<%BANNER%>

Identity and prevalence of blood parasites in wild-caught birds from Madagascar

University of Florida Institutional Repository

PAGE 1

IDENTITY AND PREVALENCE OF BLOOD PARASITES IN WILD-CAUGHT BIRDS FROM MADAGASCAR By AMY FRANCES SAVAGE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Amy Frances Savage

PAGE 3

ACKNOWLEDGMENTS I thank my friends and family for their support and lessons. Also, I thank the professors, research scientists, and other students with whom I have worked for being a source of inspiration and encouragement. Finally, I would like to thank the members of my committee. Drs. Donald J. Forrester and David W. Steadman were enthusiastic, energetic and creative. I am grateful for the time they took to teach and motivate me. I sincerely thank my committee chair, Dr. Ellis Greiner, for his kindness, patience and understanding. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi LIST OF FIGURES..........................................................................................................vii ABSTRACT.....................................................................................................................viii CHAPTER 1 INTRODUCTION........................................................................................................1 2 MATERIALS AND METHODS...............................................................................11 3 NEW SPECIES DESCRIPTIONS.............................................................................13 Haemoproteus goodmani n. sp...................................................................................14 Taxonomic Summary..........................................................................................15 Remarks...............................................................................................................16 Etymology...........................................................................................................16 Haemoproteus forresteri n. sp.....................................................................................16 Taxonomic Summary..........................................................................................17 Remarks...............................................................................................................18 Etymology...........................................................................................................18 Haemoproteus vangii n. sp.........................................................................................19 Taxonomic Summary..........................................................................................20 Remarks...............................................................................................................20 Etymology...........................................................................................................20 Haemoproteus khani n. sp...........................................................................................21 Taxonomic Summary..........................................................................................21 Remarks...............................................................................................................22 Etymology...........................................................................................................22 Haemoproteus dicruri .................................................................................................22 Taxonomic Summary..........................................................................................23 Remarks...............................................................................................................23 Leucocytozoon frasci n. sp.........................................................................................25 Taxonomic Summary..........................................................................................26 Remarks...............................................................................................................26 Etymology...........................................................................................................27 iv

PAGE 5

Leucocytozoon lairdi n. sp..........................................................................................27 Taxonomic Summary..........................................................................................27 Remarks...............................................................................................................28 Etymology...........................................................................................................28 Leucocytozoon greineri n. sp......................................................................................29 Taxonomic Summary..........................................................................................29 Remarks...............................................................................................................30 Etymology...........................................................................................................30 4 SURVEY RESULTS..................................................................................................36 Prevalence by host family...........................................................................................36 Altitude.......................................................................................................................39 Reserves......................................................................................................................39 Habitat.........................................................................................................................40 Gender.........................................................................................................................40 Age..............................................................................................................................41 Breeding Condition.....................................................................................................41 5 DISCUSSION.............................................................................................................45 Families.......................................................................................................................49 Vectors........................................................................................................................58 Altitude.......................................................................................................................61 Reserves......................................................................................................................63 Habitat.........................................................................................................................65 Gender.........................................................................................................................65 Age..............................................................................................................................66 Breeding Condition.....................................................................................................66 6 CONCLUSIONS........................................................................................................68 APPENDIX CHICKEN HEMATOZOA......................................................................70 LIST OF REFERENCES...................................................................................................73 BIOGRAPHICAL SKETCH.............................................................................................80 v

PAGE 6

LIST OF TABLES Table page 3-1 Morphometric variation in the haemoproteids from the Brachypteraciidae, Vangidae, and Dicruridae.........................................................................................31 3-2 Morphometric variations of the Leucocytozoon spp. of the Brachypteraciidae, Vangidae, and Philepittidae......................................................................................33 4-1 Prevalence of hematozoa in the avifauna of Madagascar, by avian family.............42 vi

PAGE 7

LIST OF FIGURES Figure page 3-1 Haemoproteus spp. of the Brachypteraciidae, Vangidae, and Dicruridae................34 3-2 Leucocytozoon spp. of the Brachypteraciidae, Vangidae, and Philepittidae...........35 vii

PAGE 8

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE IDENTITY AND PREVALENCE OF BLOOD PARASITES IN WILD-CAUGHT BIRDS FROM MADAGASCAR By Amy Frances Savage August 2003 Chair: Ellis C. Greiner Major Department: Pathobiology The Republic of Madagascar is an area of considerable biological interest because of the high degree of endemnism of the flora and fauna. Limited research has been performed investigating the hematozoa of the avifauna on the island; and little is known regarding the prevalence of these parasites or their effects. I examined 378 wild-caught birds from 21 families, and 15 domestic fowl, for the presence of hematozoa to determine the hematozoon fauna and prevalence of parasitism in the area. Birds were captured in mist nets; and blood smears were made in the field at four different reserves in Madagascar. Slides were stained with Giemsa and examined for the presence of hematozoa on a light microscope. Prevalence by genus of parasite was Haemoproteus spp. 13.8%, Leucocytozoon spp. 11%, microfilariae 6.1%, Plasmodium spp. 1.6%, Trypanosoma spp. 1.1%, and Babesia sp. 0.5%. Seven new species of hematozoa were recognized from the Brachypteraciidae (Ground-rollers), Vangidae (Vangas), Philepittidae (Asities), and Dicruridae (Drongos). Haemoproteus goodmani H. forresteri H. vangii H. khani Leucocytozoon frasci L. lairdi and L. greineri were described. The overall prevalence of infection observed was 24.3% (92 of 378). viii

PAGE 9

Zosteropidae and Ploceidae were the two most parasitized families. Prevalence also varied by altitude and sampling site. Birds from the lowest altitudes had the highest prevalence of parasitism. Birds from one reserve had a higher prevalence than the birds from other areas. This reserve had a high avian species density and a variety of habitats. No differences in prevalence were observed by habitat, gender, age, or breeding condition. ix

PAGE 10

CHAPTER 1 INTRODUCTION The Republic of Madagascar is a continental island located approximately 400 kilometers (km) off the southeastern coast of Africa. It is approximately 1600 km long, and 560 km across at the widest point, making it the fourth largest island in the world, after Greenland, New Guinea, and Borneo. At 590,000 km 2 (Kottack, 1980), it is roughly 2.5 times the size of Great Britain. Because of Madagascars proximity to Africa, many people assume that it is most closely associated with Africa. In actuality this is not the case. Originally, Africa, Madagascar, South America, Australia, India and Antarctica were all part of one supercontinent. During the Paleozoic era 200 million years ago, Africa is presumed to have broken off of Madagascar. Afterwards, Australia separated, and finally India. Therefore Madagascar has more recent geological ties to India than Africa. Madagascar is very similar to Africa in structure and climatic zones, perhaps because it shares the same ocean and air currents. It is an island with a range of habitats including coastal plains, tropical forest, and a semi desert. The highest peak is 2,876 meters. Humans were absent until approximately 2,000 years ago, when the earliest settlers colonized the island. The descendants of these Indonesian and African immigrants have now evolved into more than 20 different ethnic groups (Kottak, 1980). One of the important facts in ecological history about Madagascar is that for 40 million years it remained isolated, allowing the fauna and flora to evolve with little continental influence. There were no large mammals like those on Africa (Kottack, 1

PAGE 11

2 1980). Today, there are more varieties of orchids on Madagascar than anywhere else in the world. Most of the approximately 6,000 species of plants there are found only on this island. Ninety-eight percent of the nearly 500 reptile and amphibian species are endemic. The avifauna on the island is also distinct. There are 282 species of birds found on or near Madagascar and 110 of those are endemic. Five are known or thought to be extinct. Most of the endemic species (80 of 110) are forest dwelling. This represents 30 of 37 genera. Blood parasites were first observed in the 1880s. Since then, researchers have been trying to understand the intricacies of their life cycles and the role they play in human and animal disease. While a great deal has been learned since Daniewlskys discovery, there are still many unknowns facing researchers. While many projects have been designed to identify the blood parasites in birds, not every avian family has been investigated equally; and there are areas of the world where little, if any investigation has been done. One of the least investigated areas is Madagascar. There have been three studies of avian blood parasites in Madagascar. Bennett and Blancou (1974) examined 64 birds representing 32 species and found 14 birds (representing 8 species) to be infected. They concluded that the prevalence of hematozoa was low; and additionally that there were no unique species. The second study examined smears from 10 birds (Greiner et al., 1996). Blood parasites were found in 6 of the 10 birds; and mixed infections were observed for the first time. Most recently, Raharimanga et al. (2002) published a study of hematozoa of Malagasy birds from a variety of locations on the island. Unfortunately, parasites were not identified to the species level. No conclusions were drawn about the pathogenicity of parasites in these hosts. With

PAGE 12

3 such limited investigation, and in consideration of the wide variety of avifauna on the island, we have opted to work in collaboration with ornithologists studying the birds of Madagascar to conduct a more comprehensive survey of the islands blood parasites. Of several parasites that are seen commonly in the blood of birds, five were considered. These include three from the phylum Apicomplexa: Plasmodium Haemoproteus and Leucocytozoon Trypanosoma are extracellular, flagellated protozoa from the subphylum Mastigophora. Finally, microfilariae, the motile embryo from the Phylum Nematoda can also be found in the blood depending on the species. Each of these is vector borne, and has slightly different development. Species of Plasmodium, Haemoproteus and Leucocytozoon have a roughly similar life cycle. When the appropriate vector takes a blood meal from an infected host, the blood contains male and female gametocytes. Once inside the stomach of the vector, the macrogametocytes develop into macrogametes; while microgametocytes exflagellate and develop into several microgametes, that then seek the macrogamete. They unite, becoming a diploid zygote, which undergoes meiosis. The zygote elongates into the ookinete, which penetrates the stomach wall of the vector, and becomes an oocyst. Within the oocyst, haploid sporozoites develop. The mature oocyst ruptures, expelling sporozoites, which will eventually migrate to the salivary glands. Upon stimulation of probing, the sporozoites enter the acinae of the salivary glands and are injected into the host with the saliva. If it is a susceptible host, the sporozoites will be carried by the circulatory system to the reticuloendothelial system (RES), where they will develop into exoerythrocytic schizonts. The schizonts mature, releasing merozoites, that can re-enter RES cells and become a new generation of schizonts or enter the circulating blood cells. In the

PAGE 13

4 circulating erythrocytes, these merozoites will develop into trophozoites, then gametocytes; which they will remain until they are either ingested in a blood meal by an arthropod, or cleared by the spleen. Among species in these genera some life cycle differences exist that are important for identification. For instance, merozoites of Plasmodium that enter circulating erythrocytes will undergo a further generation of schizongony within the erythrocyte; which will eventually rupture, releasing the merozoites. These merozoites can invade other erythrocytes, and develop either into gametocytes or into schizonts. These additional schizogonic stages function to increase parasite numbers, as merozoites can develop into gametocytes at any stage. These erythrocytic schizonts make Plasmodium unique in the fact that you can inject blood from an infected host into a nave susceptible host and cause infection. Leucocytozoon spp. and Haemoproteus spp. only have exoerythrocytic schizonts, usually in hepatocytes and vascular endothelial cells. The number of times a parasite undergoes schizongony depends on that particular species. Whereas Plasmodium is well known for causing malaria in humans, Trypanosoma spp. are responsible for the well-known diseases Sleeping Sickness and Chagas disease. Of course, trypanosomes infect animals as well; and can be found in the blood of birds in the recognizable trypomastigote stage. Trypanosomes are also vector borne the arthropod vector takes up the parasite during a blood meal and it develops in the arthropod. A few days later, the arthropod takes another blood meal and defecates while feeding; and the parasite enters the site of feeding. Some trypanosomes develop anteriorly and are injected with the saliva of the vector. Usually, the flagellates appear in the feces in the highest number when the insect is prepared to take another blood meal, increasing the

PAGE 14

5 chances of transmission (Bennett, 1961). The parasite requires the vector to begin digestion before multiplication of the parasite begins. That is, the parasite is somewhat dependent on digestion (Bennett, 1961). Replication does occur. In the same study mosquitoes that consumed approximately 155 trypomastigotes contained approximately 130,000 at 51 hours post blood meal (Bennett, 1961). The fifth type of hematozoa found in the blood of the birds sampled is microfilariae. Microfilariae are the motile embryos of filaroid nematodes. Not all microfilariae are present in the blood; some are tissue dwelling. Those found in the blood can survive in the blood for several years. They are also vector borne, and when a suitable vector picks up the microfilariae, they develop into the L 1 (or rhabditiform larvae) that penetrate the midgut wall into the hemocoel. After two molts, they mature into the filariform larvae (L 3 ). These are the larvae that return back into the vertebrate host upon feeding of the arthropod. In the vertebrate the larvae will molt twice more to the adult form, will migrate to the preferred tissue, and the females will produce the next generation of embryos. For reference, in humans, filarial infections cause river blindness; and Bancrofts filariasis or elephantiasis. Although scientists have been studying avian blood parasites for over 100 years, relatively little is known about the pathogenicity associated with infection. One of the more-studied areas is Hawaii, where Plasmodium relictum was introduced, and has been affecting native and introduced species. Recently, Yorinks and Atkinson (2000) investigated the effects of malaria on activity budgets of juvenile Apapane ( Himatione sanguinea ). First, they found that the bite from one infected mosquito caused a fatal infection (infection resulted in acute anemia) in five of their eight birds. Additionally,

PAGE 15

6 they reported that the infected birds had a decline in several activities, becoming essentially inactive at peak parasitemias. The authors noted that these infected animals would have been at a competitive disadvantage against other birds in the area, and may have been more susceptible to predation and heat stress (Yorinks and Atkinson, 2000). This agrees with earlier findings, demonstrating canaries experimentally infected with P. relictum suffer a drop in body temperature, indicating an inability to thermoregulate (Hayworth et al., 1987). The authors hypothesize that this could increase mortality in extreme environments. This also corresponds with other work in the area, where moribund birds and birds killed by automobiles had higher prevalence of malaria than mist netted birds, supporting the theory that infected birds were more likely to be killed (van Riper et al., 1986). Penguins have also suffered significant losses as a result of avian malaria. In fact, it has been called the most important cause of death in captive penguins displayed in open-air exhibits around the world (Stoskopf, 1979). In a five-month period from 1967-1968, six African penguins at the Baltimore Zoo died as a result of malaria. By inoculating tissue and blood emulsions into healthy birds, scientists were able to identify Plasmodium elongatum Fix et al. (1988) reported on 46 Magellanic penguins, 22 of which died of malaria as the result of infection with Plasmodium relictum At necropsy, lesions typical of avian malaria were reported, including splenomegaly, hepatomegaly and pulmonary edema. Additionally, exoerythrocytic schizonts were observed in multiple tissues, including spleen, lung, liver, heart, brain and kidney. Plasmodium is not the only genus associated with disease and mortality in birds. Anemia and mortality in Pekin ducklings was associated with Leucocytozoon simondi

PAGE 16

7 although the anemia was not associated with peak parasitemia (Kocan and Clark, 1966). A central nervous system disease in kestrels is also associated with a Leucocytozoon -like parasitic infection (Raidal 2000). Circulating gametocytes were visualized and there were tissue schizonts observed in vascular endothelial cells in the brain, but none were observed in hepatic tissues, the typical location of schizogony in Leucocytozoon Other species of Leucocytozoon are known to have detrimental effects on their hosts. Leucocytozoon caulleryi has been shown to so severely affect reproductive organs in layer hens, resulting in the cessation of egg production (Nakamura et al., 2001). The genus Haemoproteus is often thought of as relatively benign, although it too is responsible for pathology and decreased performance. Haemoproteus melegridis has been the target of several studies. Atkinson et al. (1988) reported reductions in growth and weight gain in experimentally infected poults. There were four fatalities in that study, and large ruptured megaloschizonts surrounded by infiltrate were found in each (Atkinson et al., 1988a). An interesting mystery in the field of avian blood parasites is the question of host specificity. It has been believed that species of the genera Haemoproteus and Leucocytozoon are specific at the host family level (Fallis et al., 1954; Baker, 1968; Bennett et al., 1994). Because of this, many new parasites are named when previously unstudied hosts are examined. Species of Plasmodium are much less specific and are known to infect birds of different orders (Bennett et al., 1993). Microfilariae and Trypanosoma spp. (Bennett, 1961; Fallis, 1973a) are also known to not have strong host specificity. Therefore, the limiting factor in these cases is not the host, but the presence of a suitable vector. Each of these parasites is vector borne, and these vectors each have

PAGE 17

8 different behaviors and habitats, which will affect what birds they will encounter. Even if birds were introduced that were infected with one of these parasites, they are not increasing the odds of an epizootic unless there is a suitable vector present. The preferred vectors are different for each parasite. The common vector for species of Leucocytozoon and Trypanosoma are the Simuliidae (black flies). The mosquito, Culicoides (Ceratopogonidae) and louse flies (Hippoboscidae) are the vectors that transmit Haemoproteus spp. The Culicine mosquitoes vector Plasmodium species. What is important then, is the understanding that behavior and habitat of birds will influence their chances of being fed on by a vector of one of these parasites. Each of these vectors varies in flight range, flight altitude, feeding times, habitat, breeding requirements, and so on. Some may require brackish water for completion of their life cycle, while others may need running water. So then, if a bird migrates from one habitat daily, it may not become infected. For instance, if a bird spends evenings and nights in high elevations, and moves down to lower elevations during the days, it will not be exposed to those biting flies in the lower elevations that only feed at night. Some of the factors we have elected to consider have been discussed before. One is to analyze parasite prevalence by altitude. Van Riper et al. (1986) found that elevation had a marked influence on parasitemia levels (based on 16 sampling stations at 300 m intervals). They found that the highest parasitemia occurred between 900 and 1500 m in elevation, where the vector and bird populations overlapped. It has been suggested more than once that avian malaria is responsible for the dramatic decline in endemic Hawaiian birds, and even their extinction (Warner, 1968; van Riper et al., 1986). With the thirteen different altitudes ranging from 0 to 1950 meters, some with multiple sample sites at the

PAGE 18

9 same altitude, we have a good possibility of identifying patterns of infection relative to altitude. This can be used later in attempts to identify vectors of these parasites, as their range should closely match the distribution of the parasite they vector. Additionally, subsequent researchers can use the patterns of parasite prevalence to indicate what types of vectors may be present. Another factor to be examined is age, juvenile vs. adult. Van Riper (1986) found that younger birds were not more likely to be parasitized than older birds, and that younger birds had a higher parasitemia, possibly indicating lesser resistance. Using Plasmodium circumflexum Herman (1975) found that younger ducklings (one week or younger) displayed a recognizable parasitemia earlier than older ducklings, while older ducklings had a longer pre-patent period. In a seven-year period, Beier and Stoskopf (1980) reported 16 first and second year juvenile penguins died of malarial infections, but no adults died of malaria during the same time. This is of concern in birds that have a long breeding interval, longer time to sexual maturity, or produce few offspring each year. These are the populations that would not fare well if challenged with a pathogenic parasitic infection. This project is based on classical methods of parasite identification. This method is still very important and as applicable as ever, although some researchers have voiced concerns about misdiagnosing chronic, sub-clinical infections. Modern molecular techniques have been applied as diagnostic tools, but the results have not been ideal. Jarvi et al. (2002) reported that PCR tests underestimated chronic infections by 20%, but were perhaps more applicable to longitudinal studies where repeated sampling is occurring. Diagnosis of parasitism by reading blood smears is more applicable in field

PAGE 19

10 situations, particularly in remote study sites. Researchers in remote areas have more difficulties in handling vials for blood sampling, for later molecular diagnostics. It is more realistic and applicable to make blood smears on glass slides. In addition, the old and new systems need to be run in concert to validate the new approaches. Glass slides are more compact, and are easily stored indefinitely as long as they are protected from insects. Additionally, they serve as a permanent visual reference, and perhaps later can be used in conjunction with molecular techniques. It is the goal of this project to examine the prevalence of blood parasites in a retrospective survey from the years 1994 and 1995. Furthermore, I expect to find species of hematozoa previously undescribed, and possibly unique to the island. In addition to the prevalence of parasites, I plan to identify the parasites found to the species level and describe new species where possible. Finally, I hope to examine prevalence by sampling site, habitat, altitude, gender, and family groupings to identify any relationships that may exist. The goal of this project to make a strong first effort in teasing out the answers to some of these questions.

PAGE 20

CHAPTER 2 MATERIALS AND METHODS Ornithologists working at field sites in Andohahela, Anjanaharibe-Sud, Ambohitantely, and Montagne d'Ambre, Madagascar mist-netted birds at elevations of 120, 400, 440, 810, 875, 1000, 1200, 1260, 1400, 1500, 1550, 1875, and 1950 meters. Data regarding age, gender, and presence or absence of brood patch were recorded at capture when possible. Species were designated into one of three habitat preferences: forest dependent (FDE), forest dweller (FDW), and forest edge (FED). Blood smears were prepared in the field by toe clips or heart puncture. Blood was dropped onto a clean glass slide and spread into a monolayer using a second slide. Slides were air-dried and some were fixed in methanol at that time. Slides were then packed and shipped to the University of Florida. On arrival, the unfixed slides were fixed in methanol and all slides were stained with Giemsa, and stored. Slides were examined for the presence of hematozoa on a Zeiss light microscope at 100x, 160x, and 1000x (oil immersion). Due to varying quality, slides were examined for 30 minutes before being declared free of parasites. Slides with blood parasites were examined on a Nikon compound microscope, and parasites were drawn with the aid of a drawing tube. All measurements, excepting parasite length, of erythrocytes and parasites were performed as described in Bennett and Campbell (1972). Area was calculated using a drawing tube and grid, as described in Forrester et al. (1977). To calculate nuclear displacement, the formula 2X/(X+Y) was used, where Y is the distance between the periphery of the cell and the periphery of the host cell nucleus on the side which the parasite occupies. In the case of circumnuclear 11

PAGE 21

12 parasites, Y is calculated from the side on which all or most of the parasite nucleus lays. X represents the distance between the host cell membrane and host cell nucleus on the other side of the erythrocyte. One indicates no displacement, zero indicates total displacement of host cell nucleus to host cell margin. Parasite length was determined by measuring a line drawn to bisect the gametocyte along its longitudinal axis. All statistical comparisons were made with SigmaStat, using either Chi-square or Fishers Exact tests to compare prevalences. Unless otherwise indicated, alpha = 0.05.

