. Breeding of~. (1) tigripes in water containers (barrels) placed indoors and outdoors 40 3. Frequency of association of~. (1) tigripes with some mosquito prey species in the breeding places. 40 4. Characteristics of the breeding places of~. (1) tigripes 44 5. Instar mortalities of~. (1) tigripes from a man-hole (site A) 54 6. Instar mortalities of~. (1) tigripes from a concrete drain (site B) 55 7. Analysis of physico-chemical parameters and the larval incidence from a man-hole and a concrete drain (site A and B) 60 8. Mean developmental duration and number of larvae consumed during development of~. (1) tigripes in the laboratory 76 9. Sexual differences in the developmental duration of ~. (1) tigripes and in prey consumed during development in the laboratory 76 10. Duration in days of larval and pupal development of ~. (1) tigripes at different constant temperatures 83 University of Ghana http://ugspace.ug.edu.gh xiii 11. Survival of the immature stages of~. (~) tigripes reared at different temperatures. 87 12. Effect of temperature on number of prey consumed by 87 ~. (~) tigripes 13. Effect of temperature on total prey consumed during each stadium by~. (~) tigripes 89 14. Effect of temperature on weight and length of larvae and weight of pupae of~. (~) tigripes 89 15. Developmental period (days) of larvae and larval weight of~. (~) tigripes reared on different diets. 99 16. Survival of~. (~) tigripes larvae reared on different diets 100 17. Effect of water volume on predation rate of~. (~) tigripes 134 18. Effect of varying periods of starvation of £. (~) tigripes on time it takes to consume prey (handling time) and on number of prey consumed per day 139 19. Feeding preference of £. (~) tigripes for prey species 141 20. Duration of spontaneous movements of mosquito larvae 149 21. Duration of induced movements of mosquito larvae 150 22. Feeding preference of~. (~) tigripes for mosquito prey stages 152 23. Incidence of cannibalism in the newly emerged 1st-stage larvae of~. (~) tigripes 168 24. Cannibalism among different larval instars of £. (~) tigripes 169 University of Ghana http://ugspace.ug.edu.gh xiv 25. The rate of cannibalism of the 4th-stage~. (~) tigripes on different stages of the same species. 169 26. Effect of crowding of larvae of~. (~) tigripes on the rate of cannibalism 170 27. Effect of the presence of mosquito prey larvae on the rate of cannibalism of~. (~) tigripes 170 University of Ghana http://ugspace.ug.edu.gh xv List of Figures Page 1. Location map of Research Area. 19 2. Mean monthly rainfall, temperature and relative humidity of Accra (April 1989 - March 1990). 21 3. Seasonal variations in the numbers of~. (~) tigripes from two breeding sites: A man-hole (site A) and a concrete drain (site B) 46 4. Seasonal abundance of~. (~) tigripes and mosquito preys in a man-hole showing oscillations of predators and prey (site A) 48 5. Seasonal abundance of~. (~) tigripes and mosquito preys in a concrete drain showing oscillations of predators and prey (site B) 49 6. Age distribution and survivorship curve of the immature stages of~. (~) tigripes from a man-hole (site A) 52 7. Age distribution and survivorship curve of the immature stages of~. (~) tigripes from a concrete drain (site B) 53 8. Fluctuations in some physical and chemical components of a manhole (site A) breeding~. (~) tigripes 57 9. Fluctuations in some physical and chemical components of a concrete drain (site B) breeding~. (~) tigripes 58 10. Measurement of head-width of larval instars of~. (~) University of Ghana http://ugspace.ug.edu.gh xvi 77 tigripes Duration of larval development of ~. (L) tigripes at 11. different constant temperatures 79 12. Duration of pupal development of Q. (L) tigripes at different constant temperatures. 80 13. Tota~ duration of pre-adult (larva + pupa) development of ~. (L) tigripes at different constant temperatures 85 14. The numbers of Q. guinguefasciatus consumed in 24 hours by Q. (L) tigripes at different constant temperatures 86 15. Effect of different constant temperatures on the size (length and weight) of 4th-stage larva and weight of pupa of Q. (L) tigripes 88 16. Survival of adult~. (L) tigripes in laboratory 91 17a&b. Effect of predator stage, prey stage and prey dEnsity on the numbers of Q. guinguefasciatus consumed by (a) 1st-stage and (b) 2nd-stage Q. (L) tigripes larvae 126 18a&b. Effect of predator stage, prey stage and prey density on the numbers of ~. guinguefasciatus consumed by (a) 3rd-stage and (b) 4th-stage Q. (L) tigripes larvae. 127 19. The numbers of Q. guinguefasciatus consumed by different stages of~. (L) tigripes 130 20. The numbers of different stages of Q. guinguefasciatus consumed by~. (L) tigripes 131 21. The numbers of ~. guinguefasciatus consumed at different University of Ghana http://ugspace.ug.edu.gh xvii 21. The numbers of Q. guinguefasciatus consumed at different densities by Q. (1) tigripes 132 22. The influence of the duration of food deprivation on the predation rates of Q. (1) tigripes 136 23. Effect of the duration of food deprivation of £.(1) tigripes on the predatory activity 137 24. Feeding preference of £. (1) tigripes, when offered a choice of 3 prey species (showing order of prey selection) 142 25a. Rate of prey consumption by final (4th) instar larvae of Q. (1) tigripes 155 25b Rate of prey killing by final (4th) instar larvae of Q. (1) tigripes 155 26. Functional response of Q. (1) tigripes to prey density 158 27. Changes in handling time between larval stages of £. (1) tigripes feeding on 2nd stage prey larvae 163 28. Changes in handling time of final larval instar of £.(1) tigripes feeding on different stages of prey larvae 164 29. Changes in handling time of £. (1) tigripes with successive feeding on the same stage of prey larvae. 165 University of Ghana http://ugspace.ug.edu.gh xviii List Of Plates Plate 1. A manhole which serves as a soakaway for septic tanks (Site A). It was breeding~. (~) tigripes throughout the study period (April 1989 to March 1990) . 29 2. A concrete drain (Site B) which was breeding~. (~) tigripes throughout the study period (April 1989 to March 1990). 30 3. Cages used for the breeding and maintenance of adult ~. (~) tigripes in the laboratory. 71 4. Ventral side of the head of fourth stage larva of ~.(~) tigripes showing mouth brushes and a well developed mandible with a sharp claw (arrowed) (magnification X 250). 120 5. Part of the mouth brushes of ~.(~) tigripes showing numerous fine teeth on the lamellae (magnification X 500). 121 University of Ghana http://ugspace.ug.edu.gh xix Appendix Tables Page 1. Data form for extensive survey of predatory mosquito, ~. (b) tigripes 199 2. Data form for intensive survey of predatory mosquito, ~. (b) tigripes 200 3. Numbers of~. (b) tigripes and other mosquito larvae collected in the extensive survey 201 4. Frequency distribution of~. (b) tigripes from a man- hole and a concrete drain (sites A and B) 202 5. Numbers of immature stages of~. (b) tigripes and some mosquitoes collected from a man-hole (site A) 203 6. Numbers of immature stages of~. (b) tigripes and some mosquitoes collected from a concrete drain (site B) 204 7. Numbers of immature stages of~. (b) tigripes collected each day in 100 samples from a man-hole (site A) 205 8. Numbers of immature stages of ~. (b) tigripes collected each day in 100 samples from a concrete drain (siteB) 205 9. Life table for~. (b) tigripes in a man-hole (site A) 206 10. Life table for~. (1) tigripes in a concrete drain (site B) 206 11. Physico-chemical analysis of water from a man-hole and a concrete drain (sites A and B) 207 12. Ability of larvae and pupae of~. (1) tigripes to withstand high temperature. 208 University of Ghana http://ugspace.ug.edu.gh xx 13 . Summary of the colonization of Q. (b) tigripes in 209 the laboratory. 14. Number of eggs laid by Q. (b) tigripes and the time taken for eggs to mature in the laboratory 210 15. Longevity of adult Q. (b) tigripes under laboratory conditions 211 16. Average composition of Cerelac infant cereal as given by the manufacturers (Food Specialities Ghana Ltd.) 212 17a Average composition of Dog biscuits as given by the manufacturers (Nippon Pet Food Co. Ltd, J~pan) 213 17b Average composition of Milk Casein as given by the manufacturers (Wako Pure chemicals Ltd.,Japan) 213 18. Effect of predator stages, prey stages and prey densities on the number of prey consumed by Q. (b) tigripes 214 19. Duration of spontaneous movements of mosquito larvae 215 20. Duration of movements of mosquito larvae after stimulation by tapping the bowls 216 21. Dura~ion of movements of mosquito larvae after stimulation by the tapping the larvae. 217 22. Mean number of Q. guinguefasciatus eaten or killed but not eaten by the final (4th) instar larvae of Q. (b) tigripes 218 23. Functional response of Q. (b)tigripes to prey density 219 24. Feeding preference of Q. (b) tigripes, when offered a choice of 3 mosquito prey species 219 University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER 1.0 GENERAL INTRODUCTION AND LITERATURE REVIEW 1.1 Scourqe of mosquitoes There are about 3,324 species and subspecies of mosquitoes belonging to 37 genera all contained in the family culicidae (Service 1990). The family contains three sub- families: Toxorhynchitinae, Anophelinae (anophelines) and Culicinae (culicines). The distribution of mosquitoes is world-wide but about 19% of all known species, have been recorded from the Afro- tropical region, which is defined as the subsaharan mainland Africa and offshore islands including Madagascar. Malaria, yellow fever, lymphatic filariasis and viral diseases are the major mosquito borne diseases. These diseases cause a high percentage of morbidity and mortality in the human popUlations in the tropics. The role of mosquitoes as vectors of these diseases has given them economic and medical importance in these parts of the world. In many temperate countries, mosquitoes may be of little or no importance as disease vectors but they can, nevertheless cause considerable annoyance because of their bites. In the Arctic Circle, for instance, the numbers of mosquitoes biting can be so great at certain times of the year as to make almost any outdoor activity impossible. In North America, more money is spent on killing culicine mosquitoes than is expended in most tropical countries where they are important vectors of disease (Gillett 1971). University of Ghana http://ugspace.ug.edu.gh 2 Malaria Sixty species out of the 400 known species of Anopheline mosquitoes throughout the world are important vectors of malaria under natural conditions. They are also known to be vectors of filariasis and viral diseases, but it is their role as the sole transmitters of malaria probably that makes the anopheline mosquito the most important vector of human disease in the world. The World Health Organization (WHO, 1985) estimates that there are 110 million new cases of malaria a year, although about 270 million people may be carrying the malaria parasite. They further estimate that from one to two million people die each year from malaria and its complications and about 2.1 billion people (i.e. about half the World I s population) in 103 countries are at risk of malaria. About eighty percent of the African population of 400 million people live in areas where little has been done to control malaria transmission and where the problem remains virtually unchanged or is worsening (WHO 1985). It is an incapacitating and debilitating disease and thus even without causing mortality, the economic losses are too great for us to be able to estimate. Malaria was ranked as the number one major disease in order of healthy days of life lost in Ghana (Morrow et. al. 1981). Malaria cases reported in Ghana rose gradually from 593,368 in 1985 to 1.56 million in 1989 (Ahmed, K. 1990 Personal communication). University of Ghana http://ugspace.ug.edu.gh 3 Yellow fever Yellow fever is an important mosquito borne viral disease which afflicts the people in Africa and tropical areas of America. It is a zoonosis, being essentially a disease of forest monkeys which under certain conditions can be transmi tted to man. The arbovirus causing the disease is transmitted by mosquitoes. In Africa, the yellow fever virus occurs in certain cercopithecid monkeys inhabiting the forest and is transmitted amongst them mainly by Aedes africanus. Ae.africanus is a forest dwelling mosquito that breeds in tree holes and bites mainly in the forest canopy. This sylvatic or forest cycle maintains a reservoir in the monkey population (Lumsden 1951, 1952, smithburn et al 1949). Aedes simpsoni which breeds in leafaxils of banana, plantains and other plants bites monkeys at the edges of forest and thus transmits the virus from monkey to man at the edges of forest. This is transmission cycle involving Ae simpsoni, men and monkeys is sometimes referred to as the rural cycle. Aedes aegypti breeds in domestic and peri-domestic water containers and transmits the virus among human popUlations in the urban cycle of transmission, which usually occurs during epidemics. In tropical America, yellow fever is aL:;o a disease of forest monkeys, mainly cebid ones, but the most important vectors are Haemagogus and Sabethes species. Man becomes infected by disturbance of the tree top mosquitoes during tree-felling. The disease has occurred in Africa as sporadic University of Ghana http://ugspace.ug.edu.gh cases of jungle yellow fever, mainly in the forest areas or during outbreaks leading to high mortality. Between 1960 and 1962, a dramatic epidemic affected Southern Ethiopia where it was estimated that 100,000 cases and 30,000 deaths occurred in an area with a population of one million (serie et al 1968) . Estimates made on the basis of serological evidence produced the figure of approximately 40,000 infections and a death rate of about 10% during an urban yellow fever outbreak in Sudan in 1940 (Kirk 1941). Typical urban outbreaks occurred in Accra, Ghana, in 1926 and 1927, and again in 1937 (WHO 1986). Another outbreak in the Upper and Northern regions of Ghana in 1969-70 had 318 cases reported and 79 deaths. Five out of the then nine regions were affected with heavy mortalities recorded in volta and Eastern regions during an outbreak in 1977-80 (Ministry of Health 1980). Yellow fever like malaria continues to be a major threat in endemic zones of Africa where the virus reappears even after long periods of quiescence. Other Viral diseases There are about 200 different viruses known to be transmitted by arthropods, over 100 of them are transmitted by mosquitoes. More than a 100 species of mosquitoes belonging to no less than 16 of the 37 genera have so far been implicated one way or another in the transmission of viruses; at least 40 of these belong to the genus Aedes, and some 20 each to Culex and Anopheles. Some of the arboviruses transmitted to man University of Ghana http://ugspace.ug.edu.gh 5 by mosquitoes are Chikungunya in East Africa and India; o I nyongnyong , a non-fatal but painful disease found predominantly in Kenya and Uganda; Bunyamwera in Africa; west Nile in Africa, Europe, Asia, Trinidad and Panama; and Murray valley in Australia. Both the classical and haemorrhagic forms of dengue are transmitted principally by Ae aegypti. Various species of mosquitoes including, Mansonia, Aedes and are known to transmit viral encephalitis, the most important are Eastern, Western, Venezuelan, st. Louis and Japanese encephalitis; all these involve a zoonoses with birds. The first three which are sometimes referred to as equine encephalitis are very virulent to horses and man. Filariasi~ The ;:orld Health organization estimates that about 90 million people are infected with lymphatic filariasis, the cause of elephantiasis in the world WHO 1990) . It is transmitted by various species of Culex mosquitoes. Members of Culex pipiens complex, and in particular Culex guinguefasciatus which is a night biting mosquito, are the most important vectors of the nocturnally periodic form of Wuchereriq bancrofti - the parasite which causes this disease. In addition to these mosquitoes, other species of Culex and some species of Aedes are considered to be of regional importance. The non- periodic form of ~. bancrofti is transmitted mainly by Anopheles polynesiensis which has a wide University of Ghana http://ugspace.ug.edu.gh 6 distribution in the Pacific region. The culicine transmitters of Brugia malayi, another parasite which causes filariasis, belong almost entirely to the genus Mansonia of which the most widespread appears to be Mansonia uniformis. Lymphatic filariasis is not as important in Africa as it is in the Far East. In Africa, Anopheline mosquitoes which transmit malaria have been implicated in the transmission of filariasis (Coker 1986). These scourges of mosquitoes cause tremendous pain and suffering ranging from internal organ damage, disabling anaemia aLd death. Beyond their toll of individual illness and death, these diseases have insidious effects on society. They impede national and individual development, impair intellectual and physical growth, and exact a huge cost in treatment and control programmes. Control of Mosquitoes With all the improvements in drugs and vaccines which have taken place, extensive reliance still has to be placed on reducing the population of vectors of these diseases. Measures for controlling the arthropod vectors of diseases including mosquitoes may be considered in two categories, the first comprising measures which may be employed to mechanically prevent the vector from coming in contact with the human host, and the second comprising measures of control such as chemical, biological,sanitary and cultural which are aimed at the destruction of the vector during some stage of its development. Mechanical barriers include the use of University of Ghana http://ugspace.ug.edu.gh 7 repellent substances, mosquito proofing and bed nets. Insecticidal control methods are aimed at both the immature and the adult stages. Insecticides such as DDT, Dieldrin and HCH e.t.c., have been used extensively for mosquito control. The success achieved initially with insecticidal control of mosquitoes were so spectacular that for some time it was not only the control of the vectors but their total eradication was also considered. Chemical control has a number of advantages - It is very effective and the results are quickly seen. It is readily available and relatively simple to use. It is the best method during epidemics. Yet a number of problems have been found to be associated with this method of controlling vectors, some of which are: (a) The need for repetitive application for control to be effective. (b) Development of resistance to the chemicals by the mosquito populations (c) Deleterious effects of the insecticides on populations of non- target organisms, including predators and parasitoids. (d) Resurgence of treated populations, where the application of the chemical may destroy the natural enemies of the vector with the result that the vector or