PAGE 22

CHAPTER 3 NEW SPECIES DESCRIPTIONS Madagascar is home to both endemic and broadly distributed avian families. Four families, three endemic and one found more widely were investigated for the presence of hematozoa. The Brachypteraciidae is endemic to Madagascar and encompasses three genera and five species of Ground-rollers. They are found in tropical and subtropical rainforest and arid thornscrub. All are medium sized terrestrial birds, feeding mainly on small invertebrates and vertebrates encountered while foraging in leaf litter on the forest floor. Three of the five species are threatened, mostly due to loss of habitat. This family of birds requires undisturbed, pristine forest, which is being degraded by traditional slash and burn agriculture, mining, and logging. Additionally, cattle grazing the understory and well as hunting by humans for consumption are concerns (Langrand, 2001). The Vangidae and Philepittidae are also endemic. The Vangidae encompasses 12 genera and 14 species, found mainly in forested areas and also savanna and subdeserts (Clements, 2000). Only Cyanolanius madagascarinus (the Blue Vanga) is found outside of Madagascar, on Grand Comoro and Moheli Island (Langrand, 1991). In the Philepittidae, there are two genera, each with two species. They are found in a variety of forests, mainly in eastern Madagascar although Schlegels Asity ( Philepitta schlegeli ) is found in the dense forests of western Madagascar (Clements, 2000). A fourth family, the Dicruridae, was also examined. It is made up of 20 species of birds, occurring in Africa, India and Australia (Langrand, 1990). The Crested Drongo ( Dicrurus forficatus ) is the only member of the family on Madagascar, and occurs in 13

PAGE 23

14 Madagascar, and Anjouan in the Comoros. It can be found commonly in a variety of habitats from forests to sparsely wooded terrain and plantations (Morris and Hawkins, 1998). To date, three studies have examined birds from Madagascar for hematozoa (Bennett and Blancou, 1974; Greiner et al., 1996; Raharimanga et al., 2002). Raharimanga et al. (2002) reported hematozoa from Brachypteraciidae, but they were not identified to the species level. Greiner et al (1996) reported a Leucocytozoon sp. from both a Hook-billed Vanga ( Vanga curvirostris ) and a Velvet Asity ( Philepitta castanea ), but did not describe them. Only one species of Haemoproteus has been reported in the family Dicruridae. Haemoproteus dicruri was first described by de Mello (1935) from Dicrurus macrocercus and later redescribed by Peirce (1984b) from D. adsimilis I identified several new species of Haemoproteus and Leucocytozoon in these families. The Brachypteraciidae had three new species of hematozoa, two of Haemoproteus and one of Leucocytozoon One species each of Haemoproteus and Leucocytozoon were observed in Vangidae. One new species of Leucocytozoon is described from Philepittidae, and a new species of Haemoproteus is described from the Dicruridae. Several authors have discussed host specificity in both of these genera (Fallis et al., 1954; Fallis et al., 1974; Atkinson, 1986; Bennett and Peirce, 1988; Bennett et al., 1991). Based on this, the Haemoproteus and Leucocytozoon species described here are considered to be new species, specific to their respective families. Haemoproteus goodmani n. sp. Immature gametocyte: Young parasites develop laterally to host cell nucleus in mature erythrocytes, either in contact with or free from the host cell nucleus. Margins sometimes slightly amoeboid.

PAGE 24

15 Macrogametocyte: (n=23) Table 3-1, Figures 3-1(1), 3-1(2) Female gametocyte halteridial, with smooth or slightly irregular or amoeboid margins. Parasite tips commonly extend beyond erythrocyte nucleus towards limiting margin of the erythrocyte. Erythrocyte nucleus displaced laterally, with an average NDR of 0.66. Host cell nucleus not distorted however, maintaining the same average length and width observed in uninfected erythrocytes. Host-parasite complex slightly larger in area than uninfected erythrocyte. Area increases 6.2%, with parasite taking up 50% of host-parasite complex and 60% of erythrocyte cytoplasm. Outer margin of parasite not usually observed in contact with erythrocyte limiting membrane. Additionally, inner margin of parasite often, not always, in contact with host cell nucleus. A vacuolated cytoplasm, which does not stain deeply or evenly, results in a mosaic appearance. Parasite nucleus not always discernible, often located terminally and rarely centrally. From one to eight fine yellow refractile granules are rarely seen, and are terminal or central. Volutin granules commonly seen scattered throughout the cytoplasm, and are large with an average of 14 per gametocyte. Microgametocyte: (n=12) Microgametocyte has the general morphology of the macrogametocyte. Cytoplasm does not stain and appears white, with a large lightly pink-staining nucleus centrally located. Parasite nucleus diffuse, occupying 25-50% of microgametocyte. Taxonomic Summary Type host: Pitta-like Ground-roller ( Atelornis pittoides ), Lafresnaye, 1834, Brachypteraciidae. Type locality: Ambohitantely, Madagascar, latitude 18 04 to 18 14S, longitude 47 12 to 47E.

PAGE 25

16 Basis of description: Parasites are described from a blood smear taken from an adult Atelornis pittoides (Pitta-like Ground-roller). HAPANTOTYPE: Blood smear AA-83 Atelornis pittoides collected by Aristide Andrianarimisa on 13 October 1994 in Ambohitantely, Madagascar at 1500 meters altitude. Accession G463728, IRCAH Distribution: It is expected that this parasite will be found throughout the range of the Ground-rollers on Madagascar. Remarks This parasite is medium sized and often, but not always, in contact with the host cell nucleus. It causes only a slight increase in area of the infected cell, and slight displacement of the erythrocyte nucleus. Etymology This parasite is named after Dr. Steven M. Goodman, biologist and ornithologist, for his years of dedicated field work in Madagascar collecting information for biological inventories, invaluable to Malagasy officials and conservationists worldwide. Additionally, his steadfast efforts in making blood smears from birds for the evaluation of hematozoa are recognized. Haemoproteus forresteri n. sp. Immature gametocyte: Young parasites lateral to host cell nucleus in mature erythrocytes. Presses against the limiting membrane and host cell nucleus from an early stage. Microhalteridial as immature, progressing through a thick halteridial phase before reaching mature form. Terminals of developing gametocyte progress along periphery of host cell nucleus, until they connect. Parasite then grows outward until entire host cell cytoplasm is filled.

PAGE 26

17 Macrogametocyte: (n=10) Table 3-1, Figure 3-1(3) Mature macrogametocytes halteridial, becoming circumnuclear with the ends of parasite almost touching or completely touching, rarely a thick gametocyte completely displacing host cell nucleus against the host cell membrane. Host cell nucleus not distorted in length or width, but has a NDR of 0.51. Infected host cell increased in length from 15 m to 16.2 m, but decreases in width from 12.4 m to 10 m, and there is a 17.6% reduction in area. Parasite occupies 64% of host-parasite complex, and 84% of cytoplasm. Parasite margins generally smooth and occasionally amoeboid. Stains blue to light blue with Giemsa. Parasite nucleus compact, lightly staining and pink in color, and sometimes indistinct. When visible, gametocyte nucleus commonly touching outer periphery of parasite, closest to host cell limiting membrane. Pigment granules not always observed and range in number up to 12 when seen. They are fine and appear white or light yellow. Volutin fine and dust-like, commonly seen accumulated at the ends of the gametocyte. Microgametocyte: (n =11) Figure 3-1(4) Mature microgametocyte similar to macrogametocyte in its displacement of host cell nucleus to margin and becoming circumnuclear. Parasite nucleus sometimes large, but not diffuse and stains pink with Giemsa. Taxonomic Summary Type host: Rufous-headed Ground-roller ( Atelornis crossleyi ), Sharpe, 1875, Brachypteraciidae Type locality: Anjanaharibe-Sud, Madagascar, 14 44.8S, 49 26.0E Basis of description: Parasites are described from a blood smear from an adult Atelornis crossleyi (Rufous-headed Ground-roller). HAPANTOTYPE: Blood smear SG

PAGE 27

18 200 collected by Steven M. Goodman on 27 November 1994 at 1950 meters in Anjanaharibe-Sud, Madagascar. Submission G463729, IRCAH. Additional hosts: Atelornis pittoides (Pitta-like Ground-roller) Distribution: It is presumed that this parasite will be found throughout the range of the Ground-rollers on Madagascar. Remarks This parasite differs from the only other Haemoproteus sp. in Brachypteraciidae, namely H. goodmani in that the mature gametocyte is circumnuclear or completely displaces the host cell nucleus. Further, the inner margin of the parasite is usually pressed firmly against the host cell nucleus throughout its development, unlike H. goodmani The nucleus of H. forresteri is more compact and readily visualized than in H. goodmani Also, volutin granules of H. forresteri are fine and dust-like found at the terminus or periphery, where the volutin in H. goodmani is large and distributed randomly throughout the gametocyte. Finally, H. forresteri causes hypertrophy of the host cell. This parasite is described with two predominant morphologies, which, while uncommon is not unprecedented. Haemoproteus sacharovyi is described as pleomorphic with multiple common forms. Specifically, in Bennett and Peirces redescription (1990), there are four forms reported. Etymology This parasite is named after Dr. Donald Forrester of the University of Florida, in recognition of his significant contributions in the fields of parasitology and wildlife disease.

PAGE 28

19 Haemoproteus vangii n. sp. Immature gametocyte: Trophozoites and young gametocytes usually centrally located, lateral to the erythrocyte nucleus. Sometimes sub polar, near end of host cell nucleus, but still lateral. Macrogametocyte: (n=27) Table 3-1, Figure 3-1(6) Found in mature erythrocytes, macrogametocyte microhalteridial to halteridial, and stains a moderate to light blue with Giemsa. Macrogametocyte may appear to be more rod shaped, only slightly curved along the host cell nucleus, not wrapping around. Parasite ends commonly do not reach to ends of erythrocyte, but longer gametocytes can be observed. There is little distortion of the host cell by this parasite. Area of the host-parasite complex is only slightly (5.2%) larger than uninfected erythrocytes. Host cell length increases slightly from 15.8 m to 18.3 m. Host cell nucleus is not distorted in width or length, but is displaced laterally with a NDR of .071. Parasite encompasses 48.9% of the host-parasite complex, and 58.7% of the host cell cytoplasm. The parasite abuts the host cell nucleus, but is not always in contact with the host cell membrane, even at maturity. Parasite margins can be smooth or slightly amoeboid, or a combination of both. Parasite nucleus is small, averaging 6.2 m 2 typically sub-central, against or very near the outer margin of the gametocyte. Parasite nucleus stains pink and is compact and dense in appearance. Pigment granules are very fine, but still clearly refractile, and appear light yellow or white and usually are scattered through the cytoplasm but occasionally clumped. Number ranges up to 14 per cell, but the average was 7. Microgametocyte: (n=20) Figure 3-1(5) Male gametocyte morphology as described above. The gametocyte stains very lightly with Giemsa, often appearing clear. Parasite margins frequently appear indistinct. The microgametocyte is as likely to have

PAGE 29

20 amoeboid margins as the macrogametocyte. Nucleus stains lightly, and is larger than in the macrogametocyte, with an average area of 17 m 2 Taxonomic Summary Type host: Blue Vanga ( Cyanolanius madagascarinus ), Linnaeus, 1766, Vangidae Type locale: Andohahela, Madagascar, 24.6S, 46.3E Basis of description: Parasites are described from a blood smear from an adult Cyanolanius madagascarinus HAPANTOTYPE: Slide SG-551A collected by Steven M. Goodman in Andohahela, Madagascar, 3 November 1995. Submission G763730, IRCAH. Distribution: It is presumed this parasite will be found throughout the range of the Vangas on Madagascar, and possibly Grand Comoro and Moheli Island. Additional Hosts: Tylas eduardi (Tylas), Leptopterus viridis (White-headed Vanga) Remarks Haemoproteus vangii is the only species of Haemoproteus reported in Vangidae. It becomes markedly microhalteridial, clearly cupping the erythrocyte nucleus. Fine refractile granules and volutin are observed in H. vangii The gametocyte margins of H. vangii are commonly amoeboid. The author used the most mature forms of the parasite for the species description, but some of the gametocytes used for measurements may not have been fully developed. Therefore, the true area of the parasite may be slightly larger than indicated here. Based on the uniformity of this parasite, the other parameters should not be greatly affected. Etymology The parasite name is taken from nominate species of the host family, Vangidae.

PAGE 30

21 Haemoproteus khani n. sp. Macrogametocyte: (n=13) Table 3-1, Figure 3-1(7) Circumnuclear at maturity, lightly staining blue with Giemsa. Nucleus commonly indistinct, light pink or clear when visible. Margins smooth or slightly irregular, but not amoeboid. Fully developed host cellparasite complex area is 130.3 m 2 8% larger than uninfected erythrocytes. Host cell nucleus is displaced slightly, with an NDR of 0.66. Average parasite area is 110.6 m 2 host cell nucleus is reduced from 22.4 to 19.7 m 2 Parasite length is 28.1 m. Infected cells are slightly increased in length. When visible, parasite nucleus is 6.3 m 2 (6% of parasite). Refractile granules located randomly through parasite, sometimes clumped. Pigment granules small, generally round and white or yellow in color, not dark. Average number is 10.7. No volutin is observed. Parasite develops in mature erythrocytes. Developmental stages halteridial, ends wrapping around erythrocyte nucleus. First circumnuclear contact between parasite ends occurs touching host cell nucleus, then parasite grows outward filling last remaining host cell cytoplasm. Host parasite complex at this stage is 10 m 2 larger than fully mature complex, but parasite has the same area, indicating host cell shrinkage as parasite matures. Microgametocyte: (n=7) Figure 3-1(8) Same morphological characteristics as described above, with gender staining associated differences. Parasite stains virtually clear, commonly only noticed as a result of the pigment granules. Parasite nucleus, pink when observed, has average area of 27.6 m 2 25% of parasite area. Taxonomic Summary Type host: Crested Drongo ( Dicrurus forficatus ), Linnaeus, 1766, Dicruridae Type locale: Andohahela, Madagascar, 24.0S, 46.6E

PAGE 31

22 Basis of description: Parasites are described from a blood smear taken from an adult Dicrurus forficatus (Crested Drongo) HAPANTOTYPE: Blood smear SG-604 collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at 120 m. Accession G763732, IRCAH. PARAHAPANTOTYPE: Blood smear SG-605 collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at 120 m. Accession IPM-11 to LInstitut de Pasteur de Madagascar. Distribution: It is presumed that this parasite will be found throughout the range of the Drongos on Madagascar, and possibly beyond. Remarks This is the first circumnuclear haemoproteid recorded from the Dicruridae. The parasite has a characteristic development in that it generally first connects with the other end of the parasite along the host cell nucleus margin, then grows together and outward from there. Etymology The parasite is named in recognition of the significant body of work produced by Rasul A. Khan, in particular his efforts in the study of hematozoa. Haemoproteus dicruri Macrogametocyte: Table 3-1, Figure 3-1(9) Halteridial gametocyte in mature erythrocytes. Gametocyte fully displaces host cell nucleus laterally to erythrocyte margin (NDR 0.01). Ends of parasite do not wrap around erythrocyte nucleus; they do not cross the plane created by the opposite side of the erythrocyte nucleus. Parasite length is 15.9 m. Host parasite complex slightly larger than uninfected erythrocytes with an area of 128 m 2 Parasite is 106.3 m 2 occupying 98% of host cell cytoplasm and 83% of the host-parasite complex. Macrogametocyte stains light blue with Giemsa, parasite nucleus

PAGE 32

23 stains pink. Parasite nucleus usually central, commonly against host cell nucleus. Refractile granules are sometimes noticeably rod-like, with 11.9 per gametocyte. Microgametocyte: As described above, with sexual differences. Microgametocyte nucleus boundaries often indistinct. Taxonomic Summary Type host: Crested Drongo ( Dicrurus forficatus ), Linnaeus, 1766, Dicruridae Type locale: Andohahela, Madagascar, 24.0S, 46.6E Basis of description: Parasites are described from a blood smear taken from an adult Dicrurus forficatus (Crested Drongo) HAPANTOTYPE: Blood smear SG-604 collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at 120m. Accession G463733, IRCAH. PARAHAPANTOTYPE: Blood smear SG-605 collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at 120m. Accession IPM-11 to LInstitut de Pasteur de Madagascar. Distribution: It is presumed that this parasite will be found throughout the range of the Drongos on Madagascar, and possibly beyond. Remarks This is a new host record for Haemoproteus dicruri De Mello (1935b) originally described this parasite from a Black Drongo, later Peirce redescribed it from a Fork-tailed Drongo (1984b). De Mello described an ovoid, convex parasite with rod shaped pigment granules. He describes lightly staining microgametocytes with indistinct nuclei, and displaced erythrocyte nuclei (de Mello, 1935b). Peirce (1984b) also describes a pale microgametocyte with an indistinct nucleus. Also, Peirce observed the displacement of the host cell nucleus and stated that the parasite did not wrap around the host cell nucleus or become circumnuclear. Here we observe the same qualities. We observed a parasite

PAGE 33

24 that does not wrap around the erythrocyte nucleus, displaces the host cell nucleus, stains lightly with an often indistinct parasite nucleus. While Peirce did not observe rod-shaped pigment granules, we noted that the pigment granules are not always rod-shaped, and rounder granules are common. Additionally, we observed that the parasite occupies more than 90% of the host cell cytoplasm. This is slightly more than observed by Peirce (1984b) and it is unknown if this is a result of host-induced variation or those observed here were more mature parasites. Haemoproteus dicruri is differentiated from H. khani largely by its length, nuclear displacement ratio and by the tendency not to envelop the erythrocyte nucleus. Haemoproteus dicruri seems to have a more rigid composition, being almost more rod like. Haemoproteus khani causes little displacement of the host cell nucleus and grows around it easily. Haemoproteus khani becomes circumnuclear as it becomes a fully mature gametocyte, and is approximately 12 microns longer at maturity than H. dicruri The NDR of 0.01 associated with H. dicruri observed here readily supports the morphology easily visualized, a host cell nucleus completely displaced laterally. By comparison, H. khani has an NDR of 0.66, indicating only minor displacement by the gametocyte. Both parasites discussed here stain in a similar manner, and it is often easy to overlook the microgametocytes unless the refractile granules are discerned. Care should be taken when examining blood smears from Dicrurids with few mature gametocytes. Species differentiation is based on the nuclear displacement ratio and the degree to which the cap formed by the host cell nucleus encircles the host cell nucleus.

PAGE 34

25 Leucocytozoon frasci n. sp. Immature gametocyte: Young parasite displaces host cell nucleus to host cell membrane very early in development. Possibly originates from a central position, and grows outward laterally, filling host cell cytoplasm in most cases. Although hard to determine conclusively, host cell appears to be an erythrocyte. Macrogametocyte: (n=32) Table 3-2, Figure 3-2(1) Round morph most common, but distortions also commonly encountered. Macrogametocyte stains blue with Giemsa stain. Parasite nucleus clearly distinguishable, staining pink, and has an average area of 16.3 m 2 Host-parasite complex has an average area of 424.2 m 2 three times the area of uninfected erythrocytes. While the parasite distorts the host cell, making the identification of the host cell impossible, this comparison provides a relative scale for comparison. Host cell nucleus forms a cap, in an irregular manner. Sometimes very thick, other times stretched considerably around perimeter of parasite. Host cell nucleus generally covers 40% of the perimeter of parasite, ranging from 30-53%. Host cell cytoplasm is commonly recognizable within the host cell-parasite complex, occupying 3.4-38.5%. Microgametocyte: (n=14) The male gametocyte has the same characteristics as described for the macrogametocyte, with the usual gender-related staining differences. The host-parasite complex is slightly smaller than seen in the macrogametocyte, with an average area of 379 m 2 only 2.6 times the area of uninfected erythrocytes. It stains lightly with Giemsa, and nucleus not always discernible. When visible, the nucleus is diffuse with average area of 86.7 m 2

PAGE 35

26 Taxonomic Summary Type host: Rufous-headed Ground-roller ( Atelornis crossleyi ), Sharpe 1875; Scaly Ground-roller ( Geobiastes squamigerus ), Lafresnaye, 1838, Brachypteraciidae Type locality: Anjanaharibe-Sud (14 44.8S, 49 26.0E) and Andohahela (24.6S, 46.9E), Madagascar Basis of description: Parasites are described from a blood smear from an adult Atelornis crossleyi (Rufous-headed Ground-roller). HAPANTOTYPE: Blood smear SG-150 collected by Steven M. Goodman on 27 November 1994 at 1950 meters in Anjanaharibe-Sud, Madagascar. Accession G463734, IRCAH. PARAHAPANTOTYPE: Blood smear SG-505A from a Geobiastes squamigerus (Scaly Ground-roller) collected by Steven M. Goodman on 21 October 1995 at 400 meters in Andohahela, Madagascar. Accession IPM-12 to LInstitut de Pasteur de Madagascar. Additional hosts: Atelornis pittoides (Pitta-like Ground-roller) Distribution: It is presumed that this parasite will be found throughout the range of the ground-rollers on Madagascar. Remarks Leucocytozoon frasci is described from two birds of different species, and is the only species known from Brachypteraciidae. Some measurements have a wide range within the same individual. Comparing the proportions and measurements of the host-parasite complex and uninfected erythrocytes between the two birds, minor variation was observed, but no outstanding differences. The most noticeable trend was that the host parasite complex was slightly larger (2.9 times larger than uninfected erythrocyte) in the Scaly Ground-roller than in Rufous-headed Ground-roller (2.6 times larger). The uninfected erythrocytes in both species were the same size.

PAGE 36

27 Etymology This parasite is named in recognition of Dr. Salvatore Frasca Jr., a former mentor of the author for his support and guidance, and introducing her to parasitology. Leucocytozoon lairdi n. sp. Macrogametocyte: (n=23) Table 3-2, Figures 3-2(2), 3-2(3) Host cell type could not be determined due to the distortion by the parasite and there were no young parasites observed, so measurements are relative to those of an uninfected erythrocyte. Only round morph seen, staining blue with Giemsa. Parasite nucleus stains uniformly pink and generally round with area of 13.7 m 2 Host parasite complex is approximately 313.7 m 2 which is 2.5 times greater than an uninfected erythrocyte. The gametocyte encompasses 68.5% of the host parasite complex, and is rather uniform in diameter, averaging 19 m, (18-21 m.) The host cell nucleus is stretched into a band, forming an irregular cap covering 42% of the parasite perimeter. Cap not stretched excessively. It is relatively think and can be smooth or irregularly shaped along the free edge. A band is sometimes seen lying over the top of the gametocyte. Orientation of parasite can make this appear to be a split nucleus, but upon closer examination it can be seen that the host cell nucleus is lying underneath the gametocyte. Microgameotcyte: (n=15) Figure 3-2(4) Same morphological characteristics described above, with the expected gender related differences. Stains light pink to pink with Giemsa. Average area was 285.5 m 2 and nucleus varies in size, averaging 48.2 m 2 Parasite nucleus makes up 22.4% of the parasite. Taxonomic Summary Type host: Blue Vanga ( Cyanolanius madagascarinus ), Linnaeus, 1766; Helmetbird ( Euryceros prevostii ), Lesson, 1830, Vangidae.

PAGE 37

28 Type locale: Andohahela (24.6S, 46.3E) and Marojejy (14 25.6S, 49 36.5E), Madagascar. Basis of description: Parasites are described from a blood smear from an adult Cyanolanius madagascarinus HAPANTOTYPE: Slide SG-551A collected by Steven M. Goodman in Andohahela, Madagascar, 3 November 1995. Submission G463731, IRCAH. PARAHAPANTOTYPE: Slide MJR148 collected by Steven M. Goodman in Marojejy, Madagascar, 15 October 2001. Accession IPM-13 to LInstitut de Pasteur de Madagascar. Distribution: It is presumed this parasite will be found throughout the range of the Vangas on Madagascar, and possibly Grand Comoro and Moheli Island. Additional Hosts: Leptopterus viridis (White-headed Vanga), Euryceros prevostii (Helmetbird), Tylas eduardi (Tylas). Remarks This is the only species of Leucocytozoon known from the family Vangidae. The macroand microgametocytes of Leucocytozoon lairdi are described from two separate birds, as no intact microgametocytes were observed on one slide, while the quality of the other prevented full description of the parasite. It is felt that the small variation in measurements is a result of host variation, and perhaps representative of gender difference in this species. Etymology This parasite is named in honor of Marshall Laird, for his significant contributions to parasitology.

PAGE 38

29 Leucocytozoon greineri n. sp. Immature gametocyte: Displaces the host cell nucleus almost immediately, seen to develop in erythrocytes. Macrogametocyte: (n=33) Table 3-2, Figures 3-2(5), 3-2(6) Stains dark blue with pink nucleus with Giemsa stain. Round morph most commonly observed, although distorted forms present. Host parasite complex is 1.8 times larger than that of an uninfected erythrocyte with an average area of 224.4 m 2 Parasite occupies 61.6% of the host-parasite complex. The parasite has an average diameter (taken at widest point where relevant) of 16.6 m. The parasite nucleus is compact and round, having an area of 12.8 m 2 This is 9.3 % of the parasite. The host cell nucleus is stretched into a cap covering 38% of the circumference of the parasite. The cap size ranges, but usually well within 27-56%. Typically, the cap is thick with a smooth margin somewhat rounded ends. Host cell cytoplasm is commonly (50% of the time) associated with the complex. When present it makes up 10.4% of the host-cell complex, 21.2 m 2 is the average. Microgametocyte: (n=19) Same as described above, with gender staining associated differences. Stains lightly with Giemsa, usually appearing light pink. Microgametocyte nucleus is pale, sometimes with portions staining darker pink. Nucleus may be too indistinct to visualize. Nucleus has an average area of 40.8 m 2 over three times that of the macrogametocyte. This is also 29.5% of the microgametocyte. Taxonomic Summary Type host: Common Sunbird-Asity ( Neodrepanis coruscans ), Sharpe, 1875, Philepittidae. Type locale: Andohahela, Madagascar, 24.6S, 46.3E

PAGE 39

30 Basis of description: Parasites are described from a blood smear taken from an adult male Common Sunbird-Asity HAPANTOYPE: Blood smear SG-551C Neodrepanis coruscans collected by Steven M. Goodman on 3 November 1995 in Andohahela, Madagascar at 810 meters. Submission G463735, IRCAH. Distribution: It is presumed that this parasite will be found throughout the range of the Asities on Madagascar. Additional Hosts: Velvet Asity ( Philepitta castanea ) Remarks The unnamed Leucocytozoon sp. reported by Greiner et al. (1996) has been re-examined and is the same as described here. Etymology This parasite is named in recognition of Ellis C. Greiner, for his contributions to veterinary parasitology and the current knowledge of avian hematozoa.

PAGE 40

31 Table 3-1. Morphometric variation in the haemoproteids from the Brachypteraciidae, Vangidae, and Dicruridae. All measurements in microns or microns 2 H. goodmani H. forresteri H. vangii H. khani H. dicruri Host Family B rachypteraciidae B rachypteraciidae V angidae D icruridae D icruridae Infected RBC n= 34 20 48 20 12 HPC Area 140.9 (24.8) 120.3 (22.6) 124.6 (12.6) 130.3 (15.2) 128 (15.8) HPC length 18.8 (1.5) 16.2 (1.5) 18.3 (1.4) 16.2 (1.6) 15.9 (1.7) HPC width 10.9 (1.2) 10.0 (1.6) 9.6 (1.2) 9.7 (0.9) 9.8 (1.5) Parasite Area 68.97 (18.7) 76.5 (11.2) 60.9 (7.1) 110.6 (15.9) 106.3 (17.8) Parasite Length 19.1 (2.2) 21.8 (4.3) 16.4 (1.7) 28.11 (1.3) 16.1 (1.73) Macrogametocyte nucleus area 8.0 (2.6) 12.2 (12.3) 6.2 (3.1) 6.3 (1.2) 8.1 (2.9) Microgametocyte nucleus area 22.0 (3.5) 16.9 (7.5) 15.7 (9.98) 27.6 (14.6) not visible RBC nucleus length 7.3 (1.1) 6.8 (0.8) 7.3 (0.6) 6.7 (0.9) 6.7 (0.7) RBC nucleus width 4.1 (0.6) 4.7 (0.7) 3.5 (0.6) 3.4 (0.5) 3.5 (0.7) RCB nucleus area 23.97 (5.4) 27.1 (5.1) 20.8 (3.4) 19.7 (1.7) 19.38 (2.98) NDR 0.66 (0.3) 0.51 (0.4) 0.71 (0.3) 0.66 (0.3) 0.01 (0.05)

PAGE 41

Table 3-1 continued H. goodmani H. forresteri H. vangii H. khani H. dicruri Host Family B rachypteraciidae B rachypteraciidae V angidae D icruridae D icruridae Uninfected RBC n= 25 22 29 15 15 Length 16.2 (1.1) 15.0 (1.5) 15.8 (3.1) 15.4 (1.1) 15.4 (1.1) Width 10.2 (0.9) 12.4 (2.2) 9.5 (0.8) 9.7 (0.6) 9.7 (0.6) Area 130.1 (13.9) 142.7 (32.1) 118.4 (9.3) 120.4 (8.7) 120.4 (8.7) Nucleus length 7.4 (0.9) 7.1 (0.7) 7.4 (0.7) 7.1 (0.7) 7.1 (0.7) Nucleus width 3.9 (0.5) 4.6 (0.9) 3.6 (0.6) 3.7 (0.6) 3.7 (0.6) 32 HPC= host cell-parasite complex, NDR= nucleus displacement ratio

PAGE 42

33 Table 3-2. Morphometric variations of the Leucocytozoon spp. of the Brachypteraciidae, Vangidae, and Philepittidae. All measurements in microns or microns 2 L. frasci L. lairdi L. greineri Host Family B rachypteraciidae V angidae D icruridae Infected RBC N= 39 38 53 Host cell-parasite complex area --224.4 (49.7) Macrogametocyte host parasite complex 424.2 (85.1) 313.7 (48.7) -Microgametocyte host parasite complex 379.1 (135.0) 285.5 (48.6) -Parasite area 268.2 (99.8) 215.0 (42.0) 138.2 (26.8) Macrogametocyte nucleus area 16.3 (5.2) 13.7 (5.7) 12.8 (3.8) Microgametocyte nucleus area 86.7 (113.7) 48.2 (33.6) 40.8 (23.2) Host cell nucleus area 88.98 (27.4) 61.6 (19.9) 83.9 (20.6) Residual host cell cytoplasm --21.2 (19.97) Nuclear cap ratio 0.395 (0.1) 0.42 (0.1) 0.38 (0.09) Uninfected RBC N= 55 44 41 length 16.4 (1.3) 15.8 (3.1) 15.9 (1.0) width 11.3 (1.1) 9.5 (0.8) 9.7 (0.8) area 143.5 (15.6) 126.1 (14.99) 124.5 (11.3) nucleus length 7.4 (0.6) 7.4 (0.7) 8.1 (0.7) nucleus width 4.8 (0.7) 3.6 (0.6) 3.9 (0.5) nucleus area 27.3 (3.6) 23.1 (3.23) 27.4 (3.1)

PAGE 43

34 Figure 3-1. Haemoproteus spp. of the Brachypteraciidae, Vangidae, and Dicruridae. (1,2) H goodmani macrogametocytes, (3) H forresteri macrogametocyte, (4) H forresteri microgametocyte, (5) H vangii microgametocyte, (6) H. vangii macrogametocyte, (7) H khani macrogametocyte, (8) H khani microgametocyte, (9) H dicruri macrogametocyte.

PAGE 44

35 Figure 3-2. Leucocytozoon spp. of the Brachypteraciidae, Vangidae, and Philepittidae. (1) L. frasci macrogametocyte, (2,3) L. lairdi macrogametocytes, (4) L. lairdi microgametocyte, (5,6) L. greineri macrogametocytes.

PAGE 45

CHAPTER 4 SURVEY RESULTS One quarter (24.3%) of the 378 birds surveyed was parasitized by at least one species of hematozoa. Seventeen of the 21 families had infected birds. Prevalence by family was statistically different for families with at least one infected bird. At the host species level, 32 of 43 species had at least one infected representative (Table 4-1). All specimens of two species were infected by the same parasite respectively. For example, all seven Accipiter francesii were infected with Leucocytozoon toddi and 10 of 10 Ploceus sakalava were infected with Haemoproteus quela In families where at least 10 individuals were sampled, the Zosteropidae and Ploceidae were the most parasitized families, with a prevalence of 56% and 34% respectively. With reference to the prevalence of the parasite genera, Haemoproteus spp. were present in 14%, Leucocytozoon spp. in 11%, microfilariae in 6%, Plasmodium spp. in 2%, Trypanosoma spp. in 1.1%, and unidentified Babesia sp. in 0.5%. Microfilariae in birds rarely have been identified to the species, or even genus level, and these are the motile embryos of filaroid nematodes. All of the others are single celled eukaryotic parasites. Prevalence by host family There was a significant (P < 0.001) difference in prevalence of parasitism by family when all families were compared, and when families with ten or more birds sampled were compared. The following parasites were identified: ACCIPITRIDAE: Each of the seven hawks sampled were infected with Leucocytozoon toddi but no other infections were seen. 36

PAGE 46

37 ALCEDINIDAE: One of eight kingfishers was parasitized with Haemoproteus halcyonis BRACHYPTERACIIDAE: Seven of nine ground-rollers sampled were infected with at least one species of hematozoa. Four birds had multiple infections. Five birds were infected with Haemoproteus spp.; three with H. forresteri two with H. goodmani Leucocytozoon frasci is reported from five birds. Trypanosoma avium was present in one bird. Microfilariae from an unidentified nematode were observed from one bird. This is the first description of hematozoa from this family, and each species listed here are new taxa, specific to the Brachypteraciidae. DICRURIDAE: Two species of Haemoproteus H. khani and H. dicruri were observed in three drongos. Additionally, microfilariae were observed in two birds. Haemoproteus khani a new species of haemoproteid in the drongos was described. Trypanosoma avium was recorded from one bird. MONARCHIIDAE: Two of forty-eight monarch flycatchers appeared to be infected with an unidentified species of Babesia No other hematozoa were observed in any of the other individuals sampled. MOTACILLIDAE: Two of three wagtails were infected with Haemoproteus anthi NECTARIINIDAE: One sunbird was infected with Haemoproteus sequeirae PHILEPITTIDAE: Four asities were infected with Leucocytozoon greineri and microfilariae were observed in five birds. This is the first species of the Haemosporida reported from this family, and represents a new species. PLOCEIDAE: Three species of Plasmodium were seen in three weavers. Two were infected with P. rouxi while a third was infected with both P. nucleophilum and P. relictum The same species of Haemoproteus H. quelea was observed in 15 birds. Six

PAGE 47

38 birds were infected with Leucocytozoon bouffardi Two birds were infected with microfilariae. A single Trypanosoma everetti was observed in one bird. PYCNONOTIDAE: Plasmodium rouxi was observed in one bulbul. Two species of Haemoproteus were observed, H. sanguineus was observed in each of the three birds infected with this genus. One bird had a concurrent infection with H. philippinensis Two species of Leucocytozoon were observed. Three bulbuls were infected with L brimonti and two more with L pycnonoti Three birds were observed to have microfilariae. STRIGIDAE: Hematozoa were observed in three of nine owls. Haemoproteus noctuae and H. srynii were each observed in one bird, while the haemoproteid in a third was too distorted for conclusive identification. Two birds were infected with L. ziemanni SYLVIIDAE: Four Old-world warblers were infected with H. wenyoni Leucocytozoon phylloscopus was observed in one bird. Microfilariae were observed in two birds. TIMALIIDAE: One babbler was infected with microfilariae. TURDIDAE: Schizonts of Plasmodium vaughani were observed in one thrush, and two others were infected with Plasmodium rouxi Haemoproteus minutus was reported from three birds, H. fallisi from two others, and distortion prevented the positive identification of two additional infections (one mixed with a H. fallisi infection). One bird was infected with Leucocytozoon majoris Microfilariae were observed in three birds. VANGIDAE: Four of five vangas were infected, each of which with multiple infections. One bird was infected with Plasmodium rouxi Two birds were infected with Haemoproteus vangii The three birds infected with Leucocytozoon sp. each harbored L lairdi Three birds were infected with microfilariae and two more were infected with T

PAGE 48

39 avium This represents the first examination of this family for the presence of hematozoa, and the Haemoproteus and Leucocytozoon species mentioned here represent new taxa. ZOSTEROPIDAE: Only two genera of hematozoa were observed. Eight White-eyes were infected with H. zosteropis and six had L. zosteropis Altitude When prevalence was compared with altitude of bird collection, (120 m, 400 m, 800 m, 1000 m, 1200 m, 1400 m, 1500 m, 1875 m, 1950 m), there was a significant (P<0.001) difference between groups. Pair-wise comparisons revealed three groups that were different. At 120 m, prevalence was 75% with 15 of 20 birds parasitized. At the highest elevation (1950 m) prevalence was 57%, with 4 of 7 birds parasitized. Finally, the third group (1200 m) had a low level of parasitism, with 2 of 32 birds (6%) infected. Because of the small sample sizes, the altitudes were consolidated into three ranges and re-examined. The ranges were low altitude (120-800 m), medium altitude (1000-1400 m) and high altitude (1500-1950 m). Prevalence was 36%, 19% and 22% respectively. These ranges were then compared, and were found to differ from one another, so once again pair-wise comparisons were performed. The low altitude group was found to be statistically different from the others. Reserves The four regions from which the birds were sampled were compared to determine if the prevalence of parasitism was greater in any of them compared to the others. When all four were compared, a statistical difference was observed (P = 0.042). Prevalence for Ambohitantely, Anjanaharibe-Sud, Andohahela, and Montagne dAmbre were 21% (26/124), 21% (17/82), 34% (44/131) and 20% (8/41) respectively. Regions were then

PAGE 49

40 compared in pairs. Andohahela was statistically different from Ambohitantely (P = 0.023). When compared to Anjanaharibe-Sud, the p-value was 0.062. Compared to Montagne dAmbre, it appeared that there was no statistical difference (P = 0.129), but the power of the test was very low (0.311) indicating a 69% chance that no difference will be found when in fact there was a difference between the groups. No statistical significance was observed between the other groups. Habitat Birds were classified into three groups: FED, FDE, and FDW. FED species are forest edge species, living at the junction of forest and savanna habitats, and will cross open, non-forested areas. FDE species are forest dependent species, and probably never cross non-forested areas. Finally, FDW species are the forest dwelling species. These can be found in the center of a forest, or in the margin between forest and savanna, and almost certainly cross open areas. The majority of the species sampled were either forest dwelling or forest dependent species. Prevalence for each habitat was as follows: FED 31%, FDW 27%, and FDE 19%. Groups were compared to determine any trends in parasitism. There appeared to be no difference in prevalence observed in our study. Gender A total of 76 females and 111 males were conclusively identified. No gender was recorded for birds where plumage was not indicative of gender, or birds that escaped before the data could be obtained. There was no significant difference in prevalence of parasitism between males and females.

PAGE 50

41 Age Birds were classified as either adult or juvenile by field ornithologists. Prevalence by age was 23% for adults (65/284) and 8% (2/26) for juveniles. No significant difference was found in prevalence of parasitism between the two cohorts. Breeding Condition For the purposes of this survey, birds were considered to be in breeding condition when a brood patch was observed, indicating the birds were actively incubating eggs. Seventy-seven birds had brood patches, while 288 did not. Prevalence of parasitism was 27% and 23% for each group, respectively. There was no significant difference in parasitism between groups.