pest increases to a level even higher than that prior to pesticide application. University of Ghana http://ugspace.ug.edu.gh 8 (e) Elevation of secondary vector or pest to a status of primary importance. (f) General pollution of the environment and biomagnification of the pesticide in living organisms. The growing awareness of the limitations and ecological hazards of insecticidal control, has stimulated the search for biological vector control agents. The natural enemies used as biological control agents are parasitoids, predators and pathogenic organisms. Biological control has a number of distinct advantages not offered by most other approaches to vector control. It is relatively permanent once it is established. The natural enemies on which it depends are self perpetuating, and they continually adjust to changes in the population density of the pest or vector they attack. Biological control has no side effects such as toxicity or environmental pollution, and they are not harzadous to the user. Predation Predation is one of the major causes of changes in numbers of composition of natural animal populations. This has been revealed by numerous field studies of various types,for example : (1) Increased survival of prey in small parts of the environment from which predators had been excluded by screens ( Connel 1961, Hancock and Urguhart 1965, Knight University of Ghana http://ugspace.ug.edu.gh 9 1958, Smith 1966). (2) Increased survival of prey during periods of artificially restricted numbers of predator ( Elson 1962, Foerster and Ricker 1941). (3) Decreased survival of prey after the introduction of a predator into a part of the environment that was formerly inaccessible to it ( Macan 1965, Wijngaarden van and Morzer Bruyns 1961). (4) Calculations from density of prey, and density and feeding rate of the predator ( Connel 1961, Dempster 1960, Elson 1962, Horton 1961) and (5) Direct measurements of predation percentages through counts of traces remaining of prey killed by predators in prey populations of known (or simultaneously estimated density) ( Gibb 1958, Hiyama et al 1960, Maclellan 1959, Pearson 1966). In all of these studies high percentages (usually above 50%) of the prey populations were destroyed by the action of predators of a single or a few species. Predators therefore have an important role to play as biological control agents. They can be very efficient in the field where they destroy large numbers of prey larvae; they are able to search for special breeding habitats used by their preys, and they can be colonized in large numbers for release in the field. Predation on mosquitoes by larvivorous predators is considered to be one of the most economic and lasting anti- mosquito University of Ghana http://ugspace.ug.edu.gh 10 measures ( Sailer and Lienk 1954, Reddy and pondian 1974). From field and laboratory studies, aquatic insects like odonates (Hinman 1934, Hati and Gosh 1965), hemipterans (Ellis and Orden 1970), coleopterans (Christophers 1960,Bates 1965) and other invertebrates such as Planaria (Pal and Ramalingam 1981) have been found to play an important role in mosquito control. Among mosquito predators, the larvivorous fish Gambusia affinis has been studied most and it has been an agent of choice for controlling mosquitoes in many areas in the tropics. Matchavan (1976) however reported that insect predators like dragonfly numphs and hemipteran bugs devour larvae of Culex guinguefasciatus and Ae. aegypti in larger numbers than the well known larvivorous fishes , §. affinis and Poecillia reticulata. Mosquito larvae are mainly filter feeders, but a few species belonging to the subgenus Lutzia of Culex have been reported to be voracious predators (Rajasekaran and Chowdiah 1972, Surtee 1959). The larvae of all the species of subgenus Mucidus of Aedes and the genus Toxorhynchites have also been found to be predators (Service 1990, Trpis 1972). The genus Toxorhynchites contains about sixty-six species but only five species have been studied. These are Tx. brevipalpis (Brug), Tx. splendens (Wiedeman), Tx. rutilus rutilus (Coquillett), Tx. rutilus septentrional is (Dyar and Knab) and Tx. amboinensis (Doleschall) (Pal and Ramalingam 1981). The subgenus Lutzia contains only about five species which are widely distributed in the tropics. Of University of Ghana http://ugspace.ug.edu.gh 11 these only three have been studied to some extent, namely: Culex (Lutzia) fuscanus (Wiedemann), Culex (Lutzia) halifaxii (Theobald) and Q. (Lutzia) tigripes. Recently, there has been a renewed interest in predatory mosquitoes. Because the larvae are predaceous on other mosquito larvae and they co-exist with the vectors in aquatic habitats, their potential usefulness for the biological control of certain vector mosquitoes in the tropics has been favourably discussed (NAS 1973) • T.le aim of this study is to obtain information on the biology and ecology of the predatory mosquito, Q. (Lutzia) tigripes, to identify the breeding habitats and to determine some of the factors that affect the breeding and the changes in the population of this species. The study is also to understand the predator - prey interaction such as the role played by various predator stages, prey stages and prey densities on the predation rate, and the factors that affect prey selection and the predatory efficiency of Q. (Lutzia) tigripes. The information from these studies will hopefully form the basis for developing a biological control programme of mosquitoes of medical importance in an ecologically compatible and economically feasible vector management system. Considering the magnitude of the effect of malaria and other mosquito-borne diseases in Ghana and Africa as a whole, this work will increase our understanding of the biology and ecology of the predatory mosquito Q. (~) tigripes, which will be a useful basis for designing vector programmes which when University of Ghana http://ugspace.ug.edu.gh 12 coupled w~' th c h emo the rapY may reduce the scourge of mosquito borne diseases in the country. 1.2 Literature Review Morphology The larvaE~ of Culex (l!) tigripes are whitish with a brown siphon and head and they are about 10 to 11mm long. The larvae have been described as unmistakable by Hopkins (1952) because of the distinct appearance of the mouthbrushes. The mouth-brushes have been modified for predation; they have been converted into stout bristles with sharp teeth (Hopkins 1952, MacGregor 1927). The portion of the head anterior to the clypeal spines projects strongly forward and it increases the area available for attachment of mouthbrushes. The antennae are extremely short (usually less than one-quarter the length of head). It is smooth with a tuft reduced to a minute seta near the base. The siphon is also very short with an index of about 2. The siphonal index is the length of the siphon divided by its basal width. The larvae normally lie nearly parallel with the surface of the water because of the shortness of the siphon. There are about 12 subventral tufts that are arranged in zigzag manner instead of being paired (Hopkins 1952). The anal segments have been described by MacGregor (1927) as chisel-shaped because the tip of the segment t~rminates obliquely. The saddle of the anal segment is much longer than wide with the surface covered with small spicules. University of Ghana http://ugspace.ug.edu.gh 13 Normally, anal gills or papillae are used to take up salts (chloride) from water. So in general, larvae living in habitats such as tree holes in which the concentration of chloride is particularly low, are found to have exceptionally large anal papillae, whereas larvae from large swamps have exceptionally small ones, especially those from brackish waters. The predaceous larvae of g.(~) tigripes and Toxorhynchites which live in small containers form an exception: their anal gills are minute (Gibbons 1932, Wigglesworth 1933, Koch 1938), and it is presumed that these larvae obtain their salt supply indirectly from the larvae which they eat. There is very little information on the morphology of the eggs of African culicines and on their biology. MacGregor (1927) described the eggs of ~. (~) tigripes as large and cigar-shaped and laid in rafts on water surface. The adult ~. (~) tigripes are large and dull brown in colour. The proboscis and palps are also dark but usually with some pale scales near the middle of underside. The thorax and abdomen are dark brown with variable markings. The legs are usually marked by pale spots (MacGregor 1927) and are equipped with unusually long and regularly arranged spines (Gillett 1972). The adults have been observed to rest in vegetation (Gillett, 1972). They are to a great extent a sylvan species but occasionally they are found elsewhere including human habitations (Horsfall 1955). University of Ghana http://ugspace.ug.edu.gh 14 Distribution Larvae belonging to the subgenus Lutzia have been known as predators of mosquito larvae for a long time (Ikeshoji 1966) and they are distributed widely over the world. Horsfall (1955) found them mostly in the tropical regions, south Africa, Mauritius and Aden. They occur in a variety of collections of water bodies; marshes with dense vegetation, pools without vegetation, concrete troughs and clear muddy water (Nieschulz et al 1934). They inhabit rock pools, ground pools and marshes in Mauritius (MacGregor 1927). Hopkins (1952) states that it is very uncommon in barrels and domestic water vessels and there is a single record from leaf axils (Kennan 1915) . It has been recorded also from old banana leaves on the ground (Hancock 1930). It has been recorded as breeding in snail shells (Harris 1942) and even in " very 3aline water" (Someren Van 1943) even though he did not record the exact level of salinity. The breeding places seem to be limited more by the presence or absence of other larvae on which to prey than by any other factor (Hopkins 1952). Hopkins, however, states that it is much more commonly found in stagnant water such as swamps, pools of all kinds and ditches than in any other types of water. Gendre (1909) observed that they occur in the margins of streams with Anopheles species during dry seasons. The record of them in streams was by Bedford (1928) but there was no indication as to whether there was any appreciable current in the stream. University of Ghana http://ugspace.ug.edu.gh 15 Macfie and Ingram (1916) recorded it among common species collected in thick forest of the Ashanti region but not in the Northern transitional zone, and Chinery (1969) also placed it among the less common mosquito species in Accra, being found in only 1.28% of the total sample of 25317. Feeding and cannibalism Information on the feeding activities of the larvae of ~. (~) tigripes has also been scanty. MacGregor (1927) reported that the larvae will feed on chironomid larvae, small nematode worms, live insects trapped in the water and sometimes young minnows. Jackson (1953) observed that ~. (~) tigripes showed some prefurential feeding on mosquitoes offered to it. It preferred Aedes aegypti to Anopheles gambiae and Culex mosquitoes. She suggested that the preferential feeding behaviour could have been influenced by the amount of movement exhibited by the prey species. Gillett (1972) stated that adult ~. (~) tigripes attack man at times by day, mainly outside houses. Most authors have reported that this species rarely or never bite man. ~. (~) tigripes made up a major part (98.30% of 78 samples) of the catches in the bird baited traps and none in man-baited traps (Snow 1983). Precipitin tests done on blood meals of mosquitoes collected in the Gambia also showed that ~. (~) tigripes fed almost exclusively on avian blood (Snow and Boreham 1973). MacGregor (1927) reported that examination of red blood cells in the stomach of this species were always found to be those of the goat. University of Ghana http://ugspace.ug.edu.gh 16 Some workers have casually observed cannibalistic behaviour among ~. (L) tigripes larvae. MacGregor (1927) stated that ~. (L) tigripes larva retains its grip on or consumes another of the same species. Hopkins (1952), however, states that the larvae will eat each other, even in the presence of larvae of other species. This was confirmed by Haddow (1942) who observed that only the larger ~. (L) tigripes larvae cannibalised the smaller ones. Colonization Limited attempts have been made to colonize the species of the subgenus lutzia in the laboratory and the few attempts at colonization have never been successful (Pal and Ramalingam 1981, Ikeshoji 1966). Nevertheless, the establishment of a lutzia colony and mass rearing in the laboratory is considered a prerequisite for its employment as a biological control agent. ~. (Lutzia) fuscanus was partly colonized and maintained to the fourth generation but it died out apparently through poor insemination (Ikeshoji 1966). Prakash and Ponniah (1978) also colonized ~. (Lutzia) halifaxii but the colony only survived a limited time. However, ~.(L) tigripes has not been successfully colonized. Natural enemies A number of insects have been observed to attack and feed on immature stages of mosquito larvae (Sailer and Lienk 1954, Lee 1967). Mead (1980), observed that a predator, identified as n~tnErto unknown species of Mesostama (of the typhloplanoid University of Ghana http://ugspace.ug.edu.gh 17 neorhabdocele group) killed most of the common mosquito species including ~. (~) tigripes. Information on pathogenic organisms affecting this species are rare. The only informaticn of infection of ~. (~) tigripes by entomopathogenic fungus was reported by Nnakumusana (1985). He tested the susceptibility of different species of culicid larvae exposed to zoospores of the entomopathogenic fungus Coelomomyces indicus for 72 hours. Infection rates of 85-100% occurred in ~. (1) tigripes whilst in the predatory larvae of Tx brevipalpis, the infection rates were 0- 22%. vyas-Patel (1988) found that ~.(~) tigripes larvae were susceptible to Romanomermis cUlicivorax, when he evaluated the ability of this helminth, to infect, develop and emerge from Kenyan mosquito hosts. Mosquitoes associated with C. eL) tigripes Larvae of Lutzia species feed virtually on any species of mosquito larvae when they are kept together in a bowl. In the field, however, each species of lutzia is associated with a particular prey species because of similar preferences of habitats. For instance, Culex (Lutzia) vorax occurs with Culex pipiens pallens and sometimes with Aedes togoi in Japan (Sasa 1954), and with various Culicidae in Korea and Manchuria (Petriesceva and Cagien 1947). Culex (Lutzia) condor occurs with Anopheles stephensi in Calcutta (Iyengar 1920); ~.lutzia halifaxii does so with Anopheles farauti in New Zealand (Laird 1946); Culex (Lutzia) bigoti with Aedes aegypti in Bazil University of Ghana http://ugspace.ug.edu.gh 18 (Howard 1910) and ~. (L) tigripes with An gambiae in Kenya (Haddow 1942). In South Africa it occurs with ~. duttoni, Aedes caballus, Aedes lineatopennis and Aedes dentatus (Nieschulz et al 1934). Macfie and Ingram (1923) reported Ae aegypti as a close associate of ~.(L) tigripes in tropical Africa and in Nigeria ~.(L) tigripes occurs frequently with Ae. vittatus in rock holes ( Boorman 1961). 1.3 Description of the Study Area The study area is shown in Figure 1. It extends from Ho in the Volta Region in the East through Akosombo, Akuse, Ada, Accra, Aburi, Cape Coast to Takoradi in the West. Parts of the study area for example, Takoradi and Ho occur in the forest zone, while the bulk of the area falls within the coastal savanna. The two vegetational zones have two rainfall maxima with mean annual rainfall between 125 and 200 cm in the forest area and 74 and 89 cm in the coastal savanna area. The first rainy season is from May to June with heaviest rainfall ,.1round June and the second rainy season is from September to October. The coastal savanna area is the driest part in Ghana with the driest months occurring in December and January (Dickson and Benneh 1978). Accra is a typical station in the coastal savanna area. Average monthly relative University of Ghana http://ugspace.ug.edu.gh 19 r---_r--::"'----r----,---""T<:"TJT-;;---r-1r"-I" 3 0 .I 0 1 / I t I 7"1I 7 0 . '/ I \ I \ I " o 50 I I 10I 0 Km 1 Fig . LOCATION MAP OF RESEARCH AREA University of Ghana http://ugspace.ug.edu.gh 20 humidities taken in Accra at 1500 hours are highest in May - September and the highest in the morning regimes taken at 0600 hours occur in June to September. The hottest month of the year falls in March to April just before the rainy season, while August is the coolest month (Fig.2). putting all these climatic factors together, four seasons may be observed. i. Hot-wet season, which runs from March to April and then october to November. Humidity is relatively low and rainfall is heavy but short. ii. cool-wet season, from May to June and then September which is characterised by showers of long duration with moderately strong winds. iii. Cool-dry season which starts from late July through August with very little rainfall, and then iv. Hot-dry season (harmattan season) which runs from December to February. It is the driest period of the year and temperatures are usually very high. The vegetation in the forest region exhibits deciduous characteristics but destruction of forest by man has led to some parts of the forest region to be referred to as derived savannah. The vegetation in the rest of the study area is described as coastal shrub and grassland, which consists mainly of of grassland with coastal patches of shrub and occasional trees (Dickson and Benneh 1978). University of Ghana http://ugspace.ug.edu.gh 21 100 90 >- f- 80 Ci :E 70 :> r 60 cr:: 50 ~ r 40 --- Maximum f-z Minimum 0 :E 30 «z 20 w :E 10 0 A M A S 0 N 0 F M MONTH 30 28 190 26 ;> 24 lIJ 170 22 a: :> E 20 f- 150 « E 18 a: lIJ 130 16 a:. -l -l 14 ~ « 110 lIJ IJ... 12 f-. z 90 10 a: « a: 8 « 70 w 6 I,IJ l!) C) « 50 4 <1 a:: 2 a: w lIJ > 30 « 0 >« 10 0 A M A S 0 N 0 M MONTH Fig. 2 MEAN MONTHLY RAINFALL, TEMPERATURE AND RELATIVE HUMIDITY OF ACCRA (April 1989-March J990) University of Ghana http://ugspace.ug.edu.gh 22 The primary occupation in the study area is sea fishing along the whole length of the coast and food farming and animal rearing t~wards the interior. Accra, the country's capital, is the largest urban settlement in the country. It is also the country's capital. It offers more opportunities for employment than any of the urban settlements. Tema, Takoradi and Ho are the other big cities in the study area. These towns are fast expanding and with this rate of urbanization, the problems of mosquito breeding becomes magnified. The striking features in these towns and cities are the lack of good sanitary facilities and poor drainage. The lack of good drainage system often results in many stagnant water bodies around the houses which provide suitable breeding sites for mosquitoes. University of Ghana http://ugspace.ug.edu.gh 23 Chapter 2.0 BIONOMICS OF C. (LUTZIA) TIGRIPES 2.1 Introduction Mosquito larvae and pupae are found in a variety of different habitats, ranging from large expanses of water such as swamps and cultivated fields to small collections of water as found in plant axils, snail shells etc. Some species exhibit considerable plasticity in their selection of breeding places and such larvae occur in several different types of habitats while others are more restricted in their choice. The mature stage of Q. (.1) tigripes have been reported to breed in a variety of habitats by various workers (Chapter 1). There are no reports on the preferred breeding habits, distribution and the population dynamics of this species. Thus the objectives of this study are to determine: (1) the breeding habitats in which Q. (.1) tigripes are encountered. (2) the characteristics of these breeding habitats. (3) the :':requency of association of this species with other mosquito species in its natural breeding habitats. (4) the population distribution of Q. (.1) tigripes in the breeding places. (5) the seasonal abundance and variations in the numbers of the immature stages of Q. (.1) tigripes and their mosquito preys. (6) the relationship between the numbers of the immature stages of Q.