PAGE 51

42 Table 4-1. Prevalence of hematozoa in the avifauna of Madagascar, by avian family. Family & species Total birds Infected birds Plas. Haem. Leuc. Micro Tryp Babesia Accipitridae Accipiter francesii 7 7 7 Alcedinidae Alcedo vintsiodes 2 0 Ipsidina madagascariensis 6 1 1 Brachypteraciidae Atelornis pittoides 5 3 3 2 Atelornis crossleyi 3 3 2 2 Geobiastes squam igerus 1 1 1 1 Campephagidae Coracina cinerea 1 0 Columbidae Streptopelia picturata 2 0 Cuculidae Cuculus rochii 1 1 1 Dicruridae Dicrurus forficatus 8 4 3 2 1 Leptosomidae Leptosomus discolor 1 0 Monarchiidae Terpsiphone m utata 48 2 2 Total 48 2 2 % infected 4.2 4.2 Motacillidae Motacilla flaviventris 3 2 2 Nectariniidae Nectarina notata 1 0 Nectarina souimanga 9 1 1 Total 10 1 1 % infected 10 10

PAGE 52

43 Table 4-1 continued Family & species Total birds Infected birds Plas. Haem. Leuc. Micro Tryp Babesia Philepittidae Neodrepanis coruscans 3 1 1 Neodrepanis hypoxantha 1 0 Philepitta castanea 22 4 3 5 Total 26 5 4 5 % infected 19.2 15.4 19.2 Ploceidae Foudia madagascariensis 8 2 1 1 Foudia omissa 30 6 4 4 Ploceus nelicourvi 8 1 1 Ploceus sakalava 10 10 2 10 1 2 1 Total 56 19 3 15 6 2 1 % infected 33.9 5.4 26.8 10.7 3.6 1.8 Pycnonotidae Hypsipetes madagascariensis 23 7 1 2 3 1 Xanthornixis cinereiceps 8 0 Bernieria madagascariensis 25 4 1 2 2 Xanthornixis zosterops 22 0 Total 78 11 1 3 5 3 % infected 14.1 1.3 3.9 6.4 3.9 Strigidae Otus rutilus 9 4 3 2 Sylviidae Dromaeocercus brunneus 1 1 1 Harterula flavoviridis 1 0 Nesillas typica 15 3 2 1 Newtonia amphichroa 11 1 1 1 Newtonia brunneicauda 11 1 1 Total 39 5 4 1 2 % infected 12.8 10.3 2.6 5.1 Threskiornithidae Lophotibis cristata 1 0 Timaliidae Mystacornis crossleyi 1 1 1 Oxylabes madagascariensis 10 Total 11 1 1 % infected 9.1 9.1

PAGE 53

44 Table 4-1 continued Family & species Total birds Infected birds Plas. Haem. Leuc. Micro Tryp Babesia Turdidae Copsychus albospecularis 15 3 2 2 Monticola sharpei 22 5 3 3 1 1 Total 37 8 3 5 1 3 % infected 21.6 8.1 13.5 2.7 8.1 Vangidae Cyanolanius m adagascarinus 1 1 1 1 1 1 Tylas eduardi 2 2 1 1 2 1 Vanga curvirostris 1 0 Leptoperus viridis 1 1 1 1 Zosteropidae Zosterops maderaspatana 16 9 8 6 Total 16 9 8 6 % infected 56.3 50 37.5 TOTAL 378 92 6 52 40 23 4 2 % 24.3 1.6 13.8 11 6.1 1.1 0.5

PAGE 54

CHAPTER 5 DISCUSSION There have been many studies to examine the prevalence of avian hematozoa throughout the world. In the larger surveys, researchers have examined birds of the Neotropics (White et al., 1978), North America (Greiner et al., 1975), eastern and southeastern Asia (McClure et al., 1978), and central Europe (Kucera, 1981). Overall prevalence varied considerably in these areas, from 10.5% in the Neotropics, to 36.9% in North America. In each of these surveys, the most commonly observed genus of hematozoa was Haemoproteus A great deal of work has also been done in Africa, India and island groups in these regions (de Mello, 1935a; de Mello and Fonseca, 1937; Bennett et al., 1974; Bennett and Herman, 1976; Bennett et al. 1978; Peirce et al. 1977; Peirce and Feare, 1978; Peirce, 1979; Peirce, 1984a; Peirce, 1984b; Bennett, 1990). This depth of knowledge and accumulation of information over time is invaluable to current and future investigations in this region. Many new species have been described in this region, and others redescribed over time, owing to the diversity of the birds throughout Africa and India. In 1990, Bennett reported 78 valid species of Haemoproteus Leucocytozoon and Plasmodium from the Indian subcontinent alone. While a great deal of work has been done on the continent of Africa, relatively little work has been done in Madagascar, a major continental island off the coast of Mozambique. Three papers examining the avifauna for hematozoa have been published to date. Bennett and Blancou (1974) examined 64 individuals from thirty-two species. Of those, 14 (22%) birds were infected. Interestingly, Bennett and Blancou did not 45

PAGE 55

46 observe any species of Haemoproteus rather, Leucocytozoon was the most commonly observed genus of blood parasite. Bennett and Blancou (1974) concluded that the prevalence of hematozoa on the island was low, and that there was no indication of unique species of hematozoa. The results of this survey indicate that these conclusions are not the case, and that the prevalence of avian hematozoa falls within the range reported elsewhere. Additionally, the endemic families and species were seen to harbor new and distinct species of Haemoproteus and Leucocytozoon Bennett and Blancou (1974) were unable to draw conclusions on parasitism by family as a result of few birds being sampled from each family. Greiner et al. (1996) reported six of 10 birds sampled from southeastern Madagascar to be parasitized. In addition to the higher overall prevalence, they also reported the presence of the genus Haemoproteus for the first time, in addition to Plasmodium Leucocytozoon and microfilariae. To have 60% of the birds parasitized is a high number for such a small group, but the presence of four types of hematozoa observed also speaks to the richness of all the forms of life in Madagascar. Parasites in that study were identified only to the genus level. Most recently, Raharimanga et al. (2002) examined 387 Malagasy birds and reported a prevalence of 35.9% overall. They sampled birds in 1995, 1996, and 2001 at six different sites. Again, parasites were not identified to the species level. In this study, birds were collected in the years 1994 and 1995, and four different sites were sampled. Additionally, parasites were identified to the species level, providing a clearer understanding on the hematozoa in Madagascar. These investigations have been important as a means of establishing the presence of avian hematozoa in multiple species, and over several years. The current investigation

PAGE 56

47 represents the first large-scale survey of the hematozoa of the avifauna on Madagascar including identification of the parasites observed. This information will be vital to future investigations in Madagascar, and for more complete understanding of the biology of the avifauna there. This survey was intended to investigate the prevalence of hematozoa by host family, and by several other parameters including age, gender, altitude, sampling site, and habitat. Breeding condition (classified as here by the presence or absence of a brood patch) was also considered. All birds included here were wild caught, none raised or held in captivity. Additionally, the birds sampled here are from non-migrating, resident populations on the island. One quarter (24.5%) of the birds surveyed was parasitized by at least one species of hematozoa. As mentioned earlier, reported prevalence in Madagascar has varied widely, from 22% (Bennett and Blancou, 1974), to 60% (Greiner et al., 1996) and 39.5% (Raharimanga et al., 2002). Prevalence in nearby areas varied greatly as well. For instance, in the Mascarene Islands, Peirce (1979) reported 42% and 27% of the birds were parasitized, while in Zambia he reported a prevalence of 48.2%, and finally he reported a 12.7% prevalence through Africa (Peirce, 1984a; Peirce, 1984b). Bennett et al. (1974) reported 24% of the birds examined in Uganda were parasitized, while in Senegal only 11.5% were parasitized (Bennett et al., 1978). In our study, 17 of the 21 families had infected birds. One or two birds only represented each of the four families that was not seen to be infected. It is likely that with a larger sample size, parasites would have been observed in these families, and certainly further investigation is warranted. When compared, a statistical difference in

PAGE 57

48 parasitism by avian family was found for the families that had at least one infected bird. Further, there was a statistical difference in parasitism for families with at least 10 birds sampled. The Zosteropidae and Ploceidae were the most parasitized, with a prevalence of 56.3% and 33.9% respectively. These families were also the most parasitized in Ugandan birds, as reported by Bennett et al. (1974). Other surveys have also reported a variation in prevalence by avian family (Greiner et al., 1975; Bennett and Herman, 1976, White et al. 1978). Variation by family or species may be a result of preferential feeding by the vectors. Two studies have commented that some birds appeared to be more attractive than others (Bennett and Fallis, 1960; Greiner et al., 1978). Prevalence of parasite genera was also compared by avian family, for families with twenty or more birds sampled. Two of these families, Monarchiidae and Philepittidae, were not seen to be parasitized with species of Haemoproteus There was a statistical difference in prevalence of Haemoproteus spp. both with and without these families included. Pair-wise comparisons revealed that two families were statistically different from one another, the Ploceidae (26.8%) and Pycnonotidae (3.8%). Neither of these families was statistically different from the remaining families in the comparison. There was no difference in parasitism by the other genera of parasite by family. On the species level, 32 of 43 species had at least one infected representative. Some of these families were also represented in prior work. Bennett and Blancou (1974) also examined two Streptopelia picturata and like this survey, no parasites were seen in this species. They examined 15 Fodys ( Foudia madagascariensis ) in 1974, and 4 were infected with Leucocytozoon fringillinarum Raharimanga et al. (2002) reported the

PAGE 58

49 presence of either Plasmodium or Haemoproteus (not evaluated separately) or Leucocytozoon in several of the same species examined here. The prevalence of the parasite genera was similar to that reported elsewhere, with species of Haemoproteus being the most commonly seen genus. Families Prevalence was not analyzed for families represented by fewer than 10 samples. Three families included in this survey are endemic to Madagascar. These are Brachypteraciidae, Philepittidae, and Vangidae. The Brachypteraciidae, or Ground-rollers, were primarily parasitized by Haemoproteus and Leucocytozoon species. Additionally, microfilariae were seen in one individual. Plasmodium spp. were not observed. Raharimanga et al. (2002) examined five birds and reported the presence of the same hematozoa. Four of their five birds were infected with either Plasmodium or Haemoproteus one with Leucocytozoon one with Trypanosoma avium and two with microfilariae. In the Vangidae, four of five birds were parasitized and all had multiple infections. Interestingly, species of Plasmodium Haemoproteus Leucocytozoon microfilariae and Trypanosoma were observed in these birds. In previous studies, Greiner et al. (1996), reported the one Vanga examined to be infected with both a species of Leucocytozoon and microfilariae. Out of 11 Vangas, Raharimanga et al. (2002) reported four infections each of Plasmodium or Haemoproteus Leucocytozoon and microfilariae. While the samples sizes discussed are too small to make inferences with regard to parasitism in the population, there is certainly a trend indicating high levels of parasitism warranting further investigation. Finally, Philepittidae, the Asities, are small birds endemic to Madagascar. There are four species; three have been included in this sample of 26 individuals. The prevalence of parasitism is 19.2%. The same small

PAGE 59

50 Leucocytozoon was noted repeatedly. Microfilariae were present in several individuals. Because Plasmodium is typically rarely seen, it is not surprising that none were observed here. It is interesting that no haemoproteids were noted. The genus Haemoproteus is typically the most common hematozoon observed in surveys throughout the world. The presence of Leucocytozoon greineri indicated that the family was susceptible to parasitism. Greiner et al. (1996) reported the presence of an unidentified species of Leucocytozoon and microfilariae. This slide was re-examined and the Leucocytozoon was recognized to be the same species reported here and is considered L. greineri The lack of infection with haemoproteids warrants further investigation. Because it is an endemic family and not present outside of Madagascar, it is possible that no species of Haemoproteus has evolved to fill this niche. It is also possible that no appropriate vector exists in the locales sampled. The parasites found in these three families represent new taxa, and for the first time species names have been designated. This survey, then, represents the first investigation of parasitism in these families to the parasite species level. The Monarchiidae examined in this survey are represented by one of the two species of Flycatchers on Madagascar. Forty-eight Madagascar Paradise Flycatchers ( Terpsiphone mutata ) were sampled and examined for the presence of hematozoa. These birds are common in all forest sites, seen in degraded areas and in open areas, and have a wide altitudinal range (sea level to 1600 m) (S.M. Goodman, pers. comm.; Morris and Hawkins, 1998). Because of the wide range of habitats and elevations in which these birds were sampled, one might expect that at least a portion of the birds would exhibit an infection of some type. Interestingly, none of the birds sampled here was parasitized by

PAGE 60

51 any of the expected genera. What appeared to be an undescribed Babesia species was observed in two birds, but further examination of this family is needed to fully appreciate this phenomenon, as ticks were noticeably absent from birds collected (S.M. Goodman, pers. comm.). Based on the parasitism seen in other families, it was known that ornithophilic vectors do exist on the island, at the same ranges. Greiner et al. (1996) reported the two T. mutata included in their sample to be infected. The schizont of a species of Plasmodium was observed in one, while the other was infected with either Plasmodium or Haemoproteus based on seeing only pigmented trophozoites. Raharimanga et al. (2002) reported that three of 20 members of this family sampled were infected with a species of either Plasmodium or Haemoproteus The authors did not observe any erythrocytic schizonts, all of the infected birds were T. mutata adults (V. Robert, pers. comm.). Based on this information, it appears that a species of Haemoproteus infective to the Flycatchers is present in Madagascar. Further examination of this family is warranted, to determine the if the parasites are similar to those reported for this family from other locations, or if the species is new and possible only found in the Malagasy Flycatchers. Members of the same avian genus sampled in other geographical areas were parasitized. Peirce (1984) reported one of seven Terpsiphone viridis examined was infected with a Haemoproteus sp., one of two T. viridis and one of seven Batis molitors were infected with Plasmodium in Zambia. Based on these reports, we can conclude that the family is susceptible to parasitism. As yet, it was not apparent that this species exhibited any behavioral adaptations that would preclude infection. In considering the difference between the results of our survey and of the 2002 project, it is likely that the vector for the haemoproteid reported by Raharimanga et al. (2002) is not

PAGE 61

52 found throughout Madagascar, but more likely in isolated areas. It is also possible that there were chronic infections existing at a subclinical level in the birds sampled for our survey. These infections can result in a very low-level parasitemia that may go undetected on microscopic examination. Two species of thrushes (Turdidae) were represented in this survey, with a total of 37 birds. The family had a prevalence of 21.6% overall for parasitism, and had the highest prevalence of Plasmodium with 3 of 37 (8%) birds infected with species in this genus. Schizonts consistent with Plasmodium vaughani were observed in one bird and P relictum -like gametocytes were noted in two others. Developing schizonts were seen, but were not useable for diagnosis. Haemoproteus spp. were the most common parasite in the family, with both H. minutus and H. fallisi recognized. Leucocytozoon majoris and microfilariae were also noted. No parasites were seen in the single thrush examined by Bennett and Blancou (1974). Raharimanga et al. (2002) examined blood smears from 45 individuals, and although no Leucocytozoon species were observed, 14 (31%) were infected with either Plasmodium or Haemoproteus spp., and seven had microfilariae circulating at the time of sampling. The family with the highest prevalence in this survey was the Zosteropidae, or the White-eyes. Of the 16 birds sampled, nine (56.3%) were infected with at least one species of either Haemoproteus or Leucocytozoon Five birds were infected with both genera, and the two species of parasite observed were H. zosteropis and L. zosteropis The prevalence of both Haemoproteus (50%) and Leucocytozoon (37.5%) are similar to those reported by Raharimanga et al., (2002). They reported 64% (11/17) birds were infected with either Plasmodium or Haemoproteus and 36% were infected with

PAGE 62

53 Leucocytozoon sp. Additionally, they observed birds were infected with microfilariae. Although Bennett reported this family to be highly parasitized in Uganda, only 15 birds were included in his sample, and again, further investigation is necessary (Bennett et al., 1974). All three sample sizes are small, but earlier work on Aldabra Atoll examined a slightly larger group of Malagasy White-eyes reported 38 of 48 individuals infected with elongated gametocytes of a Haemoproteus sp. Also, 14 of these birds were infected with erythrocytic schizonts. Definitive identification was not possible as no gametocytes were seen (Lowery, 1971). Finally, in the Mascarene Islands, Peirce et al. (1977) sampled a total of 100 members of this family, 74 were infected with hematozoa. Peirce (1979) reported four of four White-eyes to be infected with Leucocytozoon, after reviewing a another sample from the same area. Further examination of this family in Madagascar will be important in determining if the prevalence remains as high with more animals examined. The weaver family, Ploceidae, was the second most parasitized family in this survey, with a prevalence of 33.9% in 56 birds. Four species were represented, with 8-30 birds from each. Multiple birds were infected with Plasmodium Haemoproteus Leucocytozoon and microfilariae. Plasmodium rouxi P. nucleophilum and P. relictum were observed, in addition to H. quelea and L. bouffardi Trypanosoma everetti is known to infect the Ploceidae, and was observed in a single Ploceus sakalava (Bennett et al., 1994). Interestingly, each of the 10 P. sakalava sampled was infected with H. quelea while only five of the remaining 46 weavers were infected with it. One possible reason for this may be related to the behavior of this species of bird. The three other species sampled are solitary breeders, while P. sakalava forms dense colonies (S.M Goodman,

PAGE 63

54 pers. comm.). This has been seen in this family before. In a survey from Senegal, Bennett et al. (1978) reported the prevalence of parasitism in non-colonial nesting ploceids was 3.1%, but in the colonial nesting ploceids, prevalence was substantially higher at 21.3%. Each of the prior investigations in Madagascar reported parasitism in this family from each of the parasite genera observed here (Bennett and Blancou, 1974; Greiner et al., 1996; Raharimanga et al., 2002). Because Bennett and Blancou (1974) as well as Greiner et al. (1996) reported positive individuals from small samplings (15 and 2 birds, respectively), this supports our observation of the high prevalence of parasitism in this family. If the actual prevalence of parasitism in this family were low, then small sample sizes would be less likely to contain infected individuals. There are other indications that this family may be more susceptible to parasitism, either through their behavior, or greater abundance of the vectors. For example, Bennett et al. (1974) also reported that the ploceids were the second most parasitized family surveyed in Uganda. They detected at least one species of hematozoa in the 116 of 206 weavers examined. In the Mascarene Islands, 17% of the Madagascar Fodys were infected with either Leucocytozoon or Plasmodium For the Rodrigues Fody sampled at the same time, the prevalence was 31% (Peirce 1979). With 78 birds captured, the Pycnonotidae or Bulbuls, had the largest sample. Four species were obtained, and the overall prevalence for the family was 14%. Plasmodium rouxi H. sanguineus H. philippinensis L. brimonti L. pycnonoti and microfilariae were all observed (Table 3). As the prevalence observed here is relatively low, it should not be entirely surprising that none of 26 Hypsipetes madagascariensis examined on Aldabra Atoll were parasitized (Lowery, 1968). A single, uninfected individual was included in

PAGE 64

55 the Greiner et al. (1996) survey. Raharimanga et al., (2002) reported low prevalence of Plasmodium or Haemoproteus (7%) and Leucocytozoon (8%). They reported 13 of the 84 (15%) bulbuls sampled were infected with microfilariae. In Uganda, only 14 of 244 (6.3%) bulbuls were infected (Bennett, 1974). Haemoproteus wenyoni was the most common parasite observed in the warblers (Sylviidae), with three of the five species infected. It appeared that there might be two different species of Haemoproteus in these birds, but there was not enough material to confirm this. One bird was infected with Leucocytozoon phylloscopus and two with microfilariae. Raharimanga et al. (2002) also found this to be true, with 24% of the 45 birds they examined infected with either Plasmodium or Haemoproteus They also only found one bird to be infected with Leucocytozoon This may be an indication that the vectorhost relationship was not ideal for Leucocytozoon but was functional. Another possibility was that there was a moderate vector population, which did not serve to increase the prevalence, but rather maintained the level of parasitism by Leucocytozoon spp. Of the 11 babblers (Timaliidae) surveyed, only one was infected with the microfilariae from an unknown filaroid nematode. Raharimanga et al. (2002) reported a very low prevalence of parasitism in this family. Only one of 25 birds was infected with Plasmodium or Haemoproteus one with Leucocytozoon and 5 with microfilariae. Because of the small sample sizes, it is hard to draw conclusions with regard to parasitism in this family. The presence of the one infection indicated that members of this family do come into contact with a functional vector for their respective genus.

PAGE 65

56 In the sunbirds (Nectariinidae), one of 10 birds was infected with H. sequeirae Raharimanga et al. (2002) sampled fourteen sunbirds and found that more than half were infected with either Haemoproteus or Plasmodium and two with Leucocytozoon The difference in prevalence may be a result of sampling location, as in their investigation, all parasitized birds were taken from locations other than those included in this survey. This difference may also reflect a true low prevalence, or birds with chronic infections may not be being recognized as infected. The Dicruridae is present throughout Africa, and has been examined before for hematozoa. De Mello (1935b) described Haemoproteus dicruri later redescribed by Peirce (1984c). Until now, H. dicruri was the only haemoproteid known from this family. In our survey, the four birds infected with Haemoproteus were infected both with H. dicruri and in addition, H. khani a new haemoproteid from this family. Both species were present, at high intensities, in the four infected birds. This is the first examination of this family in Madagascar. Considering the unique flora and fauna of this island, it is not surprising that there are parasites that may be unique to Madagascar. Further sampling is needed to find pure infections of both haemoproteids, for the comparison of the blood stages and perhaps elucidation of the life cycles of these parasites. If no pure infections are found, the parasites may share the same vector, or perhaps infection with one may insult the immune system enough allowing a rise in parasitemia of a chronic infection of the other. Additional research may even indicate that the two parasites are the same species, but with two distinct forms of mature gametocyte. This is rarely seen, but at least two examples exist, H. sacharovyi in Columbidae, and H. forresteri from the Vangas.