(.1) tigripes and some physical and University of Ghana http://ugspace.ug.edu.gh 24 chemical properties of the breeding places. 2.2. Materials and Methods The great diversity of mosquito larval habitats, both the size and number of which can change overnight, are inherent difficulties and problems associated with sampling mosquito larval populations. Despite these problems a few sampling techniques have been developed to obtain quantitative estimates of either larval density or population size. The small size of many of the most important mosquito breeding habitats makes them impossible to sample by many of the normal limnological methods such as drag nets, dredges, cylinders and cages (Service 1971). The different sampling methods used by different authors to estimate relative densities of mosquitoes have been discussed by Service (1976). The choice of a particular method depends on the sampling accuracy desired, the physical characteristics of the area and the behaviour of the mosquitoes. The sampling technique used in this study was restricted to the larvae and pupae only, and all samples were collected with a dipper. The dipper is undoubtedly the most commonly used tool for collecting mosquito larvae and pupae that occur in large and small collections of ground water and various container - type habitats (Service 1976, Boyd 1949, Russell et al 1963). University of Ghana http://ugspace.ug.edu.gh 25 2.2.1 Distribution of C. (L) tigripes and association with other mosquitoes An extensive survey covering an area shown in figure 1 was carried out to determine the active breeding places of ~. (L) tigripes and the types and frequency of other mosquito larvae which occur in the preferred habitats of ~. (L) tigripes. The areas sampled in Accra and Tema were areas which are known for high mosquito populations (Chinery 1969, 1970). The survey was carried out from April 1989 to March 1990, a period which covered aJl the seasons encountered in a year. Water bodies which were accesible were selected and sampling was done once a week with a dipper which was 9.5 cm in diameter and 3.5 cm deep and could hold 250 ml of water. The inside was painted white to facilitate easy detection of larvae, and a long metal handle was attached to it so that it could reach relatively inaccessible water sources. The sampling procedure was standardized as sampling was carried out by one individual, the author, who without exception always employed the same techniques for taking the samples. That is in each sampling, the dipper was quickly dipped at an angle (about 45°C) at the water surface and removed before it overflowed. If the dipper was immersed too slowly, the larvae were disturbed and they dived to the bottom with the result that many escape collection. Also each sampling site was approached carefully to avoid the shadow of the sampler or dipper falling over the water surface to disturb the larvae. An interval of about University of Ghana http://ugspace.ug.edu.gh 26 2 _ 3 minutes was allowed between each dip to allow larvae and pupae which had moved down the water to return to the surface. Preliminary trials showed that most (about 80%) of the mosquito larvae which moved to the bottom of water after being disturbed returned to the surface within one to two minutes, end nearly all surfaced within three minutes. The principle of dipping relies on the fact that nearly all mosquito larvae sooner or later must come to the water surface for oxygen (Knight 1964). Individual artificial containers were examined for the presence or absence of g.(~) tigripes but larger water bodies like ponds, lagoons, swamps etc were recorded as not breeding g. (!:) tigripes when all probable breeding sites yielded no g.(!:) tigripes after examination, and they were recorded as breeding g.(!:) tigripes if g.(!:) tigripes '.vas found in at least one dipping. A positive breeding place for g. (!:) tigripes was any site in which either a larva, pupa or egg of g.(~) tigripes was found. Samples from all positive dips were collected into specimen bottles duly labelled with the pertinent data such as the date, the place of collection, time etc. (see Appendix Table 1) and sent to the laboratory for counting and identification. Culicine larvae were identified using characters described by Hopkins (1952), and anopheline larvae using De Meillon (1947). Only the fourth stage larvae were used for the identification of species, EO when necessary younger larvae wer~ reared to the fourth stage. University of Ghana http://ugspace.ug.edu.gh 27 2.2.2. Characteristics of the breeding places All breeding places as well as the positive collections were recorded on larval survey forms (Appendix Table 1) together with the following characteristics of the breeding places, which were examined visually: (i) Permanence - as to whether the water will last during the dry season or will dry up. (ii) Flow - as to whether the water is flowing or stagnant. (iii) "Polluted" as to whether the water contains sewerage pollutants or not (ie. by visual examination). (iv) Vegetation - as to whether there are some emergent or floating vegetation in the water or not. (v) Shade - as to whether the water is directly exposed to the sun or not during the day. (vi) Other predators - as to whether there are other predators of mosquitoes in the water or not. 2.2.3. Seasonal population dynamics Two semi-permanent water sources which were found to be breeding ~. (1) tigripes and located on the campus of the University of Ghana, Legon were used for the population dynamics studies. Th e f'1 rst b reed1. ng place, Site A, is a University of Ghana http://ugspace.ug.edu.gh 28 manhole about 75x75x120 cm which is serving as a soakaway for septic tanks. It is constructed of concrete with the edges covered with green grass (Plate 1). The water has a foul smell and is polluted with urine and faecal matter. It is exposed directly to sunlight most of the time, except when the grasses at the edges grow tall and give some shade to the water during the day. The second breeding place, site B, is a concrete drain about 40 cm wide and 1200 cm long, that had been blocked at one end with materials from a construction site (Plate 2). It contained fallen dry leaves and branches. Although the water in this site favoured the accumulation of leaf litters, it remained moderately clear throughout the study period. Tap water was sometimes added to this site whenever the water level fell very low. The sampling period for the seasonal population dynamics was from April 1989 to March 1990. The same dipper as described previously was used except that in this case sampling was done three times a week (Monday,Wednesday and Friday) between 9.00 and 11.00 hours. During sampling from site A, a dip was taken near each of the four corners plus one from the middle. For site B, one dip was taken from five spots in the drain. The samples collected were taken to the laboratory and the number of eggs and the various immature stages per dip were recorded (Appendix Table 2). The mean number of larvae per dip and the degree of variability between samples (sample variance) was calculated to determine the University of Ghana http://ugspace.ug.edu.gh 29 Plate 1. A man hole which serves as a soakaway for septic tanks (site A). It was breeding £. (~ tigripes throughout the study period (April 1989-March 1990) University of Ghana http://ugspace.ug.edu.gh 31 population distribution. Eggs of mosquitoes collected were identified after they had hatched. 2.2.4. Life Budget Studies To understand the population dynamics of a particular insect, measurements of actual numbers of individuals in a generation passing through each stage of development is required. It therefore becomes necessary to construct ecological life tables of such insect. Ecological life tables can be described as the summary of the vital statistics of a population by a record of sequential measurements of individuals revealing population change of the insect throughout its life span in a natural environment. The purpose of a life table therefore is to summarize the survival and mortality rates of a population. Construction of life tables or budgets has been more extensively used in agriculture and forest entomology, but the technique has only recently been applied to mosquitoes because of lack of population data (Service 1976, WHO 1967). Life budget studies carried on mosquitoes include one on An. gambiae complex in Kenya (Service 1971, 1973), and on Ae. aegypti in Bangkok, Thailand (Southwood et al 1972). In all these studies, the data were obtained from continuous and intense studies on a single hacitat. It has been emphasized by Varley and Gradwell (1970) that the most instructive life tables will usually be based on a continuous and intensive study of a population in University of Ghana http://ugspace.ug.edu.gh 32 a single ~abitat, not by sampling different populations in a number of similar habitats in different years. Most methods of population analysis involving the construction of life budgets are designed to be used for species that have discrete generations, so that a cohort can be followed through its life. However, the majority of insects, particularly in the tropics, have overlapping generations or I ike mosquitoes, breed continuously (Southwood et al 1972). Time-specific or vertical life tables give a measure of the rate of an imaginary cohort by determining the age structure of the population at one given time. Both the age distribution of the population and deaths in the different age classes are recorded. This type of life-table is most useful with species having either overlapping generations or continual recruitment (Deevey 1947). Instar Mortalities and Survivorship Curves It has been recognized (Bates 1941) that if the duration of the different larval age classes is taken into consideration then there is a relationship between the numbers collected in the different age classes and their survivorship. The age distribution of the population should be assumed to give the same shape as the survivorship curve if the population is more or less stable (Service 1976). One hundred samples were taken daily for ten consecutive days from breeding sites A and B with a 250ml dipper. The total numbers of each larval instar as well as pupae of ~. (~) University of Ghana http://ugspace.ug.edu.gh 33 tigripes collected on each sampling day were recorded. To obtain a histogram of the age distribution of the immature stages of ~. (L) tigripes, the total numbers of different larval instars and pupae collected over the entire collecting period were divided by the appropriate instar durations. These values were plotted against age in days of the larvae and pupae, and the resultant graph represents the stage-specific age distribution. A curve drawn through the mid-points of each histogram represents the mid points in the life of each instar and give the age-specific distribution curve (Service 1976). This profile of age distribution will simulate the time-specific curve if the steady-state assumption holds. From the survivorship curve the numbers of larvae surviving to each age in days is read off to give the numbers (nx) surviving to age x. The life table is then construct~d based on the percentages of the populations entering each stage on each sampling day corrected to 1000. The columns that make up the life budget are: x - age in days nx - No of larvae surviving to age x. Ix - No per 1000 surviving to age x. dx - mortality between ages x and x+l px - probability that a larva of age x would survive to age x+l qx - probability that a larva of age x would die before reaching age x+l University of Ghana http://ugspace.ug.edu.gh 34 ex - expectation of life 2.2.5. Physical and Chemical properties of breeding places of C. eLl tigripes In order to obtain information on the physical and chemical properties of the breeding places in the study, the determination of the physico-chemical properties should be carried out at all such places during the extensive survey. However, since these properties do not remain the same in each breeding place but may change throughout the year, it would have been ideal to measure these factors throughout the seasons and at all sites. This could not be done because of logistic problems, so, two different semi-permanent habitats which breed ~. (b) tigripes throughout the seasons were chosen for the water analysis. The surface temperature and the Hydrogen ion concentration (pH) were measured twice weekly in the field immediately after the larval sampling from breeding sites A and B, using a pocket pH meter with a thermometer attached, (Iuchi Model pH 51, manufactured by Yokogawa electric works, Tokyo, Japan). The probe of the meter containing the thermometer was immersed in the water to a depth of 6cm to measure the pH and the temperature. After taking the above measurements, the water for chemical analysis was also collected in 500ml bottles, with the mouths of the bottles immersed to a depth of about 5cm below the surface. For the Winkler test to measure dissolved oxygen, 125 ml University of Ghana http://ugspace.ug.edu.gh 35 bottles were used, and the water was collected without air bubbles. The chemical properties analysed in the laboratory were chloride, dissolved oxygen and total alkalinity. The water was allowed to stand until the sediments settled at the bottom of the bottle. The supernatant water was then decanted off and analysed. Salinity or chloride content was measured by direct estimation of total chloride content by chemical reaction using a silver nitrate solution (WHO 1975). Dissolved oxygen and total alkalinity were measured by standard methods of American Public Health Association (APHA 1975) . 2.3 Results and Discussion 2.3.1 Distribution and occurrence of C. eLl tigripes in the breeding places Several arbitrary classifications have been proposed for the different types of mosquito breeding habitats (Bates 1949, Boyd 1930, Hopkins 1952, Mattingly 1969, Chinery 1969). The number of water bodies in which mosquitoes were breeding during the extensive survey were 1773, and they have been classified into six categories based on Chinery (1969). These are: 1. Water occurring in channels, which consisted mainly of concrete and earth drains, stagnant streams etc. University of Ghana http://ugspace.ug.edu.gh Table 1 Occurrence and Distribution of~ . (~) tigripes in breeding places category of No. of No. with ". Total No. Total No. breeding site Type sites Ii,,) tigripes (%) sampled in with G.l~igripes (%) sampled category 1. water in a) concrete drains 283 40 (14.13) channels b) Earth drains 117 10 (8.55, 400 50 (12.5) 2. Large a) Hydrants 109 9 (8.26) artificial b) Manholes 56 3 (5.36) 493 49 (9.94) water c) septic tanks/ containers soakaways 120 16 (13.33) d) Barrels 107 13 (12.15) e) Water tanks 101 8 (7.92) 3. Small a) Lorry tyres 412 36 (8.74) artificial b) Car parts 7 0 (0.00) 477 36 (7.55) water c) Tins/cans 52 0 (0.00) containers d) Tree holes 6 0 (0.00) 01 en 4. standing a) Ponds 82 4 (4.88) water in b) Lagoons 6 0 (0.00) 119 6 (5.04) large c) BOrrow pits 31 2 (6.45) excavations 5. Marshy a) Marshes and 76 4 (5.26) 76 4 (5.26) and swampy b) Swamps grounds 6. Standing a) Ground pools 158 4 (2.53) shallow b) Rain pools 30 0 (0.00) water on c) Tyre print 13 0 (0.00) 208 4 (1.92) the ground d) Hoof prints 7 0 (0.00) Total 1773 151 (8.52) --- -- ---- - University of Ghana http://ugspace.ug.edu.gh 37 2. Large artificial water containers, such as hydrants, septic tanks and soakaways, barrels, drums, water tanks etc. 3. Small artificial water containers, consisting mainly of discarded lorry tyres and parts, tins, cans, tree holes etc. 4. Standing water in clearly defined excavations, such as ponds, lagoons, burrow pits etc. 5. Water in marshes and swampy grounds. 6. Standing shallow water on the ground, consisting mostly of ground and rain pools, pot holes, hoof prints, lorry tyre prints, broken pipes etc. These can also be classified broadly into two main groups namely artificial or man-made and natural sources of water. Table 1 shows the occurrence and distribution of ~. (l!) tigripes in the breeding places. Water sources, both large and small, that were sampled and which were found to contain mosquito larvae were 1773 of which 151 (8.52%) of them were breeding ~.(l!) tigripes . They were encountered most often in water in channels (12.5% of breeding sites in category 1) and then in large artificial containers (9.94% in category 2), but they were found less in standing shallow water on the ground (1.92% in category 6). Concrete drains were the major source of breeding for Q. (l!) tigripes in category 1 while septic tanks and soakaways, then barrels and drums formed the University of Ghana http://ugspace.ug.edu.gh 38 major breeding places in category 2. In Jos, Nigeria, £.