PAGE 66

57 The remaining families were represented by a small sample sizes. Mentioned earlier, there was 100% infection of Accipitridae by Leucocytozoon toddi This suggests a highly successful vector inhabiting the same range as those birds sampled here. Perhaps the vector is very successful reproductively, or perhaps is well adapted to obtaining blood meals from this host. Bennett and Blancou (1974) reported the same species of Leucocytozoon in one of four Buteo brachypterus in 1964, while no members of this family were included in the other projects in the area. Nine owls were sampled and four were found to be infected with hematozoa ( H noctuae H. syrnii and L. ziemanni ). The presence of hematozoa in this family on Madagascar has been well established in earlier work. Greiner et al. (1996) reported a species of Haemoproteus in this family. Also, each of the six birds examined by Raharimanga et al. (2002) was infected with either Plasmodium or Haemoproteus one with Leucocytozoon and four with microfilariae. A small group of kingfishers (Alcedinidae) was examined, only one of which proved to be infected with H. halcyonis This is not unusual, as water birds can be less parasitized by hematozoa (Greiner et al., 1975). Raharimanga et al. (2002) examined fifteen kingfishers, and none were infected with any Apicomplexan parasites, although three were carrying microfilariae. On the other hand, Bennett et al. (1974) reported that each of the kingfishers examined in Uganda were parasitized by either H. enucleator or H. halcyonis Finally, two of three wagtails (Motacillidae) were infected with H. anthi The only previous work that included a wagtail was Greiner et al. (1996). They reported no

PAGE 67

58 parasites seen in the one individual examined. This, then represented the first record of hematozoon infection in this family in Madagascar. The remaining families: Threskiornithidae, Columbidae, Leptosomidae and Campephagidae were represented by one individual (2 in Columbidae) and no parasites were seen. These families will not be discussed. Vectors All of the parasites discussed here are vector borne parasites. Therefore, for infection to be maintained within the population, vectors and avian hosts must come into regular contact, the parasite must exist in a reservoir population for the vectors to become infected, and there have to be susceptible avian hosts for the parasite to complete its life cycle. In many ways, the search for the ornithophilic vectors of avian hematozoa in Madagascar will require starting from scratch. There are many aspects to vector biology that must be appreciated before successful investigation can be performed. Behavior, flight altitude, and altitudes at which the insects search for blood meals all need to be determined. Each of these factors varies substantially from one species of vector to another. For example, Pilaka and Elouard (1999) commented on the restriction that Simulium has to running water during the aquatic stage. Within this restriction, a preference for factors such as oxygen flow, pH, turbidity, temperature, and habitat all vary for the different species of black flies. Different insect families forage at different levels in the canopy. Bennett and Coombs (1975) found that most simuliids and ceratopogonids were captured from bait birds suspended 3-4 m off the ground, indicating a preference for feeding off the birds in the canopy. Also, ornithophilic flies are active at particular times of the day. For example, ornithophilic biting flies were collected most frequently in the evening crepuscular period (6-10 PM) in Newfoundland (Bennett and

PAGE 68

59 Coombs, 1975). Different species respond to different attractants, uropygial gland ether extracts, carbon dioxide or visual stimuli, that allow then to zero in on a potential blood meal (Fallis and Smith, 1965; Bennett et al., 1972). The simuliid flies that are arriving at sites to feed off the birds may actually not be very close to the sampling site, as Fallis and Bennett reported (1966) these flies were able to travel a distance of two miles in 5-12 days. On the other hand, it is the limited flight range of Culicoides spp, in addition to limited host dispersal and low winter temperatures, that was suggested to impact the transmission of H. meleagridis in northern Florida (Atkinson et al. 1988). Additionally, more than one species of arthropod may be capable of transmitting the parasite in question. For instance, at least five species of Culex mosquitoes have been shown to be either natural or experimental vectors of Plasmodium elongatum (Nayar et al., 1998). In Canada, L. simondi is transmitted by two species of simuliid, which are present at slightly different times, allowing for longer possible transmission of the parasite (Fallis and Bennett, 1966). Because the natural vector, Simuliim spp., of T. avium in Canada cannot be colonized, researchers were able to successfully use Aedes sp. in its place (Bennett, 1970). This, however, does not mean that any member of the same family of insects can successfully vector hematozoa. In 1990, Work et al. (1990) demonstrated that sporozoites of P. relictum failed to develop in three species of Culicidae, although the primary natural vector for that parasite is a member of the same family. Fallis and Bennett (1966) showed that sylvatic flies were not suitable vectors for L. simondi a parasite of ducks and geese. This indicated that there was a specific relationship between habitat, ornithophilic vectors, hematozoa and their avian hosts perpetuating parasitism.

PAGE 69

60 Sol et al. (2000) illustrated how geographic variation in prevalence of parasitism can be attributed to abundance of vectors using five separate populations of feral pigeons. After investigating vectors of avian hematozoa in Newfoundland, Bennett and Coombs (1975) suggested that the absence of Haemoproteus velans from the Picadae and H. canachites from the Tetraonidae may be a result of low vector density in their study site. They also noted that number of vectors is not likely to be the only factor influencing vector potential. They concluded that low vector density and high vertebrate host density can produce the same prevalence of parasitism as in high vector densities and low vertebrate host densities. The investigation and identification of the insects of Madagascar is in its early stages. Like so many areas of biological investigation there, much remains to be understood. Culicoides spp. are known to be present in Madagascar (Meiswinkel, 1991; Sebastian et al. 2001). Duchemin et al. (2003) discuss the presence of the Anopheles spp., with 26 species on Madagascar. Interestingly, the incredible endemnicity of the island holds true for this genus, with 42% being endemic. Also, 29 species of Aedini are present, ten of these are endemic. The Simuliidae, or black flies, are also present on Madagascar. Species present are known to transmit animal onchocerciasis (Elouard 2003) in Madagascar, but human onchoceriasis in not present. In Madagascar, 22 of the 27 species if Simuliidae are endemic. Elouard comments that at least 13 species remain to be described, and that less than 80% of the Malagasy Simulium spp. have been discovered. So while the same genera that contain ornithophilic vectors are known to be present in Madagascar, it remains unknown which of these may be responsible for transmitting the hematozoa observed in this survey.

PAGE 70

61 Altitude Altitude in an important consideration when examining blood parasitism in birds. With increases in altitude, vegetation and temperatures change affects which vectors will thrive in a particular area. Hawkins et al. (1998) commented that the change in bird density with elevation in Anjanaharibe-Sud somewhat reflected the decrease in canopy height with elevation. In addition, the birds present can also change, as while some species are elevation generalists, others are more restricted within a range of a few hundred meters (Hawkins et al., 1998). Initially, nine altitudinal ranges were compared to determine if any difference in prevalence by elevation was noted. A difference was noted and altitudes were then compared to one another in pairs. Only three groups were significantly different from the others. Two of which, 120 m and 1950 m, represented small sample sizes, and further investigation with larger sample sizes is needed before solid conclusions can be drawn regarding the prevalence of parasitism in these areas. Additionally, at 1200 m, only 6% of the birds were infected. While this was low, it was significantly different only from 400m. After consolidating groups, it became evident only the low group was statistically different from the others. It is not surprising that the lowest elevation had a higher prevalence of parasitism, and this concurs with Raharimanga et al. (2002) who reported the prevalence of Plasmodium or Haemoproteus and microfilariae to be significantly higher below 500 m. Based on the data obtained thus far, it would be difficult to form definite theories as to why prevalence of parasitism is higher at this elevation, as the sites sampled did not include the same ranges, and the habitats in each range are not similar in all cases. This precludes us from drawing solid conclusions, but certainly opens the door for more directed investigations. The prevalence of Plasmodium infections by altitude has been examined in the Hawaiian

PAGE 71

62 Islands (van Riper et al., 1986). In that study, elevation was found to have a significant influence upon both population prevalence and parasitemia levels. In that case it was the mid range elevations that had the highest prevalence of malaria, and it was theorized that the cause for this is the overlap of the vector population and the native (susceptible) avian host. As the vectors for the parasites in our study are identified, and the biology of their life cycles are understood, this information can be compared to the findings reported here to provide a more complete understanding of the factors driving the different prevalences reported here. Thankfully, a relatively large amount has been published in recent years about the avifauna and its altitudinal ranges. With regard to Andohahela, a reserve from which a portion of the birds for this study were sampled, Hawkins and Goodman (1999) discussed species richness with regard to low, medium, and high elevation species of birds and reported species richness slightly greater at mid elevations and lower at high elevations. They reported that the two high elevation sites (1500 and 1875 m) had distinctly fewer species, and at 1875 m there were fewer forest species than at the other altitudes (more aerial species were noted here). When graphed, species richness for Andohahela peaked at 1200 m and began to decline as altitude increased. In Anjanaharibe-Sud, species richness was clearly lower at the highest range sampled, 1950 m, but the results were unclear with regard to comparisons between the other three altitudes. The biology of the avian host is an important factor affecting parasitism, and understanding this allows identification of trends in the future. For instance, van Riper et al. (1986) suggested that avian malaria may be restricting native Hawaiian populations to higher elevations and drier areas, in addition to modifying behavior patterns to minimize contact with vectors

PAGE 72

63 of the parasite. Monitoring the prevalence of parasitism through the ranges in elevation discussed here, combined with data on the avian populations will continue to be an important tool for conservationists and biologists. Reserves Prevalence was compared between the four regions from which birds were sampled. The four regions sampled were reserves, Ambohitantely, Anjanaharibe-Sud, Andohahela and Montagne dAmbre. Ambohitantely is in the central highlands, and is an area of forested fragments. Anjanaharibe-Sud lies along a chain of mountains covered with humid forest in northeastern Madagascar; the climate is a humid and tropical (Goodman, 1998; Goodman and Lewis, 1998). Relative humidity reaches 97% and annual precipitation is reported to be slightly more than 2 m, two thirds of which falls in the rainy season. The rainy season here begins in December, and the samples taken prior to the start. On average it rains 271 days per year (Goodman and Lewis, 1998). Andohahela is at the southeastern tip of Madagascar. All of this reserve falls below the Tropic of Capricorn and is one of the southernmost tropical forests in the Old World (Goodman, 1999). A variety of habitats were found in this reserve, including humid forest, spiny forest, gallery forest, savanna, and riverine habitats. Additionally, the elevation ranges from low (440 m) to high (1875 m). Because of these factors there is a rich variety of avian species represented (Hawkins and Goodman, 1999). Andohahela is considered comparable to Anjanaharibe-Sud. Finally, there is Montagne dAmbre, an isolated area of montane forest at the far north end of the island. This is mainly a mid to high elevation site, as forest starts at 1000 m and reaches to 1300 m. A statistical difference in overall prevalence by reserve was observed (P = 0.023). Reserves were then compared in pairs. Andohahela was clearly different from

PAGE 73

64 Ambohitantely. Compared with Montagne dAmbre, the difference was not significant, but the power of the test was low. Because Montagne dAmbre had the same prevalence as both Ambohitantely and Anjanaharibe-Sud, and the power of the test was low (probably as a result of the smaller sample size in this reserve), I interpreted the results to indicate a difference does exist. When compared to Anjanaharibe-Sud, the p-value was 0.069, only slightly higher than the generally accepted cut-off of 0.05. Because it is so close to this mark as well as the pattern of parasitism for this region, I feel that there is a difference and a larger sample size would clarify this. It is not entirely unexpected that these two groups are more similar, as biologists working in the area have described the sites as comparable. Andohahela had the highest prevalence of each genus of hematozoa compared to the other three regions. The only exception was in relation to the trypanosomes, where there was one infected bird in Andohahela and one in Ambohitantely. Four birds sampled in Andohahela were infected with Plasmodium the two birds that were infected with the unnamed species of Babesia were also from this reserve. Andohahela had the same amount of Leucocytozoon as the other three groups combined, and more Haemoproteus and microfilariae infections than the other three groups combined. As described earlier, this particular reserve has a variety of habitats and elevations resulting in a high density of avian species. In addition to the variety of potential host species, and microhabitats to suit difference types of vectors, there is also known to be several types of aquatic habitats within this reserve. There is marshland, slow moving rivers, and the reserve borders on the tributary of one of the larger rivers in the area. Over 50% of the known Malagasy species of Simulium occur in the

PAGE 74

65 southeastern region, specifically Andohahela, of Madagascar (Pilaka and Elouard, 1999). No statistical significance was observed between the other groups. Habitat Birds were classified into three groups: FED, FDE, and FDW. There appeared to be no difference in prevalence observed in between groups, but all Chi-squares were performed with a low power, indicating that not enough data were included. The majority of the species sampled were either forest dweller or forest dependant species, 365 of 378 birds total. Only 13 birds considered to be forest edge species were sampled. Data on this factor should continue to be collected, so that habitat can be compared using a larger and more balanced sample. Effort should be made to investigate a proportionate amount of forest edge dwelling species for this classification to be compared to the other two. Gender Gender was established for 187 of the birds sampled, and prevalence was 22% for each cohort. Seventeen of 76 females and 24 of 111 males were infected with at least one species of hematozoa. No statistical difference was observed between groups. Raharimanga et al. (2002) also reported no difference was found in the prevalence in males compared to females. Sol et al. (2000) found no difference in prevalence of parasitism by gender, pointing out that there is probably equal exposure to vectors. Weatherhead and Bennett (1991) reported a higher prevalence of parasitism in male Red-winged Blackbirds than in females, and as already mentioned, the males were more likely to remain in the same area while females were less likely to remain or return to the area. Based on this, it is possible that the higher prevalence seen in males in that study is simply a result of greater exposure to vectors. Perhaps the other sites where these birds move to are less ideal for the vector, reducing the rate of transmission. Although

PAGE 75

66 Deviche et al. (2001) found a difference in intensity by gender, prevalence was not different between genders in Dark-eyed Juncos. Additionally, they postulated that the difference in intensity of Leucocytozoon infection observed was probably a result of a difference in behaviors exhibited by the two genders. Age Sixty-five of 284 adults and 2 of 26 juveniles were found to be infected with at least one genus of hematozoa. No significant difference in prevalence was observed between adults and juveniles. There is some concern when sampling young birds about if enough time has passed since hatching for the birds to become infected, and the infection to become patent. Since all birds were mist-netted for this survey, the juveniles were old enough to have fledged and therefore old enough for infection to become patent assuming they were fed on early after hatching. The lack of difference between the two groups indicates there is at least equal exposure to the vectors for both age groups. Sol et al. (2000) found that intensity of parasitism by age class was similar. Weatherhead and Bennett (1990) reported increasing prevalence of parasitism with age in male Red-winged Blackbirds. Peirce (1984b) reported an only slightly higher prevalence of parasitism in juveniles at one site in Zambia. In Hawaii, van Riper et al. (1986) reported no significant difference in prevalence between juveniles and adults for either native or introduced species. In Madagascar, no difference in prevalence by genus was found between adults and juveniles by Raharimanga et al. (2002). Breeding Condition Reproductive status here was solely based on the presence or absence of a brood patch, indicating that the bird was actively nesting or sitting on eggs. No statistical difference was observed in parasitism between groups. Because of the nature of the

PAGE 76

67 study design, there was no other way to determine reproductive status other than this morphological cue. Results for this can be confounded, for instance if species were sitting on nests that that time but perhaps do not develop brood patches. If they were parasitized, but not showing a brood patch, they would have been inappropriately grouped with those birds not laying at that time. Because of the variation in parental effort and brood patch development in birds, it is perhaps best to consider this factor only when examining the same species or family of bird.

PAGE 77

CHAPTER 6 CONCLUSIONS In this study, the presence of seven new species of hematozoa was reported, possibly endemic to the island. This increases the current knowledge of the flora and fauna of the island, and indicates that the parasites of these birds may also have a large degree of endemism like their hosts. Additional research is needed before solid conclusions can be reached. The prevalence of parasitism in one of the sites sampled, Andohahela, was shown to be significantly higher than the other three sites. This was also the area with a high degree of species richness and varied habitats within the sampling area to support a variety of ornithophilic flies. Prevalence also varied by avian family, which could be a result of one or a combination of factors such as avian host behavior, vector abundance, or vector feeding preference. Finally, prevalence of parasitism was significantly higher at low altitudes. No difference in parasitism was observed for habitat, age, gender or breeding condition. The prevalence of parasitism, both overall and by parasite genera, was found to be similar to that reported from other countries. As entomological and ornithological research continues in Madagascar, a more complete understanding of the parasitism can be developed. Many facets of the relationship between the parasite, arthropod vector and avian host remain to be investigated. For instance, it remains to be seen what impact parasitism by hematozoa has on the avian populations, and if endemic birds are suffering more as a result of infection than birds representing more broadly distributed species. Also, there are 68

PAGE 78

69 isolated and distinct populations of birds within a species that do not mix. It is important to compare these groups to determine if they are parasitized by the same species of hematozoa, or if one population is more commonly parasitized or less so. Only limited conclusions can be drawn considering the amount of data yet to be collected. The results of this study provide a baseline for other islands that have not been examined. A great deal remains to be learned about the flora and fauna of Madagascar. This study and those done beforehand have established a baseline of knowledge upon which future investigations can be developed.

PAGE 79

APPENDIX CHICKEN HEMATOZOA Chickens and other poultry are an important commodity throughout Africa, but are most important in small rural or impoverished communities. Chickens are of important nutritional value, both as meat animals, but also for eggs. Chickens also may be used as gifts or in celebrations. Conservationists are also interested in knowing if the importation of birds may be bringing potential pathogens into a nave population. So, for several reasons the health of chickens and other domestic fowl is important. In rural communities, chicken are usually allowed to forage freely, and are not kept caged or penned as commonly seen in larger production oriented farms. Foraging may increase the exposure to vector species of ornithophilic flies. This can be a result of birds spending more time in the range occupied by vectors, or also by being present in those environments at the preferred feeding times by those vectors. Haemoparasites of poultry have been investigated, probably largely due to the detrimental effects of Leucocytozoon caulleryi in Eastern Asia. This particular parasite is the causative agent of chicken leucocytozoonosis, which is known to cause weight loss and decreased egg production, and death (Isobe and Akiba, 1986). Several studies have been published summarizing the known hosts and parasites, and species of Leucocytozoon Plasmodium Haemoproteus Trypanosoma and Aegyptianella have all been reported (Fallis et al., 1973; Bennett et all, 1991; Earl et al., 1991; Huchzermeyer and van der Vyver, F.H., 1991; Huchzermeyer, 1993). 70

PAGE 80

71 Blood samples were taken from 15 domestic chickens, and examined for the presence of hematozoa. The majority of the chickens were obtained from a village near the limit of the Anjanaharibe-Sud forest. Two more were from Tolagnaro, while the origins of the remaining three were unknown. A single macrogametocyte of Leucocytozoon schoutedeni was observed in one blood smear from these birds. No other species of hematozoa were seen to be present in these birds. Further investigation would be helpful in understanding the prevalence of blood parasitism in domestic poultry in Madagascar, although it appears that prevalence is low. It is possible that chickens infected with Leucocytozoon schoutedeni were imported, bringing the parasite to the island. Low prevalence may indicate that the insect that is vectoring this parasite is not the optimal vector, therefore, development of the parasite in the arthropod host is somehow partially impeded. Another possible cause of low parasite prevalence is that the parasite is highly pathogenic and kills infected birds rapidly. There are no reports regarding high chicken mortality in Madagascar, and although there are no published data to this point, it does not seem likely that the low prevalence in this study was due to pathogenicity. Because these chickens were all from remote areas, it is probable that that they were raised in that area from stock that the family or village had maintained. Also, due to the low economic status of the population, it is unlikely that most individuals would be able to import birds themselves. Therefore, the suggestion that the parasite is being maintained at low levels in the country is more likely a result of inadequate capability of flies to transmit this parasite. Finally, while this parasite matches the description of L. schoutedeni it is important to examine a larger number of

PAGE 81

72 birds to determine the true parameters and morphology of Leucocytozoon species present, to conclusively determine if it is a known parasite from Africa or a species endemic to Madagascar. In addition, it will be necessary to determine import trends with regard to poultry. If eggs are imported to raise replacement stock, it is unlikely that the parasite observed here is of African origin. If young birds are more commonly imported it is possible that they could become infected prior to shipping, and develop a parasitemia and be responsible for the arrival of this parasite to Madagascar. Although the margins have been set throughout this document correctly, please pay close attention to the possibility of picture frames overlapping the margin. The base style to use is Normal

PAGE 82

LIST OF REFERENCES Apanius, V., and C.E. Kirkpatrick. 1988. Preliminary report of Haemoproteus tinnunculi infections in a breeding population of American Kestrels (Falco sparverius). Journal of Wildlife Diseases 24: 150-153. Atkinson, C.T. 1986. Host specificity and morphometric variation of Haemoproteus melegridis Levine, 1961 (Protozoa: Haemosporina) in gallinaceous birds. Canadian Journal of Zoology 64: 2634-8. Atkinson, C.T., Forrester, D.J., and E.C. Greiner. 1988a. Pathogenicity of Haemoproteus melegridis (Haemosporina: Haemoproteidae) in experimentally infected turkeys. Journal of Parasitology 74: 228-239. Atkinson, C.T., Forrester, D.J. and E.C. Greiner. 1988b. Epizootiology of Haemoproteus meleagridis (Protozoa: Haemosporina) in Florida: Seasonal transmission and vector abundance. Journal of Medical Entomology 25: 45-51. Baker, J.R. 1968. The host restriction of Haemoproteus columbae Journal of Protozoology 15: 334-335. Beier, J.C., and M.K. Stoskopf. 1980. The epidemiology of avian malaria in Black-Footed penguins ( Spheniscus demersus ). Journal of Zoo Animal Medicine 11: 99-105. Bennett, G.F. 1961. On the specificity and transmission of some avian trypanosomes. Canadian Journal of Zoology 39: 17-33. Bennett, G.F. 1970. Development of trypanosomes of the T. avium complex in the invertebrate host. Canadian Journal of Zoology 48: 945-957. Bennett, G.F. 1990. Avian Haemoprotozoa of the Indian subcontinentthe species and vectors. Proceedings of the Zoological Society, Calcutta 43: 49-58. Bennett, G.F., Bishop, M.A., and M.A Peirce. 1993. Checklist of the avian species of Plasmodium Marchiafava & Celli, 1885 (Apicomplexa) and their distribution by avian family and Wallacean life zones. Systematic Parasitology 26: 171-179. Bennett, G.F. and J. Blancou. 1974. A note on the blood parasites of some birds from the Republic of Madagascar. Journal of Wildlife Diseases 10: 239-240. 73

PAGE 83

74 Bennett, G.F., Blancou, J., White, E.M., and N.A. Williams. 1978. Blood parasites of some birds from Senegal. Journal of Wildlife Diseases 14: 67-73. Bennett, G.F., and A.G. Campbell. 1972. Avian Haemoproteidae. I. Description of n. sp. and a review of the haemoproteids of the family Turdidae. Canadian Journal of Zoology 50: 1269-1275. Haemoproteus fallisi Bennett, G.F. and R.F. Coombs. 1975. Ornithophilic vectors of avian hematozoa in insular Newfoundland. Canadian Journal Zoology 53: 1241-1246. Bennett, G.F., Earl, R.A., Peirce, M.A., Huchzermeyer, F.W. and D. Squires-Parsons. 1991. Avian Leucocytozoidae: the leucocytozoids of the Phansianidae sensu lato. Journal of Natural History 25: 1407-1428. Bennett, G.F., Earl, R.A., and D. Squires-Parsons. 1994. Trypanosomes of some sub-Saharan birds. Onderstepoort Journal of Veterinary Research 61: 263-271. Bennett, G.F. and A.M. Fallis. 1960. Blood parasites of birds in Algonquin Park, Canada, and a discussion of their transmission. Canadian Journal of Zoology 38: 261-273. Bennett, G.F., Fallis A.M., and A.G. Campbell. 1972. The response of Similium ( Eusimulium ) euryadminiculum Davies (Diptera: Simuliidae) to some olfactory and visual stimuli. Canadian Journal of Zoology 50: 793-800. Bennett, G.F., and C.M. Herman. 1976. Blood parasites of some birds from Kenya, Tanzania and Zaire. Journal of Wildlife Diseases 12: 59-65. Bennett, G.F., Okia, N.O., and M.F. Cameron. 1974. Avian hematozoa of some Ugandan birds. Journal of Wildlife Diseases 10: 458-465. Bennett, G.F., and M.A. Peirce. 1988. Morphological form in the avian Haemoproteidae and an annotated checklist of the genus Haemoproteus Kruse, 1890. Journal of Natural History 22: 1683-1696. Bennett, G.F. and M.A. Peirce. 1990. The haemoproteid parasites of the pigeons and doves (family Columbidae). Journal of Natural History 24: 311-325. Bennett, G.F., Peirce, M.A., and R.A. Earle. 1994. An annotated checklist of the valid avian species of Haemoproteus Leucocytozoon (Apicomplexa: Haemosporida) and Hepatazoon (Apicomplexa: Haemogregarinidae). Systemic Parasitology 29: 61-73. Clements, J.F. 2000. Birds of the World: A Checklist, 5th Ed, Ibis Publishing Company, Vista, California. De Mello, I.F. 1935a. A contribution to the study of the blood parasites of some Indian birds. Proceedings of the Indian Academy of Science I: 349-358.