(~) tigripes was among the two predominant species found present in septic tanks mainly in wet and early dry seasons (Irving- Bell et al 1987). Discarded lorry tyres formed the major source of breeding for £. (~) tigripes in category 3, but smaller water containers in category 3 such as tins and cans, car parts and tree holes were found not to be breeding £. (~) tigripes, so were the many standing shallow waters on the ground in category 6. These sites which were not breeding £. (~) tigripes were mostly temporary water sources which were much more likely to dry up in the absence of rains. Because they are liable to rapid desiccation, small and temporary collections of water are often free of predators (Gillies and De Meillon 1968). Ground pools which contained £. (~) tigripes were found only during the wet season when the water persisted for some days because of the frequent rains. Even though one would expect large water bodies to be favourite breeding places for £. (~) tigripes (Hopkins 1952), they were encountered less frequently in ponds, marshes and swamps and in stagnant streams. Hopkins (1952) reported that £. (~) tigripes were very uncommon in barrels and drums, but in this study they show a relatively high occurence and were encountered in 12.15% (13 out of 107) of all the barrels and drums breeding mosquitoes. The breeding preference of £. (~) tigripes with reference to location of domestic containers, i. e. whether outdoors or indoors is shown in Table 2. It indicates that £. (~) tigripes University of Ghana http://ugspace.ug.edu.gh 39 are outdoor breeders because out of the total of 107 barrels which were breeding ~. (~) tigripes (Table 1), 86 and 21 were placed outdoors and indoors respectively and 12 (13.95%) of those placed outdoors were found to be breeding ~. (~) tigripes while only 1 (4.75%) of those placed indoors contained the predator. Culex (~) tigripes constituted only 2.50% (612 out of 11240) of the larvae collected from the extensive survey. ~. guinguefasciatus was the most predominant species collected throughout the study period and they formed 47.06% of all the larvae collected . It was followed in abundance by Ae. aegypti; An. gambiae; Ae. vittatus; ~. univittatus; ~. decens; ~. thalassius and ~. duttoni forming 22.61%, 12.53%, 6.99%, 3.86%, 2.54%, 1.24% and 0.78% respectively. The population of ~.(~) tigripes seen here is very small compared to that of the other mosquitoes. Chinery (1969) found that only 1.23% of the total mosquito sample of 25317 collected in Accra were ~. (~) tigripes. He also found ~. guinguefasciatus, Ae. aegypti and An. gambiae to be the most predominant species and the rest of the mosquito species I isted above as less common species. The frequency of association of ~. (~) tigripes with some of the other mosquito prey species is given in Table 3. It indicates that ~. duttoni was the species most frequently associated with the predator in its preferred natural breeding places. It was followed by ~. guinguefasciatusAe,Ae. aegypti, University of Ghana http://ugspace.ug.edu.gh ~ble 2 Breeding of ~. Lutzia tigripes in water containers (barrels) placed indoors ~tdoors '10. of barrels outdoor Indoor ;ampled No. (%) No. with C .(L). No. (%) No. with C{t..} outdoor tigripes (%) indoor tigripes (%) 107 86 (80.37) 12 (13.95) 21(19.63) 1 (4.75) ble 3 Frequency of association of ~. i1l. tigripes with some mosquito prey species in the breeding places Breeding places with pres species % Association with ,4..)tigripes. Mosquito species with without C.~) tigripes C.~) tigripes ~. guinguefasciatus 63 611 9.35 ~. univittatus 14 267 4.98 ~. decens 12 169 6.63 ~. duttoni 7 62 10.14 ~. thalassius 3 53 5.36 Ae. aegypti 51 492 7.56 Ae. vittatus 24 181 6.83 An. gambiae 20 352 5.38 University of Ghana http://ugspace.ug.edu.gh 41 Ae. vittatus, ~. decens, An. gambiae, ~. thalassius , and ~. univittatus. The frequency of association of a particular mosquito species with ~. (L) tigripes will depend on the extent to which their preferred natural habitats are similar. Most of the preferred breeding habitats of ~.(L) tigripes are similar to those of ~. duttoni because the predator also breeds in large and small, permanent and temporary, clean or foul and all sorts of artificial water containers (Table 1), and it does not breed in brackish water just like ~. duttoni (Ikeshoji 1966). ~. (L) tigripes occurred frequently with Ae. aegypti in breeding places such as discarded lorry tyres, barrels, water tanks, hydrants and in concrete and earth drains. ~. (L) tigripes however, was not found in smaller water containers placed indoors, discarded tin cans, and tree holes which are also preferred breeding places for Ae. aegypti (Christophers 1960, Chinery 1969). It was observed that most of the occasions when An. gambiae concurred with ~. (L) tigripes in the same habitat it occurred during the dry season, and this is due to the fact that in the dry season An. gambiae breeds mainly in habitats such as concrete and earth drains, stagnant streams and even in septic tanks (Irving-Bell et al 1987), when many of its preferred breeding places (cans and tins, discarded lorry tyres, tree holes etc) are dried out. Gendre (1909) observed that ~. (L) tigripes occurred frequently with Anopheles species in the margins of streams during the dry University of Ghana http://ugspace.ug.edu.gh 42 season. Even though the population of An. gambiae in the survey was higher than that of Ae. vittatus there was not much difference between the frequency at which they occurred with ~.(~) tigripes in its preferred natural habitats (Table 3). This may be because the preferred natural habitats of Ae. vittatus and that of the predator are much more similar than those of An. gambiae. Ae. vittatus breeds in a wide variety of places, particularly in concrete drains, water hydrants, manholes and water tanks. A study of the habitats of Ae. vittatus in the Plateau Province of Northern Nigeria by Boorman (1961), showed that Ae. vittatus occurred most frequently with ~.(~) tigripes in rock holes which were the preferred natural breeding places for Ae. vittatus. Similarly, ~. duttoni was found with ~. (~) tigripes in their breeding habitats more frequently than ~. thalassius (Table 3 ), but ~. duttoni formed only 0.78% of the total number of mosquitoes collected, whilst ~. thalassius formed 1.24% of the total pop~lation. ~. thalassius breeds mostly in brackish water besides the sea and lagoons which do not support breeding of ~.(~) tigripes, whilst ~. duttoni breeds mostly in a variety of places including ponds, marshes and swamps and also in septic tanks and soakaways, which are also preferred by ~.(~) tigripes. These observations confirm the assertion that the frequency of concurrence of mosquito species with the predator will depend on the extent of similarity of their natural breeding habitats. The frequency of concurrence of University of Ghana http://ugspace.ug.edu.gh 43 the various species with the predator will also influence the type of species which will be preyed upon by the predator. 2.3.2 Characteristics of the breeding places Table 4 shows the characteristics of the breeding places of g.(L) tigtipes. 20.53% (31 out of 151) of the sites breeding g.(L) tigripes were found to be "polluted". These sites that contained sewerage pollutants sometimes had foul- smelling water. septic tanks and soakaways formed the major part of the sites with "polluted" water which were also breeding g. (L) tigripes. Concrete drains and septic tanks are particularly abundant in the towns and cities and they are usually badly kept and so contained water containing sewerage pollutants. These, although allowing some mosquitoes such as g. (L) tigripes and g. guinguefasciatus to develop, remain unfavourable to species like An. gambiae to develop. This may also explain the low frequency of occurrence of An. gambiae in the preferred breeding habitats of g. (L) tigripes. 9.27% (14 out of 151) of the sites breeding g. (L) tigripes had some type of vegetation either floating or submerged in the water. Vegetation was found in all the marshes and swamps as well as the ponds which were breeding the predator; these were breeding sites with relatively large water bodies and are usually pe rmanent. The concrete drains which had vegetations ( 5%) were those in which soil had accumulated to allow the vegetation to grow. 13.91% of the sites had the water surface shaded from direct sunlight during the day, and 11.26% University of Ghana http://ugspace.ug.edu.gh Table 4 Characteristl.cs or tne Dreea~ny ~"'(1'-"'''' v. category Type No. "Polluted" Vegetation Shade Other Predators of with !!. breeding tigril2es No. % No. % No. % No. t places 1 Concrete drain 40 6 15.0 2 5.0 0 0 4 10 Earth drain 10 0 0 4 40.0 0 0 5 50 2 Hydrant 9 2 22.22 0 0 0 0 0 0 Manholes 3 0 0 0 0 0 0 0 0 septic tanks/soakaways 16 16 100.0 0 0 13 81.25 0 0 Barrels 13 0 0 0 0 0 0 0 0 water tanks 8 0 0 0 0 0 0 0 0 3 Lorry tyres 36 0 0 0 0 8 22.22 0 0 car parts 0 0 0 0 0 0 0 0 0 tins and cans 0 0 0 0 0 0 0 0 0 tree holes 0 0 0 0 0 0 0 0 0 ~ .fi- 4 Ponds 4 4 100 2 50 0 0 4 100 lagoons 0 0 0 0 0 0 0 0 0 Burrow pits 2 2 100 0 0 0 0 2 100 5 Marshes and swamps 4 0 0 4 100 0 0 2 50.0 6 Ground pools 4 1 50 0 0 0 0 0 0 Rain pools 0 0 0 0 0 0 0 0 0 Hoof prints 0 0 0 0 0 0 0 0 0 Lorry tyre print 0 0 0 0 0 0 0 0 0 Total 151 31 20.53 14 9.27 21 13.91 17 11.26 "Polluted" - sewerage pollutants University of Ghana http://ugspace.ug.edu.gh 45 contained other known mosquito predators such as odonates, dytiscids, notonectids and tadpoles. These results suggest that ~. (~) tigripes breeds mostly in sunlight and relatively clean water, with little vegetation and other predators of mosquitoes. 2.3.3 Qeasonal population dynamics Monthly variations in the breeding of the pre-adult stages of ~.(~) tigripes in the two semi- permanent sites - a manhole serving as a soakaway for septic tanks (Site A) and a concrete drain blocked at one end (Site B), and also the mean monthly rainfall are given in Figure 3. There are fluctuations in the numbers but two obvious peaks can be seen in the population of ~. (~) tigripes in both sites A and B. The major peaks seems to coincide with the major rainy seasons, which occur in May to July and then around October. Precipitation during the major rainy seasons is usually very heavy and sometimes cause flooding of many mosquito breeding sites, washing away many of the pre-adult stages. Flooding occurred during the course of this study and affected both sites. Sites B and A were flooded in June and July respectively. These might explain the sudden decline in the numbers of the larvae at those times in both sites. The seasonal abundance of ~.(~) tigripes and mosquito prey from both sites A and B are shown in Figures 4 and 5 respectively. The fluctuations in the populations of the predators followed the pattern of the mosquito preys; the seasonal peaks in the prey abundance University of Ghana http://ugspace.ug.edu.gh M 1990 M 1990 UJ ~ 50 II: UJ >- \ p. / \ /'" J ~ b--<>', o~ 200 Zo. \/ ''- IV = 4th I ns10 r - z 100 o o 2 3- 4 5 6 7 8 9 AGE (Days) Fig. 7 AGE DISTRIBUTION AND SURVIVORSHIP CURVE OF THE IMMATURE STAGES OF C (L) TIGRIPES FROM A CONCRETE DRAIN (Site-BT University of Ghana http://ugspace.ug.edu.gh mortalities of g. (iJ tigripes from a man hole (site A) In days No. entering Death Relative Proportion aginning ins tar in inst ar proportion dying nstar dying in daily in ins tar ~~stars _)Y (- Di 1- S d .. k** ·1 (Sci: -I ) (Di) S~~_, ) \. Sl:i-1 379 69 0.1821 0.1402 0.0873 (J1 "'" 1.33 310 80 0.2581 0.2186 0.1296 .. 54 230 90 0.3913 0.3458 0.2156 J.71 140 118 0.8429 0.5052 0.8037 ~. 34 22 20 0.9091 0.6913 1. 0410 3.38 K = 2.2776 - - juration in days ce between successive values of log of number entering instar University of Ghana http://ugspace.ug.edu.gh Instar mortalities of £. (Jd tigripes from a concrete ara1n tS1~e DJ .n days No. entering Death Relative Proportion !ginning ins tar in ins tar proportion dying lstar dying in daily in ins tar ~(star ( Di 1- S ~J/d ~ k** . / (Sh-I) (Di) SI::.: _I ) \. SI:;_I 0 300 71 0.2367 0.1838 0.1173 1. 33 229 59 0.2576 0.2182 0.1294 i 170 65 0.3824 0.3376 0.2828 ("35.741 105 90 0.8571 0.5228 0.8451 U1 U1 6.34 15 14 0.9333 0.7349 1.1761 8.38 K = 2.5507 duration in days /ce between successive values of log of number entering instar University of Ghana http://ugspace.ug.edu.gh 56 A high survival expectation is shown in the early stages than in the later ones (Appendix Tables 9 and 10) and this point is further demonstrated by the k-values (Tables 5 and 6 ) where the highest mortality occurred among the later stages (4th and pupae). The results suggest that mortality of the later stages is high and therefore the survival of these stages (4th and pupae) would probably determine the adult populations of £. (L) tigripes emerging from those habitats. 2.3.5 yhysico-chemical properties of the breeding water The results of the analysis of water from the two £. (L) tigripes breeding sites (A and B) are shown graphically in Figures 8 and 9 (Appendix Table 11). There was very little fluctuations in the Hydrogen ion concentration (pH) of the water from both breeding places (site A and B). The pH value ranged from 7.1 to 8.0. Similarly, the water surface temperature measured at a depth of 6 cm. varied only slightly throughout the seasons at both breeding places: it varied from 24.3 to 28.9°C and from 25.1 to 31.1°C in sites A and B respectively. The water in site B (concrete drain) was much shallower than the water in site A, so the temperature of water in site B increased more during the hot dry season from December to March (Figures 8 and 9) . Chloride content which also gives the measure of salinity of the water did not fluctuate much. The total alkalinity which is a measure of the amount of carbonates and bicarbonates in the water ranged from University of Ghana http://ugspace.ug.edu.gh 57 ~ ______________~ ________________~ 'OO 8 ,/A\. /,,1-'--'_-' 90 ~ / \. ..... y' '. ...... -r" 80 E z 70 >- w ... _ ...... DlssolvedOJeygen I- (D >-w ........... Tolol Alkollnlly 60 z Xo _ Chloride ...J 00:6 50 20 I- - .. ;::: 01 01 E 70 80 ~ ~ •• .-0 Total Alkalinity 70 Z w 0-- -0 Dissolved Oxygen >- ~ t!lw Chloride 60 z >x-_0 60 oa: ::i 0 !SO :1: .J¥" \ ..J :..JU 50 40 0.05) between the sexes in the developmental duration of both larvae and pupae and in the mean number of prey consumed. The ratio of males to females in this study was 1:1.18. Figure 10 shows the changes in the width of the head capsule of the larval stages during alopment oUf n~.i v~e.rs ittyig orifp eGs hian nthae hltatbpo:r/a/utogrys pace.ug.edu.gh LARVAL STAGES It 2nd 3rd 4th Pupa Total ± 0.48 1.21 ± 0.48 1.17 ± 0.38 2.63 ± 0.58 2.04 + 0.46 8.38 ± 2.38 = 24) (n = 24) (n = 24) (n = 24) (n ~ 24) (n = 24) 9 ± 15.29 27.50 ± 13.83 30.29 ± 13.48 62.25 ± 10.12 - 145.67±17.33 n = 24) (n = 24) (n = 24) (n = 24) -- - as provided with prey of the same stage -.J 0\ 11 differences in developmental duration of ~. tigripes and in prey consumed during t opments in the laboratory I Pupal Total T-test Total No. T-test days larval & pupal of prey days consumed .• 70 2.00 ± 0.45 8.27 ± 0.90 T=0.72* 145.73±14.44 0.05* (n=22) (n=22) (n=22) l.02 2.08 ± 0.49 8.46 ± 1.13 145.46±19.86 0.05 I (n=26) (n=26) (n=26) ificant) University of Ghana http://ugspace.ug.edu.gh 77 0-6 0-5 0-4 0-3 0-2 0 - 1 0 0 LARVAL INSTARS OF C (L) TIGRIPES Fig _ 10 MEASUREMENT OF HEAD WIDTH OF LARVAL INSTARS OF ~ ill TIGRIPES University of Ghana http://ugspace.ug.edu.gh 78 development. The means of the head widths followed Dyar's rule, i.e. it increased geometrically by a ratio of 1.5. 3.4.2 Effect of temperature on the duration of development of the immature stages The results of rearing larvae and pupae of Q. (~) tigripes at different constant temperatures are presented in Figure 11 and Table 10. The first instar larvae reared at 12°e survived for 2-3 days and died without moulting. At 34°e the first instar larvae were lethargic. At this temperature, the larvae remained at the top of the water and moved only after they had been disturbed. The above results suggest that 12°e is close to or below the lower developmental threshold for the immature stages and 34°e is also close to the upper developmental threshold. Further work needs to be done to establish the lower and upper developmental thresholds. The total duration of larval and pupal development at 200 e was 20.93 ± 1.62 days The development for these same stages was much faster at 26° 0 and 30 e. The total pre-adult (larva + pupa) days were 10.65 ± 0.92 and 8.0 ± 0.82 respectively. Development at 32 0 e (11.32 ± 1.06) was faster than that at 20 0 e but it was slower than those reared at 26°e and 300 e (Fig 13). The development of the immature stages at 26°e was about half the time required for complete development at 20 0 e and all the larvae completed development to adults. University of Ghana http://ugspace.ug.edu.gh 79 -;;;- 10 >- ao z o 8 i= ~ 2 o ~,~t~i- -----~----_r------._----~------~------- 20 24 26 28 30 32 TEMPERATURE °C Fig." DURATiON OF LARVAL DEVELOPMENT OF ~: UJ TI G RI PE S AT DI FFE RENT CONSTANT TEMPERATURES University of Ghana http://ugspace.ug.edu.gh 80 4 · 0 3·8 3·6 3 ·4 3 ·2 1/1 :>. 0 0 3 ·0 z 2 · 8 0 I- 2· 6 ~ a::: ::> 2·4 0 2 · 2 -' ~ I- 2 · 0 z w ~ I ' 8 a.. 0 -w' '·6 >w ' · 4 0 ' · 2 ' · 0 I 0 20 24 26 28 30 32 TEMPERATURE °C Fig. 12 DURATION OF PUPAL DEVELOPMENT OF C. (L) T1GRIPES AT DIFFERENT CONSTANT T EMPERAT UR ES University of Ghana http://ugspace.ug.edu.gh 81 At 300e, the larval and pupal development was completed two days earlier than at 26°e, and all the larvae completed their development. survival of larvae reared at 32°e was 81.25% and lower than those reared at 26° and 300e but it was higher than those reared at 200e (Table 11). The duration of development of the immature stages was significantly (P< 0.05) longer at Further, there were no significant differences between the developmental durations of the larvae reared at 26°e, 30 0e and 32°e. The duration of the pupal stage was shortened from 3.83 ± 0.15 days to 2.33 ± 0.52 days when temperature was increased from 200 e to 32°e (Fig 12). The decrease was significant (P<0.05). Pupal developmental duration is thus different from that of the larvan, since the the shortest developmental time was recorded at 30oe, whereas that of the larvae was at 32°e. This difference may be attributed to the hard exoskeleton of the pupae which enables them wi thstand relatively higher temperatures than the larvae (Appendix Table 12). Blunck (1924) has defined the most suitable developmental temperature as the "optimum", and the optimum in turn as the temperature at which the greatest percentage of individuals accomplish their development within the shortest period of time. others have used the highest survival to the adult stage in the shortest time to determine the optimum temperature (Brust 1967, Trpis and Shemanchuk 1969, 1970). The optimum developmental temperature for ~. (L) tigripes in this study is University of Ghana http://ugspace.ug.edu.gh 82 30°C since it gave the shortest developmental time with the highest survival of the immature stages to the adult stage (Fig 13, Tables 10 and 11). 3.4.3 Effect of temperature on prey consumption Temperature has a significant effect (P< 0.05) on the daily consumption of mosquito larvae by Q.(L) tigripes. Considering the fact that predator larvae were fed with prey larvae of the same stage, more individual prey were consumed daily at ~Ooc than at lower or higher temperatures (Fig. 14 and Table 12) . The fact that the rearing temperature of 32°C slowed down the duration of larval development (Fig 11) suggests that the larvae were uncomfortable at that temperature and less active and therefore could not consume large numbers of prey larvae. University of Ghana http://ugspace.ug.edu.gh ,ion in days of larval and pupal development of g. ill. tiqrioes ,fferent constant temperatures 20°C 26°C 30°C 32°C 2.83 ± 0.75 1.16 ± 0.41 1.00 ± 0.00 1. 00 ± 0.00 (n = 16) (n = 16) (n = 16) (n = 16) 3.0 ± 0.00 1.83 ± 0.75 1.00 ± 0.00 1.50 ± 0.51 (n = 12) (n = 16) (n = 16) (n = 13) 4.5 ± 0.00 1.50 ± 0.52 1.00 ± 0.00 2.83 ± 0.75 (n = 10) (n = 16) (n = 16) (n = 13) 0) w 6.83 ± 0.76 3.33 ± 0.34 2.50 ± 0.58 8.99 ± 1.82 (n = 10) (n 16) (n 16) (n 13) 17.16 ± 2.06 7.82 ± 2.23 5.50 ± 0.52 8.99 ± 1.82 3.83 ± 0.15 2.83 ± 0.98 2.50 ± 0.52 2.33 ± 0.52 (n 10) (n 16) (n 16) (n = 13) ~I 20.93 ± 2.11 10.65 ± 3.21a 8.0 ± 1. 21a 11.32 ± 2.34a by the same letters are not significantly different (P > 0.05) University of Ghana http://ugspace.ug.edu.gh 84 The mean number of prey consumed per stadium by ~. (li) tigripes was however different from the number consumed in 24 hours at 0 the various temperatures. At 20 e and 32°e, the mean numbers of individual prey larvae consumed per stadium were 229.33 ± 25.30 and 184.00 ± 23.81 respectively (Table 13). The daily prey consumption was less at 20 0 e and 32°e than at 26°e and 300 e and the larval development was most rapid at 300 e (Tables 10 and 12). This suggests that at temperatures above and below the optimum, development was slower and since it took a longer period, the predator consumed more prey. This suggests that the impact of predation could be greater at the lower and higher temperatures. Similar observations were made by Trpis (1972) with Tx. brevipalpis (Theobald). 3.4.4. Effect of temperature on the size of the immature stages The relationship between the size (weight and length) of the final instar (4th-stage) larvae and the weight of pupae of Q. (li) tigripes reared at different constant temperatures are presented in Figure 15 and Table 14. As temperature was increased from 20 0 e to 32°e, the 4th-stage predator larvae and pupae became smaller (Fig. 15) although larvae reared at 32°e consumed more prey larvae than those reared at 26°e and 300 e (Table 13). As temperature increased the rate of daily prey consumption also increases while the duration of development University of Ghana http://ugspace.ug.edu.gh 85 22 III 20 >- c 0 18 z 0 16 l- e::( a:: 14 :::> 0 12 ...J e::( I- 10 z w ::2: 8 (L 0 ...J 6 W w> 4 0 2 0 ( ) I 0 20 24 26 28 30 32 TEMPERATURE °C Fig.13 TOTAL DURATION OF PRE-ADULT (Larva + pupa) D E V EL 0 P MEN T 0 Fe. ( L) T 1G RIP E SAT D IFFERE NT CONSTANTTEMPER ATU RES University of Ghana http://ugspace.ug.edu.gh 86 40 35 (f) a:: I 30 N"'" ....... 3 rd 0 w 25 ::1E ?-, :::> // \ (f) z 4 th 0 / \ u 20 / \ w / \ / A ./ a:: / ./ \ b - - --0 Length of Larvae 2 2 o o o 20 24 26 28 30 32 TEMPERATURE °C Fig.15 EFFECT OF DIFFERENT CONSTANT TEMPERATURES ON THE SIZE (LENGTH ANDWEIGHT) OF 4th- STAGE LARVA AND WEIGHT OF PUPA OF .f..ill TIGRIPES University of Ghana http://ugspace.ug.edu.gh J::7mP7rature on total prey consumea aUL' ~U~ t::o,-,u "' ... c". ..... ~ ... b.gr1pes 26°C 30°C 32°C .31 14.33 ± 4.80 18.36 ± 1.73 10.50 ± 0.84 i) (n 16) (n 16) (n 16) 52 29.50 ± 10.82 25.00 ± 0.83 21.50 ± 8.04 ) (n 16) (n 16) (n 13) 11 32.50 ± 12.68 30.37 ± 1.03 71.50 ± 19.63 ) (n 16) (n = 16) (n 13) .18 94.50 ± 16.63 86.50 ± 21.03 80.50 ± 11.17 (n 16) (n = 16) (n 13) 00 \0 1 3 0 170.83 ± 37.27 160.17 ± 21.61 184.00 ± 23.81 pf temperature on weight and length of larvae ~ht of pupae of ~ (~) tigripes 10°C 26°C 30°C 32°C 0.59 7.82 ± 0.43 6.83 ± 0.68 5.82 ± 0.60 10) (n = 16) (n = 16) (n = 13) 0.36 7.10 ± 0.37 6.73 ± 0.28 6.42 ± 0.32 :>.36 7.48 ± 0.40 6.28 ± 0.41 5.30 ± 0.14 LO) (n = 16) (n = 16) (n = 13) ---- ----- University of Ghana http://ugspace.ug.edu.gh 90 decreases. This suggests that more energy will be utilized to hasten development, and the rate of metabolism will also be expected to increase with temperature. All these factors may explain the reduction in the size of the larvae and the weight of the pupae with increasing rearing temperature. 3.4.5 Colonization in the laboratory Informatiun on the breeding of large numbers ot Q. (~) tigripes in the insectary is summarized in Appendix Table 13. Eggs brought from the field hatched within 24 hours and the percentage hatchability was 81.95% (n=703). Out of 568 adults which emerged, 265 were males and 303 were females ( a ratio of 1:1.14). Out of the 303 females, 102 (33.66%) took some blood meal and 13 (12.75%) were able to lay eggs, with a pre- oviposition period of 8.08 days (range 6-12 days) to lay the first batch of eggs (Appendix Table 14). The average number of eggs laid per females was 60.92 ± 32.55 (range 20-128) and a total of 792 eggs were obtained but none of them hatched into larvae. Figure 16 and Appendix Table 15 show the longevity curves of Q.(~) tigripes under laboratory conditions expressed as percentages surviving. 3.70% of the males were alive 56 days after emergence while 15.0% of the females were alive during the same period. All the males died by age 63 days while the females died 70 days after emergence. These results show that the females lived slightly longer than the males. All the mosquitoes were fed on 10% sugar solution. Comparison University of Ghana9 h1 ttp://ugspace.ug.edu.gh Fig. 16 SURVIVAL OF ADULT ..£.{JJ TIGRIPES IN LA 80 R ATORY University of Ghana http://ugspace.ug.edu.gh 92 could not be made with females fed solely on blood because the females usually took some sugar before taking blood meal. However the maximum survival time of females after blood meal intake was 41 days. 3.4.5.1 Rearing of the immature stages The eggs were whitish in colour when first laid but changed to yellowish - brown a few hours later. They were broadly oval in o~tline, but more pointed blunte at one end. The average lenght was 1.08 ± 0.13mm and 0.23 ± 0.46mm at the greatest breadth (n 250). They were larger than the normal eggs of Culex species encountered which averaged about 0.73mm x 0.15mm (Christophers 1945). The surface of the eggs appeared uniformly smooth except at the anterior and posterior poles, where they were stippled brown while the rest of the eggs looked yellowish-brown; this colour was quite distinct from those of other Culex eggs encountered in the field, which were greyish-brown. Cannibalism was one of the sources of loss during larval development and this occurred mostly during the first three days of development. This however, was reduced when the number of ~. (~) tigripes larvae per tray was reduced to 10 and the number of prey larvae was increased. There was no pupal loss due to cannibalism. Pupal mortality was only 0.52% and a further 1.90% loss occured due mostly to failure of the adults to emerge (Appendix Table 13). with the exception of the chicken, the adult female ~. (~) University of Ghana http://ugspace.ug.edu.gh 93 tigripes did not feed on any of the animals offered including human arms. This feeding behaviour confirms the observations made by some authors (Lewis 1947, Snow 1983, Snow and Boreham 1973) that ~. (.1) tigripes rarely feeds on man but rather prefers avian blood. No mating of the adults was observed in both the small and large cages. Lack of insemination was therefore thought to be the cause of the production of sterile eggs. A female is considered successfully inseminated only if her spermotheca contained spermatozoa (Lea and Evans 1972) . Spermothecae of thirty females were therefore dissected out in saline and crushed under cover slips and examined for living spermatozoa. None of the three spermothecae in each of the 30 females examined contained spermatozoa thus confirms the observation that the production of sterile eggs was due to the fact that the females were not inseminated. Adults of many Culex species of mosquitoes have failed to copulate in laboratory cultures (Horsfall and Taylor 1967, Christensen and Rowley 1978, Thompson 1948). The three favourable factors required for the development of the ovary in mosquitoes are full blood meal, fertilization and temperature (Gillett 1971, WHO 1975). The low number of blood fed females that laid eggs (12.75%) and the small number of eggs laid per female (20-128) may be due to insufficient intake of blood by the females. The number of eggs produced by mosquitoes have been found to be related to the size of blood meal taken by the female (Woke et al 1956, Colless and University of Ghana http://ugspace.ug.edu.gh 94 Chellapah 1960). The temperature and the relative humidity of the insectary ranged from 24 0 -28 0 and 68-75% respectively. These two factors can also affect the reproductive biology of the predator. 3.4.5.2 Artificial insemination The induced insemination of £. (L) tigripes carried out in this study was not successful. Observation of the behaviour of females when presented to males under the binocular microscope, showed that the females retracted their 8th abdominal segments thereby preventing union of the genitalia and thus, insemination. Observation of females of An. gambiae which were successfully inseminated using the same technique revealed that the receptive females responded to the males by extending their 8th abdominal segments thus facilitating union and transfer of spermatozoa. Firm union was never established between the sexes of £. (L) tigripes in any of the 62 attempts made. Several factors have been reported to affect the efficiency of copulation and the rate of insemination of mosquito. Other than the requirement for a vigorous mosquito stock, age of the mosquitoes (Lea and Edman 1972, Horsfall and Taylor 1967) and the type of anaesthesia used to immobilize females before induction of copulation (Fowler 1972) could influence them. All males and females used for the artificial copulation were at least 4 to 5 days old. The rate of both receptivity and the rate of insemination increases as the age of female £. guinguefasciatus increases University of Ghana http://ugspace.ug.edu.gh 95 (Lea and Evans 1972). Fowler (1972) reported that the rates of insemination of Aedes vexans (Meigen) were consistently high for females anaesthetized with nitrogen, intermediate for those anaesthetized with carbon dioxide and low for females anaesthetized with chloroform. He indicated that the efficiency of insemination during induced copulation is enhanced by the use of anaesthetic which allows rapid recovery of adults such as nitrogen. It is possible that the anaesthesia which were available and used in this study, chloroform and ether might have contributed to the unsuccessful insemination of the adult ~. (~) tigripes. 3.4.6 Development on non-living diets The results of the study of the developmental periods for ~. (~) tigripes larvae reared on the three different non-living diets, namely, Cerelac infant cereal, Dog biscuits and Milk casein are given in Table 15. Larvae maintained on diets of Cerelac, Dog biscuits and milk casein took an average of 18.47 ± 1.22, 17.71 ± 3.31 and 19.05 ± 1.80 days respectively to complete larval development, while those reared on living mosquito prey larvae, took a shorter period of 6.63 ± 0.50 days. The difference in the developmental duration of larvae reared on mosquito prey and those reared on the non-living diets is significant (P< 0.01). The mean weight of the 4th-instar larvae reared on mosquito prey larvae was 7.19 ± 0.96mg, and was much heavier than those reared on non-living diets which were 1.66. ± 0.20, 1.82 ± 0.27 and 2.78 ± 0.58mg University of Ghana http://ugspace.ug.edu.gh 96 for Cerelac, Dog biscuits and milk casein respectively. Duncan's multiple range test showed that there was no significant difference between the weights of the 4th-instar larvae reared on Cerelac and Dog biscuits, but all the other mean weights were significantly different from each other. Only one £. (~) tigripes larva which was reared on milk casein was able to pupate and emerge as adult (Table 15) after spending 17 and 3 days in the larval and pupal stages respectively. This pupa looked smaller and paler and weighed 5.33mg compared with those reared on mosquito preys. The male adult which emerged from it also looked smaller than normal and it died immediately after emergence. The minimum weight of the 4th-instar larvae which pupated when reared on mosquito prey larvae was 5.80mg, compared to the weight of the larva (2.78mg) which pupated upon feeding on the casein diet. The crit ical weight threshold necessary for pupation has been calculated for certain non-predatory insects, such as Manduca sexta (Nijhout 1975) and it has also been observed that prey availability may influence the weight at which predatory insects moult to the next instar (Beddington et al 1976). Based on these observations it may be suggested that failure of the 4th-instar £. (~) tigripes larvae to pupate was probably due to their low weights (Table 15). The 4th-instar larvae reared on the non-living diets looked paler in colour and the typical blackish-brown contents of the thorax and gut found in those reared on mosquito prey larvae were absent. University of Ghana http://ugspace.ug.edu.gh 97 Table 16 gives the survival of the predator larvae in the different instars when reared on different diets. The data indicates that in all cases, the percentage of survival of the larvae decreased as the larvae developed from the 1st-instar to the 4th-instar. This may be due to deficiencies in the diet or insufficient intake of food during their development. Mortality was highest in larvae reared on Cere lac followed by Dog biscuit and it was least in larvae reared on Milk casein. The mouthparts of Q. C!!) tigripes become more sclerotized as the larva grows and it would be expected that the larger larvae would find it difficult to filter feed effectively. The early instars may be able to filter feed to some extent because the mouthbrushes which are used for such feeding are not heavily sclerotized, also no movements of the mouthbrushes of the predators were observed even in the early instars as found in other mosquitoes which feed by filtering. ~ (lutzia) fuscanus larvae failed to grow and survive on either dog biscuits with vitamin complex, meat, dried fish or live Daphnia (Ikeshoji 1966). Jackson (1953) could not rear Q. (11) tigripes on dog biscuit, but she obtained two pupae, one of which took 7 days and the ensuing adult appeared normal. She did not rear the larvae from the 1st-instar as was done in this study. They were reared from the 2nd-instar collected from the field and it is not clear whether the larvae consumed some living preys before they were brought into the laboratory. Predatory larvae of Tx. rutilus rutilus. University of Ghana http://ugspace.ug.edu.gh 98 were successfully reared to adult stage on non-living diets of Tetramin, a tropical fish meal (Focks et al 1978). Development was, however, very slow and took 107.5 ± 19.8 days to complete development compared to 15.6 ± 1. 40 days when reared on Ae. aegypti larvae. The weights of the 4th-instars larvae and pupae were much smaller when they were reared on the non-living diet than on living mosquito prey. He observed that the extremely long developmental periods and the reduced size and weight of the 4th-instar larvae are undesirable aspects of using non-living diets to mass rear the larvae of Tx. rutilus rutilus, even though larval and adult survival and fecundity of that predator fed on non-living diet were not significalltly different from those fed on the prey diet. The results from this study indicates that non-living diets tested are not suitable for rearing .Q. (11) tigripes, nevertheless, a wider range of non-living diets should be tried for rearing this species until an appropriate diet is found. The results also gives an indication of the ability of the .Q. (11) tigripes larvae to survive over relatively long periods without preys. This may be important when the predator larvae are the first to appear in a small water body without preys. University of Ghana http://ugspace.ug.edu.gh !lopmental period (days) ot' .larvae ana J.arvaJ. We.L~lll. . ::; U.1. ~ (~ tigripes reared on different diets 1st 2nd 3rd 4th Total Pupa Larval weight instar instar instar ins tar (larvae) (mg) .c 2.72 ± 0.08 3.41 ± 0.98 5.61 ± 2.92 7.74 ± 1.15 18.47 ± 1.22ab 0 1. 66 ± 0.20d en = 29) (n = 29) (n = 28) (n = 19) (n = 19) (n = 19) I 2.71 ± 0.72 3.95 ± 1. 50 5.15 ± 1.90 6.24 ± 1.56 17.71 ± 3.31a 0 1.82 ± 0.27d t (n = 22) (n = 21) (n = 17) (n = 17) (n = 17) (n = 17) I 2.69 ± 0.47 3.36 ± 0.76 4.08 ± 0.86 8.86 ± 1.80 19.05 ± 1.80b 3.0 2.78 ± 0.58 n (n = 25) (n = 25) (n = 25) (n = 21) (n = 21) ( n = 1) (n = 21) Mosquito 1.32 ± 0.48 1.21 ± 0.42 1.15 ± 0.37 2.95 ± 0.40 6.63 ± 0.50 2.16 ± 0.37 7.19 ± 0.96 prey (n = 10) (n = 19) (n = 19) (n = 19) (n = 19) (n = 19) (n = 19) Mean developmental days and mean larval weights followed by the same \0 \0 letter are not significantly different based on Duncans Multiple range test (P<0.05) University of Ghana http://ugspace.ug.edu.gh 100 Table 16 Survival of ~. ilL. tigripes larvae reared on different diets 1st inst ar 2nd instar 3rd instar 4th inst ar pupae Diet No. of No. % No. % No. % No. % Larvae Cerelac 29 29 100 29 100 27 93.1 19 65.52 a a Dog biscuit 22 22 100 22 100 21 95.45 17 77.27 a a Milk casein 26 26 100 25 95.15 25 96.15 21 80.77 1 4.76 Mosquito prey 19 19 100 19 100 19 100 19 100 19 100 University of Ghana http://ugspace.ug.edu.gh 101 Chapter 4.0 PREDATORY BEHAVIOUR 4.1 Introduction A study of the predatory behaviour would help to understand the part it plays in the control of mosquitoes in their natural habitats. The first attempts to analyse the mechanism of predation in the field (eg. Tinbergen 1960) showed the natural situation to be exceedingly complex. It proved to be extremely difficult to get a satisfactory quantitative answer to such fundamental questions like: which factors determine how many individuals of each of the various prey types present are destroyed, and how these factors operate. In this study, the approach followed was done with the idea that the intricate processes which occur under natural conditions are composed of simple elements; that these elements can be identified, and their interactions studied in the laboratory; and that the results of such laboratory work will give us useful information which may be applicable to field situation. The laboratory situation however well simulated, is not the same as what actually goes on in the field. Thus in the laboratory, it is possible to restrict the number of interacting predators and prey, and the number of environmental variables that may affect the predation process, so that precise hypotheses can be tested. University of Ghana http://ugspace.ug.edu.gh 102 4.2. Materials and Methods 4.2.1. Prey capture and Feeding Habits The typical mode of feeding employed by culicine mosquito larvae is to filter out of the surrounding medium, food particles brought in by currents set up by their pre-oral mouthbrusr.