PAGE 84

75 De Mello, I.F. 1935b. New haemoproteids of some Indian birds. Proceedings of the Indian Academy of Science (B) 2: 469-475. De Mello, I.F. and L. da Fonseca.1937. Further notes on the haemoparasitology of the Indian birds. Proceedings of the Indian Academy of Science. (B) 6: 213-219. Deviche, P., Greiner, E.C., and X. Manteca. 2001. Seasonal and age-related changes in blood parasite prevalence in Dark-eyed Juncos ( Junco hyemalis Aves, Passeriformes). Journal of Experimental Zoology 289: 456-466. Duchemin, J.B., Ravoahangimalala, O.R., and G. Le Goff. 2003. Culicidae, Mosquitoes. In The Natural History of Madagascar, S. M. Goodman and J. P. Benstead (eds.), The University of Chicago Press, Chicago. Earl, R.A., Horak, I.G., Huchzermeyer, F.W., Bennett, G.F., Braack, L.E.O. and B.L. Penzhorn, 1991. The prevalence of blood parasites in helmeted guinea fowls, Numida melegris in the Kruger National Park. Onderstepoort Journal of Veterinary Research. 58: 145-147. Elouard, J.M. 2003. Simuliidae, Black Flies. In The Natural History of Madagascar, S. M. Goodman and J. P. Benstead (eds.), The University of Chicago Press, Chicago. Fallis, A.M. and G.F. Bennett. 1966. On the epizootiology of infections caused by Leucocytozoon simondi in Algonquin Park, Canada. Canadian Journal of Zoology. 44: 101-112. Fallis, A.M., Desser, S.H., and R.A. Khan. 1974. On species of Leucocytozoon Advances in Parasitology 12: 1-67. Fallis, A.M., Jacobson, R.L., and J.N. Raybould. 1973a. Experimental transmission of Trypanosoma numidae Wenyon to guinea fowl and chickens in Tanzania. Journal of Protozoology 20: 436-437. Fallis, A.M. Jacobson, R.L. and J.N. Raybould. 1973b. Haematozoa in domestic chickens and guinea fowl in Tanzania and transmission of Leucocytozoon neavi and Leucocytozoon schoutedni Journal of Protozoology 20: 438-442. Fallis, A.M., Pearson, J.C., and G.F. Bennett. 1954. On the specificity of Leucocytozoon Canadian Journal of Zoology 32: 120-124. Fallis, A.M. and S.M. Smith. 1965. Attractions of some simuliids to ether extracts from birds and to carbon dioxide. Proceedings of the XII International Congress of Entomology. London, 1964. Fix, A.S., Waterhouse, C., Greiner, E.C., and M.K. Stoskopt. 1988. Plasmodium relictum as a cause of avian malaria in wild-caught Magellanic penguins. Journal of Wildlife Diseases 24: 610-619.

PAGE 85

76 Forrester, D.J., Greiner, E.C., and G.F. Bennett. 1977. Avian Haemoproteidae. 7. A review of the haemoproteids of the family Ciconiidae (storks) and descriptions of Haemoproteus brobkorbi sp.nov. and H. peirci sp.nov. Canadian Journal of Zoology. 55: 1268-1274. Goodman, S.M. 1998. Description of the 1994 Biological Inventory of the Rserve Spciale dAnjanaharibe-Sud, Madagascar. Fieldiana: Zoology 90: 1-7. Goodman, S.M. 1999. Description of the Rserve Naturelle Intgrale dAndohahela, Madagascar, and the 1995 Biological Inventory of the Reserve. Fieldiana: Zoology 94: 1-8. Goodman, S.M. and B.A. Lewis. 1998. Description of the Rserve Spciale dAnjanaharibe-Sud, Madagascar. Fieldiana: Zoology 90: 9-16. Greiner, E.C., Bennett, G.F. White, E.M, and R.F. Coombs. 1975. Distribution of the avian hematozoa of North America. Canadian Journal of Zoology 53: 1762-1787. Greiner, E.C., Eveleigh, E.S., and W. M. Boone. 1978. Ornithophilic Culicoides spp. (Diptera: Ceratopogonidae) from New Brunswick, Canada, and implications of their involvement in haemoproteid transmission. Journal of Medical Entomology 14: 701-704. Greiner, E.C., Putnam, M.S., and S.M. Goodman. 1996. Blood parasites from birds in the Rserve Naturelle Intgrale d Andringitra, Madagascar. Fieldiana: Zoology 85: 142-143. Hawkins, A.F.A., and S.M. Goodman. 1999. Bird community variation with elevation and habitat in parcels 1 and 2 of the Rserve Naturelle Intgrale dAndohahela, Madagascar. Fieldiana: Zoology 94: 175-186. Hawkins, A.F.A., Thoillay, J-M, and S.M. Goodman. 1998. The birds of the Rserve Spciale dAnjanaharibe-Sud, Madagascar. Fieldiana: Zoology 90: 93-127 Hayworth, A.M., van Riper, C., and W.W. Weathers. 1987. Effects of Plasmodium relictum on the metabolic rate and body temperatures in canaries ( Serinus canarius ) Journal of Parasitology 73: 850-853. Herman, C.M. 1975. Plasmodium circumflexum : Age as a factor in isodiagnosis in Anatidae. Experimental Parasitology 38: 83-86. Huchzermeyer, F.W. and van der Vyver, F.H. 1991. Isolation of Plasmodium circumflexum from wild guineafowl ( Numida melegris ) and the experimental infection in domestic poultry. Avian Pathology 20: 213-223.

PAGE 86

77 Huchzermeyer, F.W. 1993. A host-parasite list of the haematozoa of domestic poultry in sub-Saharan Africa and the isolation of Plasmodium durae Herman from turkeys and francolins in South Africa. Onderstepoort Journal of Veterinary Research 60: 15-21. Isobe, T. and K. Akiba. 1986. Development of erythrocytic merozoites to gametocytes in chickens recovered from sporozoite infection with Leucocytozoon caulleryi Journal of Parasitology 72: 190-192. Jarvi, S.I., Scultz, J.J., and C.T. Atkinson. 2002. PCR Diagnostics underestimate the prevalence of avian malaria ( Plasmodium relictum ) in experimentally infected passerines. Journal of Parasitology 88: 153-158. Kocan, R.M., and D.T. Clark. 1966. Anemia in ducks infected with Leucocytozoon simondi Journal of Protozoology 13: 465-468. Kottak, C.P. 1980. The past in the present: history, ecology, and cultural variation in highland Madagascar. University of Michigan Press, Ann Arbor, Michigan. Kucera, J. 1981. Blood parasites of birds in central Europe. 1. Survey of literature. The incidence in domestic birds and general remarks to the incidence in wild birds. Folia Parasitologica (PRAHA) 28: 13-22. Langrand, O. 1990. Guide to the birds of Madagascar. Yale University Press, New Haven, Connecticut Langrand, O. 2001. Family Brachypteraciidae (Ground-rollers). In Handbook of the Birds of the World Vol. 6 Mousebirds to Hornbills, del Hoyo, J., Elliot, A., and J. Sargatal (eds.) Lynx Edicions, Barcelona, Spain, p. 378-388. Lowery, R.S. 1971. Blood parasites of vertebrates on Aldabra. Philosophical Transactions of the Royal Society, -B 260: 577-580. McClure, H.E., Poonswad, P., Greiner, E.C., M. Laird. 1978. Haematozoa in the birds of Eastern and Southern Asia. Memorial University of Newfoundland; St. Johns, Newfoundland. Meiswinkel, R. 1991. Afrotropical Culicoides : C ( avaritia ) miombo sp. nov., a widespread species closely allied to C ( A .) imicola Kieffer, 1913 (Diptera: Ceratopogoniidae). Onderstepoort Journal of Veterinary Research 58: 155-170. Morris, P. and F. Hawkins. 1998. Birds of Madagascar A Photographic Guide. Yale University Press, New Haven, Connecticut, p. 298. Nakamura, K., Ogiso M., Shibahara T., Kasuga H.,and T. Isobe. 2001. Pathogenicity of Leucocytozoon caulleryi for specific pathogen-free laying hens. Journal of Parasitology 87: 1202-1204.

PAGE 87

78 Nayar, J.K., Knight, J.W., and S.R. Telford. 1998. Vector ability of mosquitoes for isolates of Plasmodium elongatum from raptors in Florida. Journal of Parasitology. 84: 542-546. Peirce, M.A. 1979 Some additional observations on haematozoa of birds in the Mascarene Islands. Bulletin of the British Ornithologists Club 99: 68-71. Peirce, M.A. 1984a. Haematozoa of African birds: some miscellaneous findings. African Journal of Ecology 22: 149-152. Peirce, M.A. 1984b. Haematozoa of Zambian birds. I. General survey. Journal of Natural History 18: 105-122. Peirce, M.A. 1984c. Haematozoa of Zambian birds. VIII. Redescription of Haemoproteus dicruri from Dicrurus adsimilis (Dicruridae). Journal of Natural History 18: 789-791. Peirce, M.A., Cheke, A.S., and R.A. Cheke. 1977. A survey of blood parasites of birds in the Mascarene Islands, Indian Ocean, with descriptions of two new species and taxonomic discussion. Ibis 119: 451-461. Peirce, M.A. and C.J. Feare. 1978. Piroplasmosis in the masked booby Sula dactylatra melanops in the Amirantes, Indian Ocean. Bulletin of the British Ornithologists Club 98: 38-39. Pilaka, T. and J.M. Elouard. 1999. Aquatic biodiversity of Madagascar: Simulium (Diptera: Simuliidae) from the Rserve Naturelle Intgrale dAndohahela and surrounding areas. Fieldiana Zoology 94: 125-128. Raharimanga, V., Soula, F., Raherilalao, M.J., Goodman, S.M., Sadons, H., Tall, A., Randrianarivelojosia, M., Raharimalala, L., Duchemin, J.B., Ariey, F., and V. Robert. 2002. Hmoparasites des oiseaux sauvages Madagascar. Archives de l'Institut Pasteur de Madagascar 68: 90-99. Raidal, S.R., and S.M. Jaensch. 2000. Central nervous disease and blindness in Nankeen kestrels ( Falco cenchroieds ) due to a novel Leucocytozoon -like infection. Avian Pathology 29: 51-56. Sebastiani, F., Meiswinkel, R., Gomulski, L.M., Guglielmino, C.R., Mellors, P.S., Malacrida, A.R. and G. Sasperi. 2001. Molecular differentiation of the Old World Culicoides imicola species complex (Diptera, Ceratopogonidae), inferred using random amplified polymorphic DNA markers. Molecular Ecology. 10: 1773-1786. Sol, D., Jovani, R. and J. Torres. 2000. Geographical variation in blood parasites in feral pigeons: the role of vectors. Ecography 23: 307-314. Stoskopf, M.K., and J. Beier. 1979. Avian malaria in African black-footed penguins. Journal of the American Veterinary Medical Association 175: 944-947.

PAGE 88

79 Van Riper, C., van Riper, S.G., Goff, M.L., and M. Laird. 1986. The epizootiology and ecological significance of malaria in Hawaiian land birds. Ecological Monographs, 56: 327-344. Warner, R.E. 1968. The role of introduced disease in the extinction of the endemic Hawaiian avifauna. The Condor 70: 101-120. Weatherhead, P. J. and G.F. Bennett. 1991. Ecology of Red-winged Blackbird parasitism by haematozoa. Canadian Journal of Zoology 69: 2352-2359. White, E.M., Greiner, E.C., Bennett, G.F., and C.M. Herman. 1978. Distribution of the hematozoa of Neotropical birds. Revista de Biologia Tropical 26: 43-102. Work, T.M., Washino, R.K. and C. van Riper. 1990. Comparative susceptibility of Culex tarsalis Anopheles franciscanus and Culiseta inornata (Diptera: Culicidae) to Plasmodium relictum (Haemosporina: Plasmodiiae). Journal of Medical Entomology 27: 68-71. Yorinks, N., and C.T. Atkinson. 2000. Effects of malaria on activity budgets of experimentally infected juvenile Apapane ( Himatione sanguinea ). The Auk 117: 731-738.

PAGE 89

BIOGRAPHICAL SKETCH I was born in upstate New York, the youngest of four children. We moved to Cheshire, Connecticut when I was young, and that is where I grew up. Before attending the University of Florida, I attended the University of Connecticut and graduated with a double major in pathobiology and animal science. I enjoy international travel, particularly to the Andean nations. When I have the opportunity, I also enjoy reading and studying European history. 80


Permanent Link: http://ufdc.ufl.edu/UFE0001121/00001

Material Information

Title: Identity and prevalence of blood parasites in wild-caught birds from Madagascar
Physical Description: Mixed Material
Creator: Savage, Amy Frances ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001121:00001

Permanent Link: http://ufdc.ufl.edu/UFE0001121/00001

Material Information

Title: Identity and prevalence of blood parasites in wild-caught birds from Madagascar
Physical Description: Mixed Material
Creator: Savage, Amy Frances ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001121:00001


This item has the following downloads:


Full Text












IDENTITY AND PREVALENCE OF BLOOD PARASITES IN WILD-CAUGHT
BIRDS FROM MADAGASCAR















By

AMY FRANCES SAVAGE


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Amy Frances Savage















ACKNOWLEDGMENTS

I thank my friends and family for their support and lessons. Also, I thank the

professors, research scientists, and other students with whom I have worked for being a

source of inspiration and encouragement. Finally, I would like to thank the members of

my committee. Drs. Donald J. Forrester and David W. Steadman were enthusiastic,

energetic and creative. I am grateful for the time they took to teach and motivate me. I

sincerely thank my committee chair, Dr. Ellis Greiner, for his kindness, patience and

understanding.
















TABLE OF CONTENTS
page

A C K N O W L E D G M E N T S ......... ...................................................................................... iii

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

LIST OF FIGURES ......... ....................... ..... ........ ........... vii

A B STR A C T ... .................... ............................................ ... ....... ....... viii

CHAPTER

1 INTRODUCTION ................................................. .......................

2 MATERIALS AND METHODS ........................................................... ......... 11

3 NEW SPECIES DESCRIPTIONS ........................................ ........................ 13

Haemoproteus goodmani n. sp. ............................................................................14
Taxonom ic Sum m ary ................................................. ............................. 15
R e m a rk s ......................................................................................................... 1 6
E ty m o lo g y ..................................................................................................... 1 6
H aem oproteus forresteri n. sp ........................................................... ............... 16
Taxonom ic Sum m ary ................................................. ............................. 17
R e m a rk s ......................................................................................................... 1 8
E ty m o lo g y ..................................................................................................... 1 8
H aem oproteus vangii n. sp. ............................................... ............................. 19
Taxonom ic Sum m ary ................................................. ............................. 20
R e m a rk s ......................................................................................................... 2 0
E ty m o lo g y ..................................................................................................... 2 0
H aem oproteus khani n. sp................................................. .............................. 21
Taxonom ic Sum m ary ................................................. ............................. 21
R e m a rk s ......................................................................................................... 2 2
E ty m o lo g y ..................................................................................................... 2 2
H aem op roteu s dicru ri ...................................................................... ....................2 2
Taxonom ic Sum m ary ................................................. ............................. 23
R e m a rk s ......................................................................................................... 2 3
L eucocytozoon frasci n. sp. ............................................................. .....................25
Taxonom ic Sum m ary ................................................. ............................. 26
R em ark s ...................... ................ .....................................................2 6
E ty m o lo g y ..................................................................................................... 2 7










Leucocytozoon lairdi n. sp........ ............................................................................. 27
Taxonom ic Sum m ary ........................................ ...........................................27
R em arks .......................................................................... 28
E ty m o lo g y ...................................................................................................... 2 8
Leucocytozoon greineri n. sp..................................................................................29
Taxonom ic Sum m ary ........................................ ...........................................29
R em arks .......................................................................... 30
E ty m o lo g y ...................................................................................................... 3 0

4 SU RVEY RESU LTS .......................................................................... ...................36

Prevalence by host fam ily........................................... ...........................................36
A ltitu d e ..............................................................................3 9
R e se rv e s ............. ............................ .. ........................................................3 9
H habitat .......... ........................................................................................40
G e n d e r .................................................................................................................... 4 0
A ge ...................... ................................................................................................ 41
Breeding Condition................................................................................................ 41

5 D ISCU SSION ................................................................................................ 45

F a m ilie s .................................................................................................................. 4 9
V e c to r s ............. ............................ .. .........................................................5 8
A ltitu d e ..............................................................................6 1
R e se rv e s ............. ............................ .. ........................................................6 3
H habitat .......... ........................................................................................65
G e n d e r .................................................................................................................... 6 5
A ge ...................... ................................................................................................ 66
Breeding Condition................................................................................................ 66

6 CON CLU SION S ................................................................68

APPENDIX CHICKEN HEMATOZOA............................... .... ......... 70

LIST OF REFEREN CE S ......................................... ........... ..............................73

BIO GR A PH ICA L SK ETCH ........................................................................................ 80














v
















LIST OF TABLES


Table page

3-1 Morphometric variation in the haemoproteids from the Brachypteraciidae,
V angidae, and D icruridae .......................................................... ............... 31

3-2 Morphometric variations of the Leucocytozoon spp. of the Brachypteraciidae,
Vangidae, and Philepittidae...................... ....... ............................. 33

4-1 Prevalence of hematozoa in the avifauna of Madagascar, by avian family. ............42
















LIST OF FIGURES


Figure pge

3-1 Haemoproteus spp. of the Brachypteraciidae, Vangidae, and Dicruridae ...............34

3-2 Leucocytozoon spp. of the Brachypteraciidae, Vangidae, and Philepittidae...........35











Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

THE IDENTITY AND PREVALENCE OF BLOOD PARASITES IN WILD-CAUGHT
BIRDS FROM MADAGASCAR

By

Amy Frances Savage

August 2003

Chair: Ellis C. Greiner
Major Department: Pathobiology

The Republic of Madagascar is an area of considerable biological interest because

of the high degree of endemnism of the flora and fauna. Limited research has been

performed investigating the hematozoa of the avifauna on the island; and little is known

regarding the prevalence of these parasites or their effects. I examined 378 wild-caught

birds from 21 families, and 15 domestic fowl, for the presence of hematozoa to determine

the hematozoon fauna and prevalence of parasitism in the area. Birds were captured in

mist nets; and blood smears were made in the field at four different reserves in

Madagascar. Slides were stained with Giemsa and examined for the presence of

hematozoa on a light microscope. Prevalence by genus of parasite was Haemoproteus

spp. 13.8%, Leucocytozoon spp. 11%, microfilariae 6.1%, Plasmodium spp. 1.6%,

Trypanosoma spp. 1.1%, and Babesia sp. 0.5%. Seven new species of hematozoa were

recognized from the Brachypteraciidae (Ground-rollers), Vangidae (Vangas),

Philepittidae (Asities), and Dicruridae (Drongos). Haemoproteus goodmani, H.

forresteri, H. vangii, H. khani, Leucocytozoon frasci, L. lairdi, and L. greineri were

described. The overall prevalence of infection observed was 24.3% (92 of 378).