·es. The mouth- parts of culicine mosquito larvae are of the generalized type, having mouthbrushes with fine lamellae, well developed mandibles and maxillae and a central mentum composed of the fused segments of a degenerate labium (Surtees 1959) . Throughout the various genera, specializations have taken place so that the basic filter-feeding method has given way to browsing and predation, with changes in the mandibles and maxillae. The former has become larger and more important whilst the latter become smaller. Concurrent with these changes, the mouthbrushes have become shorter, stronger and usually serrated. The objectives of the present investigation are (1) to elucidate the methods of attack, prey capture and the feeding habits of £. (~) tigripes. (2) to study the morphological modifications associated with these activities. The methods of attack, prey capture and feeding were first observed with the naked eyes and then under a bionocular microscope, after the larvae of £. (~) tigripes and other mosquito larvae (An. gambiae, £. guinguefasciatus and Ae. aegypti) were placed together in a transparent bowl. £.(~) University of Ghana http://ugspace.ug.edu.gh 103 tigripes larvae which had captured prey were transferred into petri-dishes containing water to examine the feeding closely under the high power of the binocular microscope. The part of prey larvae (ie neck, thorax, abdomen, tail and siphon) seized during attack by the predator was also recorded. For the detailed examination of the morphology of the mouth parts, some of the ~. (L) tigripes larvae were quickly pipetted into hot water to be killed, and were mounted in Berlese's fluid on glass slides. 4.2.2. Effect of predator stage. prey stage and prey density on the predation rate Many aquatic predator species have been qualitatively evaluated for their possible use in mosquito control programmes (Sailer and Lienk 1954, Prakash and Ponniah 1978, Muspratt 1951, Haddow 1942, Jackson 1953, Rajasekaran and Chowdiah 1972). However, none of these workers studied the effectiveness of the predators at different stages of predator-prey development and densities. The objective of this investigation is to determine the effect of predator stage, prey stage and prey density on the rate of predation of ~. (L) tigripes. The four larval stages of ~·(L) tigripes were tested against ~.guinguefasciatus larvae of identical stages at three different densities. One predator larva was exposed to 10, 20, and 40 Culex prey larvae in 30ml of water in an 80rnl capacity plastic bowl (6.2 x 3cm) for a period of 24 hours. These were designated densities D1, University of Ghana http://ugspace.ug.edu.gh 104 02, and 03 respectively. Prey larvae which were consumed were replaced with larvae of the same developmental stage. The total number of combination of predator stage (4), prey stage (4) and prey density (3) was 48. All observations on the rate o of predation were conducted at room temperature (25 ± 2.0 C). 4.2.3. Effect of water volume on predation rate water volume and container size are known to influence the activities of some predators. The larvivorous fish Gambusia affinis kills more ~ guinguefasciatus larvae with increasing volume of water in the aquarium up to 1500ml (Reddy and Pandian 1973). The predatory activity of the final or fourth instar of Tx. rutilus rutilus was significantly affected by the container size. The rate of prey consumption varied inversely with container size (Padgett and Focks 1980), but the predatory capacity of Culex (lutzia) raptor was not influenced by differences in the volume of aquarium water ranging from 150 to 700mls (Prakash and Ponniah 1978). The objective of this study is to determine the effect of different water volumes on the predation rate of .Q. (l!) tigripes. Round plastic bowls (9.5 x 5.5cm) were filled with 30, 60, 90, 120, 150mls of water. Into each of these bowls, one 4th-stage .Q. (l!) tigripes larva which had previously been deprived of food for 24 hours was also introduced. The predator was deprived of food for 24 hours to minimize differences in individual hunger levels (Nakamura 1977). The number of Culex larvae killed or consumed in 24 hours was University of Ghana http://ugspace.ug.edu.gh 105 recorded for each volume of water. 4.2.4. Feeding Preference of C. (Ll tigripes The feeding preference of a mosquito predator such as Q. (L) tigripes ~ay be affected by several factors, including the prey size, mobility, defense or avoidance behaviour, palatability, abundance and the extent to which the predator and prey habitats overlap (Ellis and Borden 1970). Some of the characteristics of a prey which will elicit an attack response of a predator, and presumably, those preys possessing those characteristics which tend to elicit an attack would be most vulnerable to predation. The preferred prey will therefore be determined by the degree to which the predator responds to some of the factors mentioned above. The objectives of this study are: (1) to determine the effect of prey mobility on the feeding preference of Q. (L) tigripes. (2) to determine the common species of mosquito preys preferred by Q. (L) tigripes. (3) to determine the feeding preference of Q. (L) tigripes with respect to prey stages. 4.2.4.1 Spontaneous and Induced Movements of C.(Ll tigripes and mosquito preys Healthy 4th-stage larvae of Q. (L) tigripes, An.gambiae, ~. guinguefasciatus and Ae. aegypti were selected and put individually into plastic bowls (6.2 x 3cm) filled with 30ml of water. These bowls were watched in rotation and the total University of Ghana http://ugspace.ug.edu.gh 106 time in seconds in which each larva moved spontaneously in one minute was recorded. Fifty replications were made for each species. Fifty healthy 4th stage larvae of the four species were again selected for induced movement studies. Individual larvae were put in separate bowls with the same design as above. Larvae were individually stimulated by tapping the edge of each bowl once with a glass rod, and the time in seconds during which movement of the larva occurred after the stimulatic.'n was recorded. Twenty five larvae of each of the four species were subjected to this type of stimulation. The other twenty larvae were stimulated individually by touching the abdomen of the larvae once with a glass rod, and the time in seconds during which the larvae continued to move was recorded. stop watches were used to time the movements of the larvae within the tested period. In all these trials the larvae were allowed to settle down and remain motionless before any stimulation was given. All external stimulation such as vibrations were eliminated by making sure that no movement occurred during the trials. 4.2.4.2 Feeding Preference by Prey Species Fixed numbers of 4th-stage larvae (10 and 20) of An. gambiae, ~. guinguefasciatus and Ae. aegypti were selected and put in individual plastic bowls (6.2 x 3cm) filled with 30ml of water. One 4th-stage larva of ~. (1) tigripes was introduced into each of the bowls. In all cases, the ~. (1) tigripes larvae had previously been starved for 24 hours. The number University of Ghana http://ugspace.ug.edu.gh 107 of each species of larva killed or consumed in 24 hours were recorded. In another experiment, three of the same bowls were set up, each containing one 4th-stage larva of An. gambiae, g. guinguefasciatus and Ae. aegypti. One 4th-stage g. (~) tigripes was introduced into each bowl and the bowls were examined every 30 minutes. The order in which the prey were consumed was recorded. Forty-two replicates of this experiment were performed. Similar feeding preference tests were carried out on chironomid and £...,.. quinquefasciatus larvae because chironomid larvae were observed to be consumed by Q. (~) tigripes in its natural habitats. One 4th-stage g. (~) tigripes larva treated as above was exposed to 10 and 20 each of chironomid and 4th stage £...,.. guinquefasciatus larvae. The sizes of chironomid larvae were about the same as the Culex. Trials were made using tap and field water because chironomid larvae were observed to form tubes around themselves in their natural habitats. The numbers of each species consumed or killed in 24 hours were recorded. 4.2.4.3 Feeding Preference by Mosquito Prey Stages One 4th-stage larva of Q. (~) tigripes was exposed to equal numbers (10 and 20) of 4th-stage larvae and pupae of £...,.. quinquefasciatus in a bowl (6.2 x 3cm) containing 30ml of water. T~e number of larvae and pupae consumed or killed in 24 hours was recorded. Similar experiments were conducted by introducing one 4th-stage Q. (~) tigripes larva into bowls University of Ghana http://ugspace.ug.edu.gh 108 containing 1st and 3rd stage Culex larvae, and 2nd and 4th-stage Culex larvae. The number of each larval stage consumed or killed in 24 hours was recorded. 4.2.5 Wasteful Killing It has been observed by many authors that some predatory organisms killed more prey than they could possibly consume. This behaviour is not uncommon among larvivorous predators; for instance the larvivorous mosquito Tx. brevipalpis kills 26% more larvae than it can consume (Trpis 1972) and a similar behaviour has been found in other species of Toxorhynchites (Muspratt 1951, Corbet 1963, Corbet and Griffiths 1963). Notonecta undulata (Say) is also known for its characteristic overkilling of Culex larvae (Toth and Chew 1972, Ellis and Borden 1970). Q. (lutzia) raptor was also found to exhibit this killing behaviour (Prakash and Ponniah 1978). Q. (l!) tigripes has not been studied with regard to this killing behaviour. This phenomenon has been referred to as wasteful killing or compulsive killing behaviour. The former usually refers to cases of partial consumption of prey leaving parts of the pr£y uneaten (Johnson et al 1975) while the latter is killing without consuming any part of the prey (Trimble and Smith 1978). This behaviour is similar to the normal predatory behaviour except the dead larva is not consumed (Crans and Slaff 1977), and factors which affect this behaviour have been found to be similar to those which affect normal predation. High temperature was found to increase the University of Ghana http://ugspace.ug.edu.gh 109 daily number of prey killed but not consumed by both Tx. brevipalpis (Trpis 1972) and Tx. rutilus spetentrionalis from two geographical regions in North America (Trimble and Smith 1978). Both the number of prey killed but not eaten and the length of the killing phase in Tx. brevipalpis increased and were linearly correlated with increase in prey density (Lounibos 1979). Russo (1985) reported that surplus killing of five species of Toxorhynchites never began before the species achieved the minimal larval weight required for pupation. However, Lounibos (1979) confirmed (and it was also suggested by Corbet and Griffiths 1963) that the killing behaviour was not a necessary prerequisite for metamorphosis, and that the weight threshold of Tx. brevipalpis for pupation was substantially lower than that for killing behaviour. Compulsive killing behaviour was reported to be displayed by all larval instars of Tx. splendens both before and after each moult, but it was most pronounced just before pupation (Chan 1968). It was generally found to be most intense after the pre-pupa has formed and feeding is no longer possible (Crans and Slaff 1977, Muspratt 1951, Corbet 1963, Corbet and Griffiths 1963, Furumizo and Rudnick 1978). The ecological significance of this behaviour is unclear but it has been suggested that pre-pupal killing is a significant intraspecific interference mechanism with the potential for stabilizing the predatory-prey interaction (Hassell and May 1973). The most generally accepted theory to explain this University of Ghana http://ugspace.ug.edu.gh 110 phenomenon is that this habit of pre-pupal killing is probably a protective mechanism to ensure the continuance of the species, since the pupae are more vulnerable to attack and injury by other predaceous larvae, and also because they cannot defend themselves when attacked. The objectives of this study therefore are (1) to determine if g.(~) tigripes exhibits wasteful killing behaviour. (2) and the conditions that may affect this behaviour. Third-stage .Q. (~) tigripes larvae were reared individually until they reached the 4th-stage. Each 4th-stage larva was provided daily with either 40, 60, or 100 Culex larvae of equivalent stage as the predator. This feeding procedure was followed until the larvae pupated. The larvae were monitored for the onset of killing behaviour by checking daily, the number of prey larvae consumed and for the presence of prey larvae killed but not eaten. Normally a .Q. (~) tigripes larva will take in the entire prey larva captured and will discard the head capsule and sometimes the siphon in larger preys. In this study, a prey larva that was more than half eaten was regarded as eaten whilst the one that was less than half eaten was recorded as not eaten (ie it died as a result of wasteful killing). Dead prey larvae which showed signs of injuries on the body presumably caused by the mandibles of the predator, were also recorded as killed but not eaten. University of Ghana http://ugspace.ug.edu.gh 111 4.2.6. Functional response of C. (Ll tigripes to prey density Of the many aspects of predator behaviour relevant to predator-prey interactions, the functional response of the predator to changes in prey density is one of the most important. Information on functional response is essential for a clear understanding of the predator-prey interaction. The total number of prey destroyed by predators is the product of the number killed per predator and the number of predators that are present (Holling 1966). Solomon (1949) first proposed the terms to describe this two-fold nature of the predation process - the response of a predator to changes in prey density. He applied the term functional response to changes in the number of prey consumed by individual predators, and the term numerical response to changes in the density of predators. Functional response of predators has been classified into three categories by Holling (1959a) as follows: type 1 where there is a linear rise to a maximum in the number of prey eaten per predator as prey density increases; type 11 - where the response rises at a decreasing rate towards a maximum value, and type III where the response :s sigmoid with an initially increasing slope and again approaching an upper asymptote. Type III responses are thought to be more characteristic of vertebrate predators that can learn to concentrate on a prey as it becomes abundant (Holling 1965). Type II responses are generally associated University of Ghana http://ugspace.ug.edu.gh 112 with invertebrate predators, and has attracted the most theoretical attention. The best known description being the disc equation of Holling (1959b); Na aNT P 1 + a Th.N. Where Na the number of prey killed (attacked) p the number of predators N the density of prey a = a constant, the attack rate of the predators or the predators rate of successful search Th a constant, the handling time, including the time spent pursuing, subduing and digesting each prey T total time predator and prey are exposed to each other. A number of factors determine the characteristics of these functional response curves. The most important ones have been identified to be: (a) the time predator and prey are exposed to each other (b) the rate of searching of the predator (which influences the magnitude and character of the functional response), and (c) the time spent in handling prey (which affects the response by decreasing the time available for active search. University of Ghana http://ugspace.ug.edu.gh 113 All these are basic factors incorporated in the disc equation to obtain the basic functional response equation. other subsidiary factors such as hunger and characteristics of the environment of the predator and of the prey may affect the basic responses by changing their magnitude rather than their form. The aim of the present study is to determine the predation rate of g. (1) tigripes larvae in response to chang ing mosquito prey larvae densities. Fourth stage larvae of g. (1) tigripes which have been deprived of food for 24 hours were exposed to various numbers of g. guinguefasciatus larvae in plastic bowls (6.2 x 3cm) filled with 30mls of water. The prey densities were 10, 20, 40, 60, 80, 100 and 120 larvae per bowl of 30 ml of water. These densities were kept constant by replacing larvae which have been destroyed by the predator, and the number of prey larvae destroyed in 24 hours was recorded. 4.2.6.1 Effect of Handling time on Predatory Activity One of the basic factors affecting functional response of predator to prey density is the handl ing time. It has subcomponents which include (1) the time spent orientating to, pursuing and subduing prey . (2) the t~me spent eating prey and (3) the time spent in a digestive pause during which the predator is not hungry enough to eat further prey. University of Ghana http://ugspace.ug.edu.gh 114 They are all time consuming activities and once initiated they preclude further search (Holling 1966). The objectives of this investigation are: (1) to determine the changes in handling time between different stages of ~.(~) tigripes larvae attacking the same stage of prey larvae. (2) to determine the changes in handling time within the same stage of ~. (~) tigripes larvae attacking different stages of prey. (3) to determine changes in handling time of the predator with successive feeding on the same size of prey larvae. Freshly moulted first to fourth instar larvae of ~.(~) tigripes were each provided with 20 of 2nd stage Culex larvae. Handling time was recorded with a stop watch starting from the time a ~.(~) tigripes larva seized a prey in its mouth -parts to the time it disposes of the unwanted part (ie the head capsule) from the mouth. In another experiment individual 4th-stage larvae of the predator were provided with 20 of either 1st, 2nd, 3rd or 4th stage ~ guinguefasciatus larvae in a bowl, and the handling time on each prey consumed was recorded. Lastly, individual 4th-stage larvae of the predator were provided with 20 of 2nd stage larvae of Culex. Handling times were recorded as before for the first five prey larvae consumed. University of Ghana http://ugspace.ug.edu.gh 115 All ~.(~) tigripes larvae used in these tests were starved for 24 hours prior to each test. 4.2.6.2 Effect of food deprivation on predation rate The length of time the predator goes without food may exert an influence on its subsequent food intake. It is assumed that food deprivation time influences behaviour because of increasing hunger (Holling 1966). Hunger has been defined as the emptiness of the gut and it is the internal drive motivating all components of feeding behaviour in Hollings predation model (Holling 1966). A number of workers (Beukema 1968, Miner 1955, Ware 1972) have used the amount of food consumed by fish in relation to deprivation as an assessment of hunger. With a view to study the effect of hunger on the predatory behaviour of ~.