Zosteropidae and Ploceidae were the two most parasitized families. Prevalence also

varied by altitude and sampling site. Birds from the lowest altitudes had the highest

prevalence of parasitism. Birds from one reserve had a higher prevalence than the birds

from other areas. This reserve had a high avian species density and a variety of habitats.

No differences in prevalence were observed by habitat, gender, age, or breeding

condition.














CHAPTER 1
INTRODUCTION

The Republic of Madagascar is a continental island located approximately 400

kilometers (km) off the southeastern coast of Africa. It is approximately 1600 km long,

and 560 km across at the widest point, making it the fourth largest island in the world,

after Greenland, New Guinea, and Borneo. At 590,000 km2 (Kottack, 1980), it is roughly

2.5 times the size of Great Britain.

Because of Madagascar's proximity to Africa, many people assume that it is most

closely associated with Africa. In actuality this is not the case. Originally, Africa,

Madagascar, South America, Australia, India and Antarctica were all part of one

supercontinent. During the Paleozoic era 200 million years ago, Africa is presumed to

have broken off of Madagascar. Afterwards, Australia separated, and finally India.

Therefore Madagascar has more recent geological ties to India than Africa. Madagascar

is very similar to Africa in structure and climatic zones, perhaps because it shares the

same ocean and air currents. It is an island with a range of habitats including coastal

plains, tropical forest, and a semi desert. The highest peak is 2,876 meters. Humans

were absent until approximately 2,000 years ago, when the earliest settlers colonized the

island. The descendants of these Indonesian and African immigrants have now evolved

into more than 20 different ethnic groups (Kottak, 1980).

One of the important facts in ecological history about Madagascar is that for 40

million years it remained isolated, allowing the fauna and flora to evolve with little

continental influence. There were no large mammals like those on Africa (Kottack,









1980). Today, there are more varieties of orchids on Madagascar than anywhere else in

the world. Most of the approximately 6,000 species of plants there are found only on this

island. Ninety-eight percent of the nearly 500 reptile and amphibian species are endemic.

The avifauna on the island is also distinct. There are 282 species of birds found on

or near Madagascar and 110 of those are endemic. Five are known or thought to be

extinct. Most of the endemic species (80 of 110) are forest dwelling. This represents 30

of 37 genera.

Blood parasites were first observed in the 1880s. Since then, researchers have been

trying to understand the intricacies of their life cycles and the role they play in human and

animal disease. While a great deal has been learned since Daniewlsky's discovery, there

are still many unknowns facing researchers. While many projects have been designed to

identify the blood parasites in birds, not every avian family has been investigated equally;

and there are areas of the world where little, if any investigation has been done. One of

the least investigated areas is Madagascar.

There have been three studies of avian blood parasites in Madagascar. Bennett and

Blancou (1974) examined 64 birds representing 32 species and found 14 birds

(representing 8 species) to be infected. They concluded that the prevalence of hematozoa

was low; and additionally that there were no unique species. The second study examined

smears from 10 birds (Greiner et al., 1996). Blood parasites were found in 6 of the 10

birds; and mixed infections were observed for the first time. Most recently, Raharimanga

et al. (2002) published a study of hematozoa of Malagasy birds from a variety of

locations on the island. Unfortunately, parasites were not identified to the species level.

No conclusions were drawn about the pathogenicity of parasites in these hosts. With









such limited investigation, and in consideration of the wide variety of avifauna on the

island, we have opted to work in collaboration with ornithologists studying the birds of

Madagascar to conduct a more comprehensive survey of the island's blood parasites.

Of several parasites that are seen commonly in the blood of birds, five were

considered. These include three from the phylum Apicomplexa: Plasmodium,

Haemoproteus, and Leucocytozoon. Trypanosoma are extracellular, flagellated protozoa

from the subphylum Mastigophora. Finally, microfilariae, the motile embryo from the

Phylum Nematoda can also be found in the blood depending on the species. Each of

these is vector borne, and has slightly different development. Species of Plasmodium,

Haemoproteus and Leucocytozoon have a roughly similar life cycle. When the

appropriate vector takes a blood meal from an infected host, the blood contains male and

female gametocytes. Once inside the stomach of the vector, the macrogametocytes

develop into macrogametes; while microgametocytes exflagellate and develop into

several microgametes, that then seek the macrogamete. They unite, becoming a diploid

zygote, which undergoes meiosis. The zygote elongates into the ookinete, which

penetrates the stomach wall of the vector, and becomes an oocyst. Within the oocyst,

haploid sporozoites develop. The mature oocyst ruptures, expelling sporozoites, which

will eventually migrate to the salivary glands. Upon stimulation of probing, the

sporozoites enter the acinae of the salivary glands and are injected into the host with the

saliva. If it is a susceptible host, the sporozoites will be carried by the circulatory system

to the reticuloendothelial system (RES), where they will develop into exoerythrocytic

schizonts. The schizonts mature, releasing merozoites, that can re-enter RES cells and

become a new generation of schizonts or enter the circulating blood cells. In the









circulating erythrocytes, these merozoites will develop into trophozoites, then

gametocytes; which they will remain until they are either ingested in a blood meal by an

arthropod, or cleared by the spleen.

Among species in these genera some life cycle differences exist that are important

for identification. For instance, merozoites of Plasmodium that enter circulating

erythrocytes will undergo a further generation of schizongony within the erythrocyte;

which will eventually rupture, releasing the merozoites. These merozoites can invade

other erythrocytes, and develop either into gametocytes or into schizonts. These

additional schizogonic stages function to increase parasite numbers, as merozoites can

develop into gametocytes at any stage. These erythrocytic schizonts make Plasmodium

unique in the fact that you can inject blood from an infected host into a naive susceptible

host and cause infection. Leucocytozoon spp. and Haemoproteus spp. only have

exoerythrocytic schizonts, usually in hepatocytes and vascular endothelial cells. The

number of times a parasite undergoes schizongony depends on that particular species.

Whereas Plasmodium is well known for causing malaria in humans, Trypanosoma

spp. are responsible for the well-known diseases Sleeping Sickness and Chagas disease.

Of course, trypanosomes infect animals as well; and can be found in the blood of birds in

the recognizable trypomastigote stage. Trypanosomes are also vector borne the arthropod

vector takes up the parasite during a blood meal and it develops in the arthropod. A few

days later, the arthropod takes another blood meal and defecates while feeding; and the

parasite enters the site of feeding. Some trypanosomes develop anteriorly and are

injected with the saliva of the vector. Usually, the flagellates appear in the feces in the

highest number when the insect is prepared to take another blood meal, increasing the









chances of transmission (Bennett, 1961). The parasite requires the vector to begin

digestion before multiplication of the parasite begins. That is, the parasite is somewhat

dependent on digestion (Bennett, 1961). Replication does occur. In the same study

mosquitoes that consumed approximately 155 trypomastigotes contained approximately

130,000 at 51 hours post blood meal (Bennett, 1961).

The fifth type of hematozoa found in the blood of the birds sampled is

microfilariae. Microfilariae are the motile embryos offilaroid nematodes. Not all

microfilariae are present in the blood; some are tissue dwelling. Those found in the blood

can survive in the blood for several years. They are also vector borne, and when a

suitable vector picks up the microfilariae, they develop into the L1 (or rhabditiform

larvae) that penetrate the midgut wall into the hemocoel. After two molts, they mature

into the filariform larvae (L3). These are the larvae that return back into the vertebrate

host upon feeding of the arthropod. In the vertebrate the larvae will molt twice more to

the adult form, will migrate to the preferred tissue, and the females will produce the next

generation of embryos. For reference, in humans, filarial infections cause river

blindness; and Bancroft's filariasis or elephantiasis.

Although scientists have been studying avian blood parasites for over 100 years,

relatively little is known about the pathogenicity associated with infection. One of the

more-studied areas is Hawaii, where Plasmodium relictum was introduced, and has been

affecting native and introduced species. Recently, Yorinks and Atkinson (2000)

investigated the effects of malaria on activity budgets of juvenile Apapane (Himatione

sanguinea). First, they found that the bite from one infected mosquito caused a fatal

infection (infection resulted in acute anemia) in five of their eight birds. Additionally,









they reported that the infected birds had a decline in several activities, becoming

essentially inactive at peak parasitemias. The authors noted that these infected animals

would have been at a competitive disadvantage against other birds in the area, and may

have been more susceptible to predation and heat stress (Yorinks and Atkinson, 2000).

This agrees with earlier findings, demonstrating canaries experimentally infected with P.

relictum suffer a drop in body temperature, indicating an inability to thermoregulate

(Hayworth et al., 1987). The authors hypothesize that this could increase mortality in

extreme environments. This also corresponds with other work in the area, where

moribund birds and birds killed by automobiles had higher prevalence of malaria than

mist netted birds, supporting the theory that infected birds were more likely to be killed

(van Riper et al., 1986).

Penguins have also suffered significant losses as a result of avian malaria. In fact,

it has been called the most important cause of death in captive penguins displayed in

open-air exhibits around the world (Stoskopf, 1979). In a five-month period from 1967-

1968, six African penguins at the Baltimore Zoo died as a result of malaria. By

inoculating tissue and blood emulsions into healthy birds, scientists were able to identify

Plasmodium elongatum. Fix et al. (1988) reported on 46 Magellanic penguins, 22 of

which died of malaria as the result of infection with Plasmodium relictum. At necropsy,

lesions typical of avian malaria were reported, including splenomegaly, hepatomegaly

and pulmonary edema. Additionally, exoerythrocytic schizonts were observed in

multiple tissues, including spleen, lung, liver, heart, brain and kidney.

Plasmodium is not the only genus associated with disease and mortality in birds.

Anemia and mortality in Pekin ducklings was associated with Leucocytozoon simondi,









although the anemia was not associated with peak parasitemia (Kocan and Clark, 1966).

A central nervous system disease in kestrels is also associated with a Leucocytozoon-like

parasitic infection (Raidal 2000). Circulating gametocytes were visualized and there

were tissue schizonts observed in vascular endothelial cells in the brain, but none were

observed in hepatic tissues, the typical location of schizogony in Leucocytozoon. Other

species of Leucocytozoon are known to have detrimental effects on their hosts.

Leucocytozoon caulleryi has been shown to so severely affect reproductive organs in

layer hens, resulting in the cessation of egg production (Nakamura et al., 2001).

The genus Haemoproteus is often thought of as relatively benign, although it too is

responsible for pathology and decreased performance. Haemoproteus melegridis has

been the target of several studies. Atkinson et al. (1988) reported reductions in growth

and weight gain in experimentally infected poults. There were four fatalities in that

study, and large ruptured megaloschizonts surrounded by infiltrate were found in each

(Atkinson et al., 1988a).

An interesting mystery in the field of avian blood parasites is the question of host

specificity. It has been believed that species of the genera Haemoproteus and

Leucocytozoon are specific at the host family level (Fallis et al., 1954; Baker, 1968;

Bennett et al., 1994). Because of this, many new parasites are named when previously

unstudied hosts are examined. Species of Plasmodium are much less specific and are

known to infect birds of different orders (Bennett et al., 1993). Microfilariae and

Trypanosoma spp. (Bennett, 1961; Fallis, 1973a) are also known to not have strong host

specificity. Therefore, the limiting factor in these cases is not the host, but the presence

of a suitable vector. Each of these parasites is vector borne, and these vectors each have









different behaviors and habitats, which will affect what birds they will encounter. Even if

birds were introduced that were infected with one of these parasites, they are not

increasing the odds of an epizootic unless there is a suitable vector present.

The preferred vectors are different for each parasite. The common vector for

species of Leucocytozoon and Trypanosoma are the Simuliidae (black flies). The

mosquito, Culicoides (Ceratopogonidae) and louse flies (Hippoboscidae) are the vectors

that transmit Haemoproteus spp. The Culicine mosquitoes vector Plasmodium species.

What is important then, is the understanding that behavior and habitat of birds will

influence their chances of being fed on by a vector of one of these parasites. Each of

these vectors varies in flight range, flight altitude, feeding times, habitat, breeding

requirements, and so on. Some may require brackish water for completion of their life

cycle, while others may need running water. So then, if a bird migrates from one habitat

daily, it may not become infected. For instance, if a bird spends evenings and nights in

high elevations, and moves down to lower elevations during the days, it will not be

exposed to those biting flies in the lower elevations that only feed at night.

Some of the factors we have elected to consider have been discussed before. One is

to analyze parasite prevalence by altitude. Van Riper et al. (1986) found that elevation

had a marked influence on parasitemia levels (based on 16 sampling stations at 300 m

intervals). They found that the highest parasitemia occurred between 900 and 1500 m in

elevation, where the vector and bird populations overlapped. It has been suggested more

than once that avian malaria is responsible for the dramatic decline in endemic Hawaiian

birds, and even their extinction (Warner, 1968; van Riper et al., 1986). With the thirteen

different altitudes ranging from 0 to 1950 meters, some with multiple sample sites at the









same altitude, we have a good possibility of identifying patterns of infection relative to

altitude. This can be used later in attempts to identify vectors of these parasites, as their

range should closely match the distribution of the parasite they vector. Additionally,

subsequent researchers can use the patterns of parasite prevalence to indicate what types

of vectors may be present.

Another factor to be examined is age, juvenile vs. adult. Van Riper (1986) found

that younger birds were not more likely to be parasitized than older birds, and that

younger birds had a higher parasitemia, possibly indicating lesser resistance. Using

Plasmodium circumflexum, Herman (1975) found that younger ducklings (one week or

younger) displayed a recognizable parasitemia earlier than older ducklings, while older

ducklings had a longer pre-patent period. In a seven-year period, Beier and Stoskopf

(1980) reported 16 first and second year juvenile penguins died of malarial infections, but

no adults died of malaria during the same time. This is of concern in birds that have a

long breeding interval, longer time to sexual maturity, or produce few offspring each

year. These are the populations that would not fare well if challenged with a pathogenic

parasitic infection.

This project is based on classical methods of parasite identification. This method is

still very important and as applicable as ever, although some researchers have voiced

concerns about misdiagnosing chronic, sub-clinical infections. Modern molecular

techniques have been applied as diagnostic tools, but the results have not been ideal.

Jarvi et al. (2002) reported that PCR tests underestimated chronic infections by 20%, but

were perhaps more applicable to longitudinal studies where repeated sampling is

occurring. Diagnosis of parasitism by reading blood smears is more applicable in field









situations, particularly in remote study sites. Researchers in remote areas have more

difficulties in handling vials for blood sampling, for later molecular diagnostics. It is

more realistic and applicable to make blood smears on glass slides. In addition, the old

and new systems need to be run in concert to validate the new approaches. Glass slides

are more compact, and are easily stored indefinitely as long as they are protected from

insects. Additionally, they serve as a permanent visual reference, and perhaps later can

be used in conjunction with molecular techniques.

It is the goal of this project to examine the prevalence of blood parasites in a

retrospective survey from the years 1994 and 1995. Furthermore, I expect to find species

of hematozoa previously undescribed, and possibly unique to the island. In addition to

the prevalence of parasites, I plan to identify the parasites found to the species level and

describe new species where possible. Finally, I hope to examine prevalence by sampling

site, habitat, altitude, gender, and family groupings to identify any relationships that may

exist. The goal of this project to make a strong first effort in teasing out the answers to

some of these questions.














CHAPTER 2
MATERIALS AND METHODS

Ornithologists working at field sites in Andohahela, Anjanaharibe-Sud,

Ambohitantely, and Montagne d'Ambre, Madagascar mist-netted birds at elevations of

120, 400, 440, 810, 875, 1000, 1200, 1260, 1400, 1500, 1550, 1875, and 1950 meters.

Data regarding age, gender, and presence or absence of brood patch were recorded at

capture when possible. Species were designated into one of three habitat preferences:

forest dependent (FDE), forest dweller (FDW), and forest edge (FED). Blood smears

were prepared in the field by toe clips or heart puncture. Blood was dropped onto a clean

glass slide and spread into a monolayer using a second slide. Slides were air-dried and

some were fixed in methanol at that time. Slides were then packed and shipped to the

University of Florida. On arrival, the unfixed slides were fixed in methanol and all slides

were stained with Giemsa, and stored. Slides were examined for the presence of

hematozoa on a Zeiss light microscope at 100x, 160x, and 1000x (oil immersion). Due to

varying quality, slides were examined for 30 minutes before being declared free of

parasites. Slides with blood parasites were examined on a Nikon compound microscope,

and parasites were drawn with the aid of a drawing tube. All measurements, excepting

parasite length, of erythrocytes and parasites were performed as described in Bennett and

Campbell (1972). Area was calculated using a drawing tube and grid, as described in

Forrester et al. (1977). To calculate nuclear displacement, the formula 2X/(X+Y) was

used, where Y is the distance between the periphery of the cell and the periphery of the

host cell nucleus on the side which the parasite occupies. In the case of circumnuclear






12


parasites, Y is calculated from the side on which all or most of the parasite nucleus lays.

X represents the distance between the host cell membrane and host cell nucleus on the

other side of the erythrocyte. One indicates no displacement, zero indicates total

displacement of host cell nucleus to host cell margin. Parasite length was determined by

measuring a line drawn to bisect the gametocyte along its longitudinal axis. All statistical

comparisons were made with SigmaStat, using either Chi-square or Fisher's Exact tests

to compare prevalences. Unless otherwise indicated, alpha = 0.05.














CHAPTER 3
NEW SPECIES DESCRIPTIONS

Madagascar is home to both endemic and broadly distributed avian families. Four

families, three endemic and one found more widely were investigated for the presence of

hematozoa. The Brachypteraciidae is endemic to Madagascar and encompasses three

genera and five species of Ground-rollers. They are found in tropical and subtropical

rainforest and arid thornscrub. All are medium sized terrestrial birds, feeding mainly on

small invertebrates and vertebrates encountered while foraging in leaf litter on the forest

floor. Three of the five species are threatened, mostly due to loss of habitat. This family

of birds requires undisturbed, pristine forest, which is being degraded by traditional slash

and burn agriculture, mining, and logging. Additionally, cattle grazing the understory

and well as hunting by humans for consumption are concerns (Langrand, 2001). The

Vangidae and Philepittidae are also endemic. The Vangidae encompasses 12 genera and

14 species, found mainly in forested areas and also savanna and subdeserts (Clements,

2000). Only Cyanolanius madagascarinus (the Blue Vanga) is found outside of

Madagascar, on Grand Comoro and Moheli Island (Langrand, 1991). In the Philepittidae,

there are two genera, each with two species. They are found in a variety of forests,

mainly in eastern Madagascar although Schlegel's Asity (Philepitta schlegeli) is found in

the dense forests of western Madagascar (Clements, 2000).

A fourth family, the Dicruridae, was also examined. It is made up of 20 species of

birds, occurring in Africa, India and Australia (Langrand, 1990). The Crested Drongo

(Dicrurus forficatus) is the only member of the family on Madagascar, and occurs in









Madagascar, and Anjouan in the Comoros. It can be found commonly in a variety of

habitats from forests to sparsely wooded terrain and plantations (Morris and Hawkins,

1998).

To date, three studies have examined birds from Madagascar for hematozoa

(Bennett and Blancou, 1974; Greiner et al., 1996; Raharimanga et al., 2002).

Raharimanga et al. (2002) reported hematozoa from Brachypteraciidae, but they were not

identified to the species level. Greiner et al (1996) reported a Leucocytozoon sp. from

both a Hook-billed Vanga (Vanga curvirostris) and a Velvet Asity (Philepitta castanea),

but did not describe them. Only one species of Haemoproteus has been reported in the

family Dicruridae. Haemoproteus dicruri was first described by de Mello (1935) from

Dicrurus macrocercus, and later redescribed by Peirce (1984b) from D. adsimilis.

I identified several new species of Haemoproteus and Leucocytozoon in these

families. The Brachypteraciidae had three new species of hematozoa, two of

Haemoproteus and one of Leucocytozoon. One species each of Haemoproteus and

Leucocytozoon were observed in Vangidae. One new species of Leucocytozoon is

described from Philepittidae, and a new species of Haemoproteus is described from the

Dicruridae. Several authors have discussed host specificity in both of these genera (Fallis

et al., 1954; Fallis et al., 1974; Atkinson, 1986; Bennett and Peirce, 1988; Bennett et al.,

1991). Based on this, the Haemoproteus and Leucocytozoon species described here are

considered to be new species, specific to their respective families.

Haemoproteus goodmani n. sp.

Immature gametocyte: Young parasites develop laterally to host cell nucleus in

mature erythrocytes, either in contact with or free from the host cell nucleus. Margins

sometimes slightly amoeboid.









Macrogametocyte: (n=23) Table 3-1, Figures 3-1(1), 3-1(2) Female gametocyte

halteridial, with smooth or slightly irregular or amoeboid margins. Parasite tips

commonly extend beyond erythrocyte nucleus towards limiting margin of the

erythrocyte. Erythrocyte nucleus displaced laterally, with an average NDR of 0.66. Host

cell nucleus not distorted however, maintaining the same average length and width

observed in uninfected erythrocytes. Host-parasite complex slightly larger in area than

uninfected erythrocyte. Area increases 6.2%, with parasite taking up 50% of host-

parasite complex and 60% of erythrocyte cytoplasm. Outer margin of parasite not

usually observed in contact with erythrocyte limiting membrane. Additionally, inner

margin of parasite often, not always, in contact with host cell nucleus. A vacuolated

cytoplasm, which does not stain deeply or evenly, results in a mosaic appearance.