(~) tigripes, the length of time of food deprivation is used as the criterion for assessing hunger. The objectives of the present study are to (1) determine the effect of different lengths of food deprivation on predation rate. (2) to determine the effect of different lengths of food deprivation time on the handling duration of prey. Fourth stage larvae were initially exposed to abundant supply of 4th stage Culex mosquito larvae, and ther6after were deprived of food for 6, 12, 24, 36, and 48 hours, and subsequently exposed to a constant number University of Ghana http://ugspace.ug.edu.gh 116 of 40 4th-stage ~ guinguefasciatus larvae for 24 hours. The number of prey consumed in 24 hours was recorded. In another series of experiments 4th-stage £. (L) tigripes larvae which had been deprived of food for various hours as above were exposed individually to 40 ~ guinguefasciatus larvae and the time taken to consume the first five preys was recorded. 4.2.7. Cannibalism 4.2.7.1 Effect of predator stages on cannibalism Two experiments were conducted, the first was to determine the degree of cannibal ism among the larval stages of the predator. Twenty each of 1st, 2nd, 3rd, and 4th-stage larvae of £. (L) tigripes were put into separate plastic bowls (6.2 x 3cm) containing 30ml of water. The number of larvae killed or consumed in 24 hours from each bowl was recorded. The second experiment was to determine the effect of different stages of the predator on cannibalism. One 4th-stage of the predatory larva was introduced into separate bowls containing 20 of either 1st, 2nd, 3rd, 4th stage predatory larvae or pupae. The number of various stages of the £.(L) tigripes killed or consumed in 24 hours was recorded. 4.2.7.2. Effect of crowding on cannibalism Four plastic bowls (6.2 x 3cm) were filled with 15ml, 30ml, 60ml and 120mls water, and then 15 of 4th-stage larvae of the predator were introduced into each of the bowls. The number of larvae killed or consumed in 24 hours from each bowl was University of Ghana http://ugspace.ug.edu.gh 117 recorded. The different larval densities (ie 1, 0.5, 0.25 and 0.125 larvae per millilitre of water) were supposed to simulate different levels of crowding. 4.2.7.3 Effect of presence of prey on cannibalism One 4th stage ~.(~) tigripes larva was exposed to 10 larvae each of 2nd stage larvae of ~.(~) tigripes and ~ guinguefasciatus in a bowl. In another set-up one 4th-stage predatory larva was also exposed to 10 each of 2nd stage larvae of the predator and 4th-stage larvae of ~ guinguefasciatus. The number of larvae of different stages of the two species consumed in 24 hours was recorded. These results were to be compared with the results obtained after ten of 2nd-stage larvae of the predator have been exposed to one 4th-stage larva of the predator in a bowl for 24 hours. University of Ghana http://ugspace.ug.edu.gh 118 4.3 Results and Discussion 4.3.1. Prey capture and feeding habits The larvae of ~.(L) tigripes are generally bigger than the larvae of an equivalent stage of almost all the mosquito larvae on which they prey. The mandibles are very large with strongly sclerotized and pointed claws (Plate 4). Associated with the claws are large stiff spines which also aid in grasping the prey, particularly those posterior to the main claws. The mouth-brushes are thick and consist of several lamellae which are also sclerotized with the apical portions containing numerous sharp teeth arranged in rows (Plate 5). The mouth-brushes of ~.(L) tigripes have been described as hairs cemented together to form prehensile fangs with which the prey is seized and held while it is consumed (MacGregor 1927). The mentum is well developed with only nine large and pointed teeth. The maxillae are reduced in size but have strong setae attached to them. The antennae and the other spines on the head are also greatly reduced in size. The larva of ~. (L) tigripes usually stays motionless at the surface of water and waits until a prey comes within a striking distance, and then with a very rapid darting movement, the prey larva is captured. This sudden dart made by ~. (L) tigripes has been described by Hopkins (1936) as reminiscent of a snake and as resembling the ferocity of a crocodile by MacGregor (1927). Hopkins (1936) reported that University of Ghana http://ugspace.ug.edu.gh 119 prey larvae were seized only when they came into contact with the mouth-brushes of Q.(L) tigripes. This certainly occurs but observation in this study showed that more often, prey larvae are seized when they are within a striking distance of the predator, without necessarily coming into contact first with the mouth-parts. sometimes Q. (L) tigripes larva can even bend its body round to seize a prey larva moving almost behind it. The sight of the predator, especially the 4th-instars may playa role in its predatory activity. For instance, when the predator is either resting or obtaining its air supply at the surface, and there is a prey larva also resting near by, the predator will move its head towards it and sometimes align itself in that direction, but prey capture only occurs if the prey movement carries it towards the predator. Haddow (1942) stated that Q. (L) tigripes larvae almost always seize large or small anopheline or culicine larvae by the tail. Observations made" in this study showed that Q. (L) tigripes often seizes prey by the tail although it can seize its prey by any part of the body depending on the orientation of the prey larva at the time of attack. The observations on this show that 8% of the prey were seized by the thorax 2% by the siphon, 20% by the abdomen, 26% by the neck region and 44% by the tail. Careful observation showed that most of the prey larvae normally swam backwards and this frequently brought them either in contact or within striking distance of the predator with the tail end first. University of Ghana http://ugspace.ug.edu.gh 120 Plate 4. Ventral side of the head of 4th-stage larva of ~. (.b) tigripes showing mouth brushes and a well developed mandible with a sharp claw (arrowed) (magnification X 250) University of Ghana http://ugspace.ug.edu.gh 121 Plate 5. Part of the mouth brushes of ~.(~) tigripes showing numerous fine teeth on the lamellae University of Ghana http://ugspace.ug.edu.gh 122 Q. (l!) tigripes always seized and devoured pupae of mosquitoes from the tail because it is softer compared to the heavily sclerotized cuticle of the cephalothorax. In general however, Q. (l!) tigripes seizes its prey from the lateral side. Haddow (1942) reported that the specialized mouth-brushes of Q.(L) tigripes appear to penetrate the soft abdominal integument of the prey dlmost immediately. In this study, it was rather observed that in all instances the predator used only the mandibles and not the mouth-brushes to seize and hold prey. The sharp teeth of the mandibles penetrate the body of the prey and become pointed downwards with the tips directed towards the oral cavity while the mouth-brushes are still extended outwards from the head. The mandibles are therefore the main prehensile organs in Q. (L) tigripes. Surtees (1959) has, however, suggested that both the mouth-brushes and the stiff spines at the posterior end of the mandibles aid in grasping prey as reported earlier in this section. Observations on another predatory mosquito, Megarhinus septentrional is now called Tx. septentrional is by Breland (1949), revealed that the proper functioning of the mandibles was dependent in some way upon the mouth-brushes. The methods by which Q. (L) tigripes catches prey depends primarily on the location of the prey in the water i.e. at the water surface o~ at the bottom of the container. Q. (L) tigripes larva normally stay at the water surface but when it moves to the bottON after it has been disturbed, it stays there for University of Ghana http://ugspace.ug.edu.gh 123 sometime before it surfaces again. Thus in addition to catching prey at the surface catches prey larvae which dive to the bottom after disturbance at the water surface and also browse on submerged food particularly Aedes species. tigripes was not observed to attack or seize a prey within the water column. After capturing a prey ~.(~) tigripes often changes its normal posture of lying almost parallel to the water surface i.e. anopheline-like, and assumes the normal Culex posture, that is it slants at about 45° to the water surface. This may be due to the extra weight added to the anterior end of the predator by the captured prey. However, in this posture, the prey larva or pupa which is held firmly in the mouth parts, will become fully submerged in the water and is denied contact with the water surface leading to accelerated death by suffocation. The seized prey is often held firmly in the mandibles and the predator either stays at one place or it drifts along in the water if the prey is large and struggles to free itself. Mosquito larvae caught by ~. (~) tigripes cannot usually free themselves but captured mosquito pupae frequently struggle and break free but they may suffer physical damage caused by the mandibles of the predator. ~ (~) tigripes larvae do not usually wait for their victims to die before they consume them. Slow movements of the victims often continue after feeding has began; sometimes till almost the entire abdomen has been consumed. MacGregor (1927) observed that larvae captured by ~. (~) tigripes, usually University of Ghana http://ugspace.ug.edu.gh 124 struggles violently for a few seconds, and are slowly consumed alive until the devouring mandibles finally crush the large nerve ganglia in the thorax. ~. (L) tigripes usually feed on its prey in such a way that the thorax is eaten last while the head capsule is either partially eaten or completely discarded depending on the size of the prey relative to the predator. First instars whose head capsules are not fully sclerotized are usually consumed whole, without discarding the head capsule. It was observed in the present study that most 4th-stage ~.(L) tigripes larvae usually defaecate when they start consuming captured prey. This behaviour was probably to empty the gut to make way for the new food. This behaviour is not peculiar to ~.(L) tigripes because a fully satiated dragonfly nymph (Mesogomphus lineatus) exposed to a constant supply of mosquito larvae did not attack until such time that the nymph defaecated. Apparently in the dragonfly, the defaecation facilitated the transfer of a part of chyme from the stomach to the intestine, so the stomach or intestine evacuation appeared to control the return of appetite (Mathavan 1976). stomach or intestinal evacuation has also been shown to play a maj or role in returning appetite in fishes (Windell 1967, Brett and Higgs 1970, Pandian 1967). University of Ghana http://ugspace.ug.edu.gh 125 4.3.2 Effect of predator stage, prey stage and prey density on predation rate The results of the mean predation rates of the various combinations of predator stage, prey stage and prey density are given in Appendix Table 18 and illustrated in Figures 17 and 18. The first and second stages of the predator consumed more prey larvae of their own stage and the mean numbers consumed daily increased as the prey density also increased. The mean number of prey consumed at prey density 03 when 2nd-stage ~. (L) tigripes fed on 3rd-stage prey was an exception; less prey larvae were consumed at density 03 than 02 (Fig.17). Both the 1st and 2nd-stage ~. (L) tigripes were not very effective at feeding on larger stages (3rd and 4th) of the prey. The predation rate decreased at almost all densities as the prey-stage increased. The 3rd-stage ~. (L) tigripes is capable of consuming large numbers of prey of all stages except the 4th-stage prey larvae (Fig.18). The 4th-stage larvae of ~. (1) tigripes consumed relatively large numbers of the prey of all stages, including prey of its own stage (4th). The trend of predation rates shown by the various stages of the predator is as expected because more prey larvae University of Ghana http://ugspace.ug.edu.gh 127 40 0 (Q) :3 rd Stage predator IIJ ..J ..J ~ 30 IIJ a: Vl Q: 0 :l: ' a: z z 0 :r: V N "0 W ...J ...J ~ w 10 c:{ > It: c:{ ...J li... 0 It: W CD :::E :::J Z 5 z c:{ w :::E OL-__- ,.-__- ..-__- ,-__- --,... o 4 PREY LARVAL STAGES Fig. 20 TH E NUMBERS OF DIFFERENT STAGES OF ~. FATIGANS CONSUMED BY C. (L) TI GR IPES University of Ghana http://ugspace.ug.edu.gh 132 "0:: : 30 o ~ o W 0::: 25 a.. "o w ~ :J 20 (/)(/) Zo::: O:J U o wI 0.05) as the water volume was increased from 30 through 60mls, to 90mls but beyond this the difference became significant. This result is not surprising because it is expected that as water volume increases, the density of the preys will decrease so the chances of prey larvae corning into contact or moving closer to the predator will be decreased and therefore the number captured will be expected to decrease. Padgett and Focks (1980) found that the size of water container affected the rate of predation of Tx. rutilus rutilus. The rate of predation varied inversely with container size and directly with prey density, and they suggested that the difference may be from the variable ratios of container surface area to predator. Water volume affected the predation rate of £. (L) tigripes in the same way as container sizes in Tx. rutilus rutilus ; it varied inversely with predation rate. University of Ghana http://ugspace.ug.edu.gh 134 Table 17 Effect of water volume on predation rate of £. ~. tigripes Water volume No. of prey No. Consumed No. of F ratio (mls) offered per or killed in trials predator 24hrs 30 40 37.2 ± 3.01b 20 18.33 P< 0.01 60 40 35.0 ± 3.74 ab 20 v It) 90 40. 31.1 ± 3.25 a 20 120 40 26.8 ± 4.94 20 150 40 22.4 ± 6.38 20 ----- -- L_ . _ ____ -----~ Means followed by the same letter are not significantly different (Duncans Multiple range P > 0.05) University of Ghana http://ugspace.ug.edu.gh 135 4.3.4. Effect of food deprivation on predation rate The results of the rate of predation of ~. (~) tigripes after different periods of food deprivation are shown in Figure 22 and the results of the time taken to consume a fixed number of five prey larvae after different periods of food deprivation are shown in Figure 23. The predation rate increased as the length of food deprivation increased and reached a peak after 24 hours of food deprivation, then started to decline gradually with further increase in the length of food deprivation (Figure 22). The difference in the mean values were, however, not significantly different (p>o. 05) . The time taken by ~.(~) tigripes to consume five prey larvae decreased as the length of food deprivation was increased (Fig.23), and the difference was highly significant (Pw- 35 0:: Cl. 34 l.L. 0 33 o · z 32 z <.( w 31 ~ 30 0 6 12 18 24 30 36 40 46 DURATION OF FOOD DEPRlVATION ( hou rs ) Fig.22 THE INFLUENCE OF THE DURATIONOF FOOD DEPRIVATION ON THE PREDATION RATES OF £ . lh.) T I G RIP E S University of Ghana http://ugspace.ug.edu.gh 137 400 VI c E 360 >- W a:: 340 a. 10 W 280 ~ ::J CI) Z 0 240 u 0 ~ 200 w ~ ~ 160 Z c:( w ~ 120 100 0 6 12 18 24 30 36 42 48 DURATION OF FOOD DEPR IVATION (hrs) Fig. 23 EFFECT OF THE DURATION OF FOOD DEPRIVATION OFC.(L)T/GRfPES ON THE PREDATORY ACTIVITY University of Ghana http://ugspace.ug.edu.gh 138 and the average daily prey consumption almost levelled off indicating stability in hunger. The pattern of prey consumption therefore did not have any significant effect on the total prey larvae consumed in a day, after the return of maximum appetite. The length of food deprivation (which gives an indication of the hunger level) affects the predatory activities of some predators such as N.undulata (Ellis and Borden 1970). The predator searches more actively for its prey after different lengths of food deprivation and when preys are captured, less time is used to handle individual prey with the resultant increase in daily numbers of prey destroyed as the length of deprivation increases. University of Ghana http://ugspace.ug.edu.gh 139 Table 18 Effect of varying periods of starvation of ~. (iJ tigripes on time it takes to consume prey (handling time) and on number of prey consumed per day Hours of Handling No. of prey starvation Time (mins) consumed 6 324 ± 55.82a 34.2 ± 2.78 (n = 20) (n = 20) 12 300 ± 54.37a 35.2 ± 3.68 (n = 20) (n = 20) 24 168 ± 25.00b 37.6 ± 1. 26 (n = 20) (n = 20) 36 158 ± 23.94b 36.6 ± 2.46 (n = 20) (n = 20) 48 138 ± 19.89b 35.8 ± 2.25 (n = 20) (n = 20) F ratio 49.68 P<0.001 2.49 P>0.05 NS) NS = not significant (P > 0.05) M7an followed by the same letter are not significantly d1fferent (P > 0.05 Duncan's multiple test) University of Ghana http://ugspace.ug.edu.gh 140 4.3.5. Feeding preference 4.3.5.1. Feeding preference by prey species The results of the prey species preference tests of £. (l!) tigripes are presented in Table 19. Mosquitoes which are the most common species associated with the predator in its natural breeding places were used against the most voracious predator stage (4th). The results indicate that £. (1) tigripes larvae showed a strong preference for Ae. aegypti as compared to An. gambiae and £. guinguefasciatus particularly at the higher prey density of 20 larvae per bowl (p<.OOl). At the lower prey density of 10 larvae/bowl, more Ae. aegypti larvae were again consumed daily than either £. guinguefasciatus or An. gambiae (p0.05) (Table 19). Larvae of chironomid were commonly observed to be consumed by £.(l!) tigripes in the field and preference tests carried out indicated that £. guinguefasciatus University of Ghana http://ugspace.ug.edu.gh 141 Table 19 Feeding preference of £. (1) tigripes for prey larval species Predator No. and species of No. of Mean No. of prey instar prey bowl trials species Test consumed/24hrs X ± SD 10 An. gambiae 8.0 ± 2.08 a F =8.58 4th 10 £. guinguefasciatus 30 7.3 ± 2.71 a P<0.01 10 Ae. aegypti 9.4 ± 0.86 20 An. gambiae 11.7 ± 1.45 b F = 29.91 4th 20 £. guinguefasciatus 20 12.2 ± 1.58 b P< 0.001 ! 20 Ae. aegypti 15.2 ± 1.62 4th 10 C. guinguefasciatus 6.0 ± 1.25 t = 2.25 10 Chironomid 21 5.05 ± 0.15 P<0.05 4th 20 C. guinguefasciatus 16.5 ± 1.05 t = 15.36 20 Chironomid 20 10.5 ± 1.40 P< 0.001 4th 10 C. guingyefasciatus 7.89 ± 1.55 t = 9.65 10 Chironomid 15* 2.40 ± 1.65 P- 60 IIJ a:: 11. 50 ll... 0 40 II) IIJ 0.05). The total duration University of Ghana http://ugspace.ug.edu.gh 144 of movements by all the fifty Q. (~) tigripes larvae was 83.0 seconds compared to 726, 311, and 205 seconds moved by Ae. aegypti, Q. guinguefasciatus and An. gambiae respectively. The longer duration of spontaneous movement of Ae. aegypti means it will increase the chances of its coming within the striking distance of the predator compared to Q. guinguefasciatus and An. gambiae. In fact the lack of difference in the preference by Q. (~) tigripes for either of these two species is butressed by the lack of difference in their spontaneous movement. Thus only 6.29% of all the spontaneous larval movements were shown by Q. (~) tigripes compared to 54.79%, 23.47% and 15.47% for Ae. aegypti, Q. guinguefasciatus and An. gambiae respectively. The results indicate that Q. (~) tigripes larvae seldom moves spontaneously in comparison to the three prey species. The results of induced movement of the larvae of the four mosquito species caused by two different methods of stimulation are given in Table 21 (Appendix Tables 21 and 22) . Ae. aegypti larvae continued to move for a longer time than all the other larvae after stimulation by tapping the edge of the container as well as by tapping the larvae. The durations of movement of Q.(~) tigripes after both methods of stimulati::.n were longer than the durations shown by Q. guinguefasciatus and An. gambiae. No significant difference (P>O.05) was observed between the two methods of stimulation. These results may explain what has been observed both in the University of Ghana http://ugspace.ug.edu.gh 145 field and in the laboratory with ~. (L) tigripes. It usually stays at one place in the water and seldom moves around by itself. Prey larvae which corne into contact with it or close to it are captured and consumed. Thus prey larvae, like Ae. aegypti which move more frequently in the water habitat are more likely to either get into contact with the predator or move into the predator's striking distance to be captured. Considering the posture that the mosquitoes assume, Anopheles larvae lie horizontal to the water surface, whereas Culex and Aedes larvae hang at about 45° to the surface. Ae. aegypti larvae normally rest at the surface or at the bottom of the water. While at the bottom, they usually browse for particulate food but they must surface every few minutes for air thus exposing themselves to the predator more often than the other two species. Also,it was often observed that Ae. aegypti larvae were also captured at the bottom by ~. (L) tigripes. It appears that the angle at which the prey larvae and the predator come to rest at the surface did not affect the feeding preference of the predator. ~. (L) tigripes, particularly the early instars rest at the surface almost in the same posture as Anopheles species, but this posture did not seem to enhance the predator's chances of capturing this prey. Though ~. guinguefasciatus and An. gambiae showed different postures at the surface of the water, the difference in the predation rate on the two species was not significant (p>O.05) (Table 19) even though more An. gambiae were taken University of Ghana http://ugspace.ug.edu.gh 146 than ~. guinguefasciatus. Factors such as prey mobility and avoidance behaviour have been observed in these studies as some of the characteristics that influence the type of prey selected by ~.(L) tigripes. The tendency to remain motionless more often when not stimulated seem to be a good attribute for the predator such as ~. (L) tigripes which does not actively search for its prey but lays in ambush for it.It will be advantageous to stay motionless in order not to attract the attention of any potential prey moving towards it. Turnball (1960) observed a somewhat similar situation in the selection of prey by the spider Linyphia triagularis (Clerck), and stated that an ideal prey was one which, among other things, was highly mobile, as this attribute increased its chances of becoming ensnarled in the spiders web. However, when ~.(L) tigripes is stimulated it moves for a relatively longer period than Culex and Anopheles. Usually the movement of a prey may cause some ripples or movements in the water or cause body contact with the predator, so it may consider such a stimulus as indicating the presence of a prey. The increase in the duration of movement of ~. (L) tigripes after stimulation may therefore be a follow up to capture a potential prey. The factors which may affect prey selection by a predator include prey size, mobility, palatability, avoidance behaviour or defense, abundance and the extent to which the predator and prey concur in the same habitats, (See Section 4.2.3). Young University of Ghana http://ugspace.ug.edu.gh 147 (1967) and Turnball (1960) pointed out that, each of these factors is part of a complex and the degree to which the predator responds to this complex will determine the preferred prey. For selecting a prey in nature, an even more important factor may perhaps be the degree to which the predator and prey share the same microhabitat. ~. (~) tigripes larva spends much time at the surface film but often moves to the bottom of water container and remains there for a long time before surfacing again. Because it spends much of its time at the surface film it comes into contact most often with prey which are regularly found at that level in the water column such as mosquito larvae, and this may explain why more Culex larvae were taken than chironomid larvae. Larvae of Culex were used instead of Aedes, which is the preferred species, because the former occurs more frequently with the predator in the same breeding habitats. Preys such as Aedes larvae which in nature spend a .'.ong time crawl ing at the bottom of containers browsing for food may again be more vulnerable to predation by ~. (~) tigripes particularly when the latter is at the bottom of the water. ~. (~) tigripes larvae did not feed on other organisms such as psychodidae larvae and pupae and nematodes when offered these organisms even though they occurred in the same habitat as ~. (~) tigripes. Similarly, adult insects which fell into the water were not preyed upon by ~. (~) tigripes. This is in contrast to the observation made by MacGregor (1927) that ~. (~) tigripes will eat almost University of Ghana http://ugspace.ug.edu.gh 148 anything including some of the organisms mentioned above. 4.3.5.3. Feeding preference by prey stages The results of the predation rates obtained when 4th-stage Culex larvae and pupae were offered simultaneously to £.(~) tigripes indicate that £.(~) tigripes selectively preyed on the larvae (Table 22) and that the mean numbers of larvae and pupae consumed in 24 hours at the two densities (10 and 20 per container) are significantly different (P0.05 - Duncan's multiple range test) Table 25 The rate of cannibalism of a 4th-stage g. (Jd tigripes on other stages No./ Stage of No of Mean No. bowl ll. tigripes/ trials k.!lled/24hrs F i bowl X ± SD 20 1st 17 19.35 ± 0.79a 234.73 20 2nd 17 18.14 ± 1.86a P X 52 > X p (X2) < 0.001 P (X2) < 0.001 University of Ghana http://ugspace.ug.edu.gh pupa 4 139 13 0 Eggs 18 3946 0 0 Total 80 5589 29 1 Oct. I 50 3811 0 0 II 12 1616 0 0 III 0 1017 0 6 IV 13 1404 1 1 Pupa 6 650 0 0 Eggs 232 6097 0 0 Total 313 14,595 1 7 Nov. I 84 867 0 0 II 15 550 0 0 III 29 508 0 0 IV 28 985 0 0 Pupa 22 176 0 0 Eggs 315 2948 0 0 Total 493 6034 0 0 Dec. I 17 784 0 0 II 22 635 0 0 III 26 976 0 0 IV 38 724 0 0 Pupa 21 125 0 0 Eggs 183 760 0 0 Total 317 4004 0 0 Jan . '90 I 46 1033 6 0 II 6 802 0 0 III 2 459 0 0 IV 41 557 6 0 Pupa 11 290 0 0 Eggs 294 3386 0 0 Total 400 6527 12 0 Feb. I 6 299 0 0 II 8 199 0 0 III 12 296 0 0 IV 25 698 0 0 Pupa 10 624 0 0 Eggs 0 2104 0 0 Total 61 4220 0 0 March I 34 868 0 0 II 21 761 2 0 University of Ghana2 h0t.t p://ugspace.ug.edu.gh III 4 301 0 0 IV 23 692 0 0 Pupa 15 366 12 0 Eggs 82 2783 0 0 Total 179 5771 14 0 Appendix Table 6 The number of various species and stages of mosquito larvae collected from a concrete drain (Site B) MQnth Immature stages lutzia Culex sJ;!J;!. Aedes sJ;!J;!. AnoJ;!heles sQ12 April '89 I 33 72 68 0 II 27 231 175 0 III 4 30 98 0 IV 20 150 97 0 Pupa 2 25 101 0 Eggs 74 618 0 0 Total 160 1126 939 May I 117 247 201 0 II 98 117 176 17 III 65 207 515 0 IV 144 150 937 Pupa 4 13 31 320 Eggs 0 96 332 0 Total 0 533 984 2149 21 June I 11 137 171 II 13 25 86 99 III 69 20 67 51 IV 21 86 101 230 Pupa 39 27 40 260 Total 27 236 746 811 169 July I 11 276 376 II 54 29 192 39 III 44 90 90 128 IV 88 130 138 Pupa 419 97 62 92 Eggs 263 38 70 371 Total 68 13 392 1159 1293 334 Aug. I 81 342 II 219 12 13 119 III 307 20 10 IV 197 63 2 24 258 Pupa 115 8 17 Eggs 81 42 3 60 Total 310 193 205 0 1326 939 45 University of Ghana http://ugspace.ug.edu.gh Sept. I 23 210 477 30 II 11 183 137 23 III 14 33 210 8 IV 20 127 418 32 Pupa 0 104 77 8 Eggs 0 101 0 0 Total 68 758 1319 101 Oct. I 88 533 238 63 II 32 217 430 18 III 20 201 121 9 IV 50 417 500 80 Pupa 29 102 91 20 Eggs 39 690 10 0 Total 258 2160 1390 190 Nov. I 50 83 38 1 II 27 30 127 3 III 41 215 69 0 IV 22 437 139 10 Pupa 10 120 27 1 Eggs 0 411 0 0 Total 150 1296 400 15 Dec. I 41 67 15 27 II 16 41 4 40 III 30 19 0 1 IV 38 288 11 8 Pupa 13 37 29 16 Eggs 0 200 0 0 Total 138 652 59 92 Jan. 1990 I 12 321 0 1 II 9 111 1 6 III 4 131 0 0 IV 7 248 26 0 Pupa 1 11 51 1 Eggs 0 0 0 0 Total 33 821 78 8 Feb. I 11 173 3 0 II 19 84 1 0 III 11 37 2 0 IV 7 141 0 0 Pupa 4 38 0 0 Eggs 0 107 0 0 Total 50 580 6 0 University of Ghana http://ugspace.ug.edu.gh March I 37 191 67 73 II 21 20 29 34 III 7 47 28 0 IV 28 120 2 2 Pupa 5 47 0 0 Eggs 0 217 0 0 Total 98 642 126 109 University of Ghana 2h0tt6p ://ugspace.ug.edu.gh Appendix Table 7 Numbers of immature stages of g. (~) tigripes collected each day in 100 samples from a man hole (site A) Number collected Collecting Instar I Instar II Instar III Instar IV Pupa Days 1 41 33 23 7 1 2 54 46 20 11 0 3 35 22 36 10 3 4 25 42 19 8 0 5 63 25 27 21 2 6 51 32 20 14 4 7 45 40 28 19 0 8 35 29 21 9 1 9 63 21 13 12 2 10 49 28 14 8 1 Total 461 328 221 119 14 Appendix Table 8 Numbers of immature stages of g. (~) tigripes collected each day in 100 samples from a man hole (site B) No. Collected Collecting Instar I Instar II Instar III Instar IV Pupa Days 1 21 20 17 10 1 2 31 23 13 7 0 3 26 26 27 6 2 4 23 26 18 10 0 5 22 20 16 8 1 6 28 25 17 9 1 7 31 21 14 8 0 8 23 28 20 10 3 9 29 25 15 8 0 10 30 23 19 12 1 Total 264 237 170 88 9 UniveLrsifiety of Ghan2a0 h4t tp://ugspace.ug.edu.ghAppendix Table 9 Table for~. (~) tigripes in a manhole (Site A) x nx Ix dx px qx ex 0 379 1000 137 0.8630 0.1370 3.0090 1 327 863 164 0.8100 0.1960 2.4079 2 265 699 206 0.7053 0.2947 1. 8560 3 187 493 216 0.5619 0.4381 1. 4220 4 105 277 172 0.3791 0.6209 1.1410 5 40 105 39 0.6286 0.3714 1.1900 6 25 66 53 0.1970 0.8030 0.5980 7 5 13 Appendix Table 10 Life Table for~. (~) tigripes in a concrete drain (Site B) x nx Ix dx px qx ex 0 300 1000 167 0.833 0.1670 3.0025 1 250 833 187 0.7755 0.2245 2.5042 2 194 646 163 0.7477 0.2523 2.0844 3 145 483 183 0.6211 0.3789 1. 6190 4 90 300 150 0.5000 0.5000 1.3017 5 45 150 93 0.3800 0.6200 1.1033 6 17 57 27 0.5263 0.4737 1. 6877 7 9 30 23 0.2333 0.7667 0.6167 8. 2 7 University of Ghana http://ugspace.ug.edu.gh Appendix Table 11 Physico-chemical analysis of water from a man hole and a concrete drain (sites A and B) Months pH Chloride Total Dissolved Temp oC Total No. alkalinity oxygen Predator Preys A B A B A B A B A B A B A B APR 8.0 7.1 45 41 87.58 80.7 4.67 4.91 26.30 27.7 318 160 3031 2065 MAY 7.6 7.2 45 43 89.13 83.4 4.93 5.06 26.5 28.3 952 533 8676 4131 JUN 7.4 7.2 46 41 90.33 85.91 5.01 5.4 28.1 27.8 944 236 7081 1726 N o JUL 7.S 7.1 48 48 80.25 84.7 5.53 4.2 27.2 25.9 80 392 1557 2786 ...... AUG 7.5 7.3 48 4S 90.03 90.03 5.08 4.71 24.3 25.1 323 205 3213 2310 SEP 7.2 7.2 46 49 98.33 91.2 3.38 4.39 25.6 26.7 80 68 5619 2178 OCT 7.3 7.1 42 37 78.75 91.7 4.58 4.S 27.5 27.3 313 258 14603 3740 NOV 7.3 7.S 44 41 81.43 91.11 3.43 4.0 27.5 28.1 493 150 6034 1711 1 DEC 7.4 7.4 45 44 91. 63 85.13 3.97 4.11 28.2 29.6 317 138 4004 803 JAN 7.3 7.4 41 46 92.17 83.7 4.31 4.83 27.8 29.8 400 33 6539 908 FEB 7.4 7.5 41 46 89.25 83.2 4.83 4.07 28.6 30.7 61 50 4220 586 MAR 7.3 7.8 42 49 90.73 82.6 3.68 4.0 28.9 31.1 179 98 5785 877 - -- ~--- - University of Ghana http://ugspace.ug.edu.gh Appendix Table 12 Ability of larvae and pupae of £. (~ tigripes to withstand high temperature Instar Period of exposure No. Surviving/ % (mins) Temp (oC) No. tested surviving Remarks Control Room Temp. 15/15 100 15 34 16/16 100 30 34 15/15 100 4th 60 34 15/15 100 15 36 15/15 100 30 36 15/15 100 Active 60 36 13/15 86.67 sluggish 15 ·37 10/12 83.33 Larvae moribund 30 37 5/12 41. 67 60 37 0/12 0.00 All died Control Room Temp. 15/15 100 15 34 10/10 100 30 34 10/10 100 Pupae 60 34 10/10 100 15 36 10/10 100 30 36 10/10 100 60 36 12/13 92.31 Active. 15 37 12/12 100 30 37 7/12 58.33 60 37 4/12 33.33 15 38 4/15 26.67 Pupae were inactive. 30 38 1/15 6.67 Pupae died after exposure to water at room temperature 60 38 0/15 0 All died University of Ghana http://ugspace.ug.edu.gh Appendex Table 13 Summary of the Colonization of~. (k) tigripes in the laboratory NO. % Surviving % Mortality Total Eggs (from field) 1352 0 Eggs hatched into larvae 1108 81.95 (hatching) No. of 1st instar larvae 853 76.99 23.01 (1st instar) entering 2nd instar No. of larvae pupated 582 68.23 31.77 (developing larvae) No. of adults emerged 579 99.48 0.52 Pupae No. of adults that 568 98.10 1.90 Adult survived Adult Females 303 Adult Males 265 :tv Total Females 303 Females blood fed 102 (33.66%) Blood fed females which laid eggs 13 (12.75%) No. of eggs/female 20-128 University of Ghana http://ugspace.ug.edu.gh ::1.10 Appendix Table 14 Number of eggs- :tct"id by Q. (1..) tigripes and the time taken to mature eggs in the laboratory Adult No. of eggs Time to mature eggs mosquito laid (days) 1 43 7 2 56 7 3 20 8 4 81 6 5 65 10 6 37 6 7 58 7 8 67 6 9 101 6 10 39 12 11 99 11 12 128 10 13 28 9 Total 792 105 X ± SO \ 60.92 ± 32.55 8.08 ± 2.10 Range 20-128 6-12 University of Ghana http://ugspace.ug.edu.gh 211 Appendix Table 15 Longevity of adult g. cl-) tiqripes under laboratory conditions Day Male % Female % 1 54 100 60 100 7 53 98.10 59 98.30 14 50 92.59 57 95.00 21 43 79.62 49 81.67 28 31 57.41 40 66.67 35 22 40.74 29 48.33 42 12 22.22 19 31. 67 49 5 9.26 15 25.00 56 2 3.70 9 15.0 63 0 0.00 4 - - 6.67 70 0 0.00 University of Ghana http://ugspace.ug.edu.gh 21'- Appendix Table 16 Average composition of Cerelac Infant cereal per 100 as given by the manufacturers (Food specialities Ghana Ltd) Fat 9.0 vitamin E 3.0 Protein 15.5 vitamin C 35.0 Carbohydrate 66.9 Folic Acid (mcg) 22.5 Dietary fibre 2.8 Thiamine B1 (mg) 0.8 Mineral Ash 3.3 Riboflavin (B2) 0.3 Moisture (water) 2.5 Niacin (PP) mg 4.0 Energy value KCal 411.0 vit B6 mg 0.3 KJ 1720.0 vit B12 mg 25.0 Linoleate 1.1 Pantothenic acid mg 1.5 Vit A (IU) 1030.0 Ca (mg) 530.0 vit D 200.0 Phosphorus 430.0 Iron (mg) 7.5 Sodium (mg) 160.0 University of Ghana http://ugspace.ug.edu.gh 211 _..J';' Appendix Table 17a Average composition of Dog biscuits by weight as given by the manufacturers (Nippon Pet Food Co. Ltd., Japan) Crude Protein 32.0% Crude Lipid 13.0% Crude fiber 3.0% Crude Ash 10.0% Water 1.2% Ca 1.2% P 1.0 Appendix table 17b Average composition of Milk Casein by weight per 100gm as given by the manufacturers (Wako Pure Chemicals Ltd., Japan) Protein 86.2 Ca 26 (mg) Energy 378KCal P 120.0 Water 10.6 Fe 0.8 Lipid 1.5 Na 10.0 Ash 1.7 K 2.0 University of Ghana http://ugspace.ug.edu.gh 2)4- Appendix Table 18 Effect of Predator stages, prey stages and prey densities on the number of ~. guinguefasciatus consumed by~. (1) tigripes. Prey Prey Predator instar 1 Predator instar 2 Predator ins tar 3 Predator instar 4 stage density n (X ± SO) n (X ± SO) n (X ± SO) n (X ± SD) 1st 01 15 8.27 ± 2.15 12 9.5 ± 0.08 10 5.0 ± 2.83 10 6.0 ± 2.0 02 9 13.14± 4.14 11 17.67± 2.07 12 16.5 ± 2.74 12 18.0 ±3.52 03 12 17.8 ± 4.34 12 31.5 ± 7.66 12 34.67 ± 9.18 34 34.43±2.88 2nd 01 14 3.43 ± 0.53 14 5.21 ± 1.53 12 8.5 ± 1.17 16 8.31± 2.47 02 12 9.67 ± 5.39 12 12.58 ± 3.50 14 16.33± 4.18 14 16.5 ± 8.09 D3 14 12.57 ± 0.79 17 20.89 ± 7.98 10 28.3 ± 4.50 11 37.86± 1.46 3rd 01 10 1.1 ± 0.88 11 2.71 ± 0.95 12 7.33 ± 1.83 10 9.2 ± 0.79 02 12 4.67 ± 1.21 12 12.83 ± 1.72 12 12.83± 4.83 14 14.0 ± 2.37 03 12 6.33 ± 0.52 12 7.75 ± 2.76 1426.56±3.47 12 38.17 ±2.86 4th 01 11 0.14 ± 0.38 10 1.2 ± 0.42 11 2.38 ± 1.30 16 7.19 ± 2.10 02 11 1.55 ± 1.37 12 3.67± 1.97 11 6.83 ± 1.72 14 16.0 ± 2.45 03 15 2.2 ± 1.15 12 4.0 ± 1.83 12 4.33 ± 0.82 12 32.89± 7.66 J - ----- 01 10 larvae/bowl 02 20 larvae/bowl 03 40 larvae/bowl n = number of replicates x mean of n trials so standard deviation University of Ghana http://ugspace.ug.edu.gh Appendix Table 19 Duration of spontaneous movements of mosquito larvae Expt. An. gambiae £. guinguefasciatus Ae. aegYQti £. (lutzia) tigriQes 1. 12,16,7,10,6 9,5,3,17,7 11,16,32,41,38 1,2,3,1,2 2. 3,5,8,12,8 12,18,11,9,20 31,7,29,23,20 2,1,3,3,2 3. 2,1,1,1,1 2,6,11,9,10 5,14,25,14,20 1,2,3,4,2 4. 3,1,19,10,6 2,1,6,1,1 23,8,14,30,24 2,3,1,2,2 5. 5,3,3,2,2 5,7,1,4,9 26,6,16,4,5 1,1,1,1,1 6. 1,4,6,3,5 3,15,20,2,3 17,11,6,19,8 2,1,1,2,1 7. 5,1,1,4,2 1,3,2,4,1 10,27,19,20,15 2,2,1,2,1 8. 1,2,1,2,1 20,11,9,7,9 3,15,21,10,7 1,2,1,1,1 9. 3,1,1,2,1 3,1,1,2,2 2,2,8,8,1 2,1,1,2,1 10. 2,4,3,1,1 1,2,1,1,1 7,1,3,2,2 1,2,1,3,1 Total 205 311. 0 726.0 83.0 X±SD 4.3 ± 4.44 6.22 ± 5.68 14.52 ± 10.30 1. 66 ± 0.77 _------i_ -- Each value is the duration of spontaneous movement of each larva within one minute b) The total number of seconds moved by a larva out of 5 mins. University of Ghana http://ugspace.ug.edu.gh Appendix Table 20 Duration of Induced movements of mosquito larvae Expt. An. gambiae g. guinguefasciatus Ae. aegYQti g. (lutzia) tigriQes 1. 3,1,1,1,2 2,4,6,2,2 12,6,7,7,21 7,4,6,6,2 2. 2,2,3,1,1 2,4,6,1,4 7,3,7,9,7 2,6,4,2,4 3 . 3,2,2,4,4 1,3,3,2,3 11,8,6,10,7 2,4,6,7,3 4. 4,3,4,2,1 1,3,1,2,2 36,9,6,5,3 7,5,3,5,4 5. 5,3,0,5,1 3,3,2,2,1 20,7,6,11,4 3,5,7,4,2 Total 60 65 235 110 X±SD 2.4 ± 1.38 2.6 ± 1.38 9.4 ± 7.03 4.4 ± 1.76 - Each value is the duration of movement in seconds per minute of each larva after stimulation by tapping the bowls Appendix Table 21U nDivuerartsiointy o of fI nGduhcaedn am ohvettmpe:n/t/su ogfs mpoascquei.tou gla.revdaeu .gh Expt. An. gambiae c. guinguefasciatus Ae. aegY12ti c. (lutzia) tigr 1. 2,1,2,3,1 4,3,3,1,3 7,10,5,16,8 5,2,3,4,1 2. 2,1,3,2,1 3,3,5,2,6 20,7,24,9,11 4,2,2,3,5 3. 2,3,1,3,2 2,4,1,2,3 1,23,7,14,9 5,2,3,3,3 4. 2,2,1,2,3 3,4,2,1,2 13,9,3,7,6 4,3,5,3,2 5 4,1,3,2,1 2,3,4,2,3 2,18,6,3,9 7,3,4,2,1 Total 50 71 247 81 X±SD 2.0 ± 0.87 2.84 ± 1. 21 9.88 ± 6.24 3.24 ± 1. 42 - Each value is the duration of movement in seconds per minute of each larva after stimulation by tapping the larvae University of Ghana http://ugspace.ug.edu.gh 21~ Appendix Table 22 Mean number of Q. guinguefasciatus eaten or killed but not eaten by the final(4th instar larvae of Q. (L) tigripes Day 1 Day 2 Day 3 Day 4 Prey density No. of larvaej A B A B A B A B container 40 26.2 ± 4.95 0.00 24.9 ± 7.04 0.00 19.5 ± 1.B1 0.00 - - (n = 20) (n = 20) (n = 20) 60 30.1 ± 2.27 0.00 26.7 ± 1.22 0.00 1B.2 ± 2.61 2.0 ± 1.21 B.OB ± 1.51 6.5 ± 1.07 (n = 20) (n = 20) (n = 16) (n = 4) 100 38.9 ± 2.07 0.00 29.8 ± 1.2B 0.00 20.4 ± 2.63 9.45 ± 2.99 4.65 ± 1.B2 14.36±2.76 (n = 20) (n = 20) (n = 14) (n = 6) I A Number of prey killed and eaten B Number of prey killed but not eaten University of Ghana http://ugspace.ug.edu.gh 2J9 Appendix Table 23 Functional Response of £. 111. tigripes to prey density No. of Prey/ Mean No. of Prey killed or 30mls of water consumed/predator in 24 hrs No. of trials 10 6.63 ± 2.45 18 20 12.64 ± 3.26 18 40 21.33 ± 3.37 16 60 27.38 ± 3.34 16 80 29.00 ± 1.35 14 100 30.00 ± 1.36 14 120 30.14 ± 1.63 15 Appendix Table 24 Feeding preference of £. 111. tigripes, when offered a choice of 3 mosquito prey species. Order of Prey Selection An. gambiae C.guinguefasciatus Ae. aegypti 1st 2nd 3rd 1st 2nd 3rd 1st 2nd 3rd No. consumed 7.0 18.0 17 .0 17.0 10.0 15.0 27.0 14.0 1.0 % consumed 16.7 42.9 40.5 42.9 23.8 35.7 64.3 33.3 2.38