Parasite nucleus not always discernible, often located terminally and rarely centrally.

From one to eight fine yellow refractile granules are rarely seen, and are terminal or

central. Volutin granules commonly seen scattered throughout the cytoplasm, and are

large with an average of 14 per gametocyte.

Microgametocyte: (n=12) Microgametocyte has the general morphology of the

macrogametocyte. Cytoplasm does not stain and appears white, with a large lightly pink-

staining nucleus centrally located. Parasite nucleus diffuse, occupying 25-50% of

microgametocyte.

Taxonomic Summary

Type host: Pitta-like Ground-roller (Atelornis pittoides), Lafresnaye, 1834,

Brachypteraciidae.

Type locality: Ambohitantely, Madagascar, latitude 180 04' to 18 14'S, longitude

47 12' to 47020'E.









Basis of description: Parasites are described from a blood smear taken from an

adult Atelornis pittoides (Pitta-like Ground-roller). HAPANTOTYPE: Blood smear AA-

83 Atelornis pittoides collected by Aristide Andrianarimisa on 13 October 1994 in

Ambohitantely, Madagascar at 1500 meters altitude. Accession G463728, IRCAH

Distribution: It is expected that this parasite will be found throughout the range of

the Ground-rollers on Madagascar.

Remarks

This parasite is medium sized and often, but not always, in contact with the host

cell nucleus. It causes only a slight increase in area of the infected cell, and slight

displacement of the erythrocyte nucleus.

Etymology

This parasite is named after Dr. Steven M. Goodman, biologist and ornithologist,

for his years of dedicated field work in Madagascar collecting information for biological

inventories, invaluable to Malagasy officials and conservationists worldwide.

Additionally, his steadfast efforts in making blood smears from birds for the evaluation of

hematozoa are recognized.

Haemoproteus forresteri n. sp.

Immature gametocyte: Young parasites lateral to host cell nucleus in mature

erythrocytes. Presses against the limiting membrane and host cell nucleus from an early

stage. Microhalteridial as immature, progressing through a thick halteridial phase before

reaching mature form. Terminals of developing gametocyte progress along periphery of

host cell nucleus, until they connect. Parasite then grows outward until entire host cell

cytoplasm is filled.









Macrogametocyte: (n=10) Table 3-1, Figure 3-1(3) Mature macrogametocytes

halteridial, becoming circumnuclear with the ends of parasite almost touching or

completely touching, rarely a thick gametocyte completely displacing host cell nucleus

against the host cell membrane. Host cell nucleus not distorted in length or width, but

has a NDR of 0.51. Infected host cell increased in length from 15 [im to 16.2 rim, but

decreases in width from 12.4 [im to 10 rim, and there is a 17.6% reduction in area.

Parasite occupies 64% of host-parasite complex, and 84% of cytoplasm. Parasite margins

generally smooth and occasionally amoeboid. Stains blue to light blue with Giemsa.

Parasite nucleus compact, lightly staining and pink in color, and sometimes indistinct.

When visible, gametocyte nucleus commonly touching outer periphery of parasite,

closest to host cell limiting membrane. Pigment granules not always observed and range

in number up to 12 when seen. They are fine and appear white or light yellow. Volutin

fine and dust-like, commonly seen accumulated at the ends of the gametocyte.

Microgametocyte: (n =11) Figure 3-1(4) Mature microgametocyte similar to

macrogametocyte in its displacement of host cell nucleus to margin and becoming

circumnuclear. Parasite nucleus sometimes large, but not diffuse and stains pink with

Giemsa.

Taxonomic Summary

Type host: Rufous-headed Ground-roller (Atelornis crossleyi), Sharpe, 1875,

Brachypteraciidae

Type locality: Anjanaharibe-Sud, Madagascar, 14 44.8'S, 490 26.0'E

Basis of description: Parasites are described from a blood smear from an adult

Atelornis crossleyi (Rufous-headed Ground-roller). HAPANTOTYPE: Blood smear SG-









200 collected by Steven M. Goodman on 27 November 1994 at 1950 meters in

Anjanaharibe-Sud, Madagascar. Submission G463729, IRCAH.

Additional hosts: Atelormis pittoides (Pitta-like Ground-roller)

Distribution: It is presumed that this parasite will be found throughout the range of

the Ground-rollers on Madagascar.

Remarks

This parasite differs from the only other Haemoproteus sp. in Brachypteraciidae,

namely H. goodmani, in that the mature gametocyte is circumnuclear or completely

displaces the host cell nucleus. Further, the inner margin of the parasite is usually

pressed firmly against the host cell nucleus throughout its development, unlike H.

goodmani. The nucleus ofH. forresteri is more compact and readily visualized than in H.

goodmani. Also, volutin granules ofH. forresteri are fine and dust-like found at the

terminus or periphery, where the volutin in H. goodmani is large and distributed

randomly throughout the gametocyte. Finally, H. forresteri causes hypertrophy of the

host cell.

This parasite is described with two predominant morphologies, which, while

uncommon is not unprecedented. Haemoproteus sacharovyi is described as pleomorphic

with multiple common forms. Specifically, in Bennett and Peirce's redescription (1990),

there are four forms reported.

Etymology

This parasite is named after Dr. Donald Forrester of the University of Florida, in

recognition of his significant contributions in the fields of parasitology and wildlife

disease.









Haemoproteus vangii n. sp.

Immature gametocyte: Trophozoites and young gametocytes usually centrally

located, lateral to the erythrocyte nucleus. Sometimes sub polar, near end of host cell

nucleus, but still lateral.

Macrogametocyte: (n=27) Table 3-1, Figure 3-1(6) Found in mature erythrocytes,

macrogametocyte microhalteridial to halteridial, and stains a moderate to light blue with

Giemsa. Macrogametocyte may appear to be more rod shaped, only slightly curved

along the host cell nucleus, not wrapping around. Parasite ends commonly do not reach

to ends of erythrocyte, but longer gametocytes can be observed. There is little distortion

of the host cell by this parasite. Area of the host-parasite complex is only slightly (5.2%)

larger than uninfected erythrocytes. Host cell length increases slightly from 15.8 jm to

18.3 jm. Host cell nucleus is not distorted in width or length, but is displaced laterally

with a NDR of .071. Parasite encompasses 48.9% of the host-parasite complex, and

58.7% of the host cell cytoplasm. The parasite abuts the host cell nucleus, but is not

always in contact with the host cell membrane, even at maturity. Parasite margins can be

smooth or slightly amoeboid, or a combination of both. Parasite nucleus is small,

averaging 6.2 jim2, typically sub-central, against or very near the outer margin of the

gametocyte. Parasite nucleus stains pink and is compact and dense in appearance.

Pigment granules are very fine, but still clearly refractile, and appear light yellow or

white and usually are scattered through the cytoplasm but occasionally clumped.

Number ranges up to 14 per cell, but the average was 7.

Microgametocyte: (n=20) Figure 3-1(5) Male gametocyte morphology as

described above. The gametocyte stains very lightly with Giemsa, often appearing clear.

Parasite margins frequently appear indistinct. The microgametocyte is as likely to have









amoeboid margins as the macrogametocyte. Nucleus stains lightly, and is larger than in

the macrogametocyte, with an average area of 17 im2.

Taxonomic Summary

Type host: Blue Vanga (Cyanolanius madagascarinus), Linnaeus, 1766, Vangidae

Type locale: Andohahela, Madagascar, 24o35.6'S, 4644.3'E

Basis of description: Parasites are described from a blood smear from an adult

Cyanolanius madagascarinus HAPANTOTYPE: Slide SG-551A collected by Steven M.

Goodman in Andohahela, Madagascar, 3 November 1995. Submission G763730,

IRCAH.

Distribution: It is presumed this parasite will be found throughout the range of the

Vangas on Madagascar, and possibly Grand Comoro and Moheli Island.

Additional Hosts: Tylas eduardi (Tylas), Leptopterus viridis (White-headed

Vanga)

Remarks

Haemoproteus vangii is the only species of Haemoproteus reported in Vangidae. It

becomes markedly microhalteridial, clearly cupping the erythrocyte nucleus. Fine

refractile granules and volutin are observed in H. vangii. The gametocyte margins of H.

vangii are commonly amoeboid. The author used the most mature forms of the parasite

for the species description, but some of the gametocytes used for measurements may not

have been fully developed. Therefore, the true area of the parasite may be slightly larger

than indicated here. Based on the uniformity of this parasite, the other parameters should

not be greatly affected.

Etymology

The parasite name is taken from nominate species of the host family, Vangidae.









Haemoproteus khani n. sp.

Macrogametocyte: (n=13) Table 3-1, Figure 3-1(7) Circumnuclear at maturity,

lightly staining blue with Giemsa. Nucleus commonly indistinct, light pink or clear when

visible. Margins smooth or slightly irregular, but not amoeboid. Fully developed host

cell- parasite complex area is 130.3 jim2, 8% larger than uninfected erythrocytes. Host

cell nucleus is displaced slightly, with an NDR of 0.66. Average parasite area is 110.6

jlm2, host cell nucleus is reduced from 22.4 to 19.7 jim2. Parasite length is 28.1 im.

Infected cells are slightly increased in length. When visible, parasite nucleus is 6.3 jim2

(6% of parasite). Refractile granules located randomly through parasite, sometimes

clumped. Pigment granules small, generally round and white or yellow in color, not dark.

Average number is 10.7. No volutin is observed. Parasite develops in mature

erythrocytes.

Developmental stages halteridial, ends wrapping around erythrocyte nucleus.

First circumnuclear contact between parasite ends occurs touching host cell nucleus, then

parasite grows outward filling last remaining host cell cytoplasm. Host parasite complex

at this stage is 10 jim2 larger than fully mature complex, but parasite has the same area,

indicating host cell shrinkage as parasite matures.

Microgametocyte: (n=7) Figure 3-1(8) Same morphological characteristics as

described above, with gender staining associated differences. Parasite stains virtually

clear, commonly only noticed as a result of the pigment granules. Parasite nucleus, pink

when observed, has average area of 27.6 jim2, 25% of parasite area.

Taxonomic Summary

Type host: Crested Drongo (Dicrurus forficatus), Linnaeus, 1766, Dicruridae

Type locale: Andohahela, Madagascar, 24o49.0'S, 4636.6'E









Basis of description: Parasites are described from a blood smear taken from an

adult Dicrurus forficatus (Crested Drongo) HAPANTOTYPE: Blood smear SG-604

collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at

120 m. Accession G763732, IRCAH. PARAHAPANTOTYPE: Blood smear SG-605

collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at

120 m. Accession IPM-11 to L'Institut de Pasteur de Madagascar.

Distribution: It is presumed that this parasite will be found throughout the range

of the Drongos on Madagascar, and possibly beyond.

Remarks

This is the first circumnuclear haemoproteid recorded from the Dicruridae. The

parasite has a characteristic development in that it generally first connects with the other

end of the parasite along the host cell nucleus margin, then grows together and outward

from there.

Etymology

The parasite is named in recognition of the significant body of work produced by

Rasul A. Khan, in particular his efforts in the study of hematozoa.

Haemoproteus dicruri

Macrogametocyte: Table 3-1, Figure 3-1(9) Halteridial gametocyte in mature

erythrocytes. Gametocyte fully displaces host cell nucleus laterally to erythrocyte margin

(NDR 0.01). Ends of parasite do not wrap around erythrocyte nucleus; they do not cross

the plane created by the opposite side of the erythrocyte nucleus. Parasite length is 15.9

lm. Host parasite complex slightly larger than uninfected erythrocytes with an area of

128 .m2. Parasite is 106.3 .im2, occupying 98% of host cell cytoplasm and 83% of the

host-parasite complex. Macrogametocyte stains light blue with Giemsa, parasite nucleus









stains pink. Parasite nucleus usually central, commonly against host cell nucleus.

Refractile granules are sometimes noticeably rod-like, with 11.9 per gametocyte.

Microgametocyte: As described above, with sexual differences. Microgametocyte

nucleus boundaries often indistinct.

Taxonomic Summary

Type host: Crested Drongo (Dicrurus forficatus), Linnaeus, 1766, Dicruridae

Type locale: Andohahela, Madagascar, 24o49.0'S, 4636.6'E

Basis of description: Parasites are described from a blood smear taken from an

adult Dicrurus forficatus (Crested Drongo) HAPANTOTYPE: Blood smear SG-604

collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at

120m. Accession G463733, IRCAH. PARAHAPANTOTYPE: Blood smear SG-605

collected by Steven M. Goodman on 10 December 1995 in Andohahela, Madagascar at

120m. Accession IPM-11 to L'Institut de Pasteur de Madagascar.

Distribution: It is presumed that this parasite will be found throughout the range

of the Drongos on Madagascar, and possibly beyond.

Remarks

This is a new host record for Haemoproteus dicruri. De Mello (1935b) originally

described this parasite from a Black Drongo, later Peirce redescribed it from a Fork-tailed

Drongo (1984b). De Mello described an ovoid, convex parasite with rod shaped pigment

granules. He describes lightly staining microgametocytes with indistinct nuclei, and

displaced erythrocyte nuclei (de Mello, 1935b). Peirce (1984b) also describes a pale

microgametocyte with an indistinct nucleus. Also, Peirce observed the displacement of

the host cell nucleus and stated that the parasite did not wrap around the host cell nucleus

or become circumnuclear. Here we observe the same qualities. We observed a parasite









that does not wrap around the erythrocyte nucleus, displaces the host cell nucleus, stains

lightly with an often indistinct parasite nucleus. While Peirce did not observe rod-shaped

pigment granules, we noted that the pigment granules are not always rod-shaped, and

rounder granules are common. Additionally, we observed that the parasite occupies more

than 90% of the host cell cytoplasm. This is slightly more than observed by Peirce

(1984b) and it is unknown if this is a result of host-induced variation or those observed

here were more mature parasites.

Haemoproteus dicruri is differentiated from H. khani largely by its length, nuclear

displacement ratio and by the tendency not to envelop the erythrocyte nucleus.

Haemoproteus dicruri seems to have a more rigid composition, being almost more rod

like. Haemoproteus khani causes little displacement of the host cell nucleus and grows

around it easily. Haemoproteus khani becomes circumnuclear as it becomes a fully

mature gametocyte, and is approximately 12 microns longer at maturity than H. dicruri.

The NDR of 0.01 associated with H. dicruri observed here readily supports the

morphology easily visualized, a host cell nucleus completely displaced laterally. By

comparison, H. khani has an NDR of 0.66, indicating only minor displacement by the

gametocyte.

Both parasites discussed here stain in a similar manner, and it is often easy to

overlook the microgametocytes unless the refractile granules are discerned. Care should

be taken when examining blood smears from Dicrurids with few mature gametocytes.

Species differentiation is based on the nuclear displacement ratio and the degree to which

the cap formed by the host cell nucleus encircles the host cell nucleus.









Leucocytozoon frasci n. sp.

Immature gametocyte: Young parasite displaces host cell nucleus to host cell

membrane very early in development. Possibly originates from a central position, and

grows outward laterally, filling host cell cytoplasm in most cases. Although hard to

determine conclusively, host cell appears to be an erythrocyte.

Macrogametocyte: (n=32) Table 3-2, Figure 3-2(1) Round morph most common,

but distortions also commonly encountered. Macrogametocyte stains blue with Giemsa

stain. Parasite nucleus clearly distinguishable, staining pink, and has an average area of

16.3 .m2. Host-parasite complex has an average area of 424.2 .im2, three times the area

of uninfected erythrocytes. While the parasite distorts the host cell, making the

identification of the host cell impossible, this comparison provides a relative scale for

comparison. Host cell nucleus forms a cap, in an irregular manner. Sometimes very

thick, other times stretched considerably around perimeter of parasite. Host cell nucleus

generally covers 40% of the perimeter of parasite, ranging from 30-53%. Host cell

cytoplasm is commonly recognizable within the host cell-parasite complex, occupying

3.4-38.5%.

Microgametocyte: (n=14) The male gametocyte has the same characteristics as

described for the macrogametocyte, with the usual gender-related staining differences.

The host-parasite complex is slightly smaller than seen in the macrogametocyte, with an

average area of 379 rim2, only 2.6 times the area of uninfected erythrocytes. It stains

lightly with Giemsa, and nucleus not always discernible. When visible, the nucleus is

diffuse with average area of 86.7 im2.









Taxonomic Summary

Type host: Rufous-headed Ground-roller (Atelornis crosslevi), Sharpe 1875; Scaly

Ground-roller (Geobiastes squamigerus), Lafresnaye, 1838, Brachypteraciidae

Type locality: Anjanaharibe-Sud (14 44.8'S, 49 26.0'E) and Andohahela

(24o37.6'S, 4645.9'E), Madagascar

Basis of description: Parasites are described from a blood smear from an adult

Atelornis crosslevi (Rufous-headed Ground-roller). HAPANTOTYPE: Blood smear SG-

150 collected by Steven M. Goodman on 27 November 1994 at 1950 meters in

Anjanaharibe-Sud, Madagascar. Accession G463734, IRCAH.

PARAHAPANTOTYPE: Blood smear SG-505A from a Geobiastes squamigerus (Scaly

Ground-roller) collected by Steven M. Goodman on 21 October 1995 at 400 meters in

Andohahela, Madagascar. Accession IPM-12 to L'Institut de Pasteur de Madagascar.

Additional hosts: Atelornis pittoides (Pitta-like Ground-roller)

Distribution: It is presumed that this parasite will be found throughout the range of

the ground-rollers on Madagascar.

Remarks

Leucocytozoon frasci is described from two birds of different species, and is the

only species known from Brachypteraciidae. Some measurements have a wide range

within the same individual. Comparing the proportions and measurements of the host-

parasite complex and uninfected erythrocytes between the two birds, minor variation was

observed, but no outstanding differences. The most noticeable trend was that the host

parasite complex was slightly larger (2.9 times larger than uninfected erythrocyte) in the

Scaly Ground-roller than in Rufous-headed Ground-roller (2.6 times larger). The

uninfected erythrocytes in both species were the same size.









Etymology

This parasite is named in recognition of Dr. Salvatore Frasca Jr., a former mentor

of the author for his support and guidance, and introducing her to parasitology.

Leucocvtozoon lairdi n. sp.

Macrogametocyte: (n=23) Table 3-2, Figures 3-2(2), 3-2(3) Host cell type could

not be determined due to the distortion by the parasite and there were no young parasites

observed, so measurements are relative to those of an uninfected erythrocyte. Only round

morph seen, staining blue with Giemsa. Parasite nucleus stains uniformly pink and

generally round with area of 13.7 im2. Host parasite complex is approximately 313.7

lm2, which is 2.5 times greater than an uninfected erythrocyte. The gametocyte

encompasses 68.5% of the host parasite complex, and is rather uniform in diameter,

averaging 19 rim, (18-21 inm.) The host cell nucleus is stretched into a band, forming an

irregular cap covering 42% of the parasite perimeter. Cap not stretched excessively. It is

relatively think and can be smooth or irregularly shaped along the free edge. A band is

sometimes seen lying over the top of the gametocyte. Orientation of parasite can make

this appear to be a split nucleus, but upon closer examination it can be seen that the host

cell nucleus is lying underneath the gametocyte.

Microgameotcyte: (n=15) Figure 3-2(4) Same morphological characteristics

described above, with the expected gender related differences. Stains light pink to pink

with Giemsa. Average area was 285.5 rim2 and nucleus varies in size, averaging 48.2

lm2. Parasite nucleus makes up 22.4% of the parasite.

Taxonomic Summary

Type host: Blue Vanga (Cyanolanius madagascarinus), Linnaeus, 1766;

Helmetbird (Euryceros prevostii), Lesson, 1830, Vangidae.









Type locale: Andohahela (2435.6'S, 4644.3'E) and Marojejy (14 25.6'S, 490

36.5'E), Madagascar.

Basis of description: Parasites are described from a blood smear from an adult

Cyanolanius madagascarinus HAPANTOTYPE: Slide SG-551A collected by Steven M.

Goodman in Andohahela, Madagascar, 3 November 1995. Submission G463731,

IRCAH. PARAHAPANTOTYPE: Slide MJR148 collected by Steven M. Goodman in

Marojejy, Madagascar, 15 October 2001. Accession IPM-13 to L'Institut de Pasteur de

Madagascar.

Distribution: It is presumed this parasite will be found throughout the range of the

Vangas on Madagascar, and possibly Grand Comoro and Moheli Island.

Additional Hosts: Leptopterus viridis (White-headed Vanga), Euryceros prevostii

(Helmetbird), Tylas eduardi (Tylas).

Remarks

This is the only species of Leucocytozoon known from the family Vangidae. The

macro- and microgametocytes of Leucocytozoon lairdi are described from two separate

birds, as no intact microgametocytes were observed on one slide, while the quality of the

other prevented full description of the parasite. It is felt that the small variation in

measurements is a result of host variation, and perhaps representative of gender

difference in this species.

Etymology

This parasite is named in honor of Marshall Laird, for his significant contributions

to parasitology.









Leucocytozoon greineri n. sp.

Immature gametocyte: Displaces the host cell nucleus almost immediately, seen

to develop in erythrocytes.

Macrogametocyte: (n=33) Table 3-2, Figures 3-2(5), 3-2(6) Stains dark blue with

pink nucleus with Giemsa stain. Round morph most commonly observed, although

distorted forms present. Host parasite complex is 1.8 times larger than that of an

uninfected erythrocyte with an average area of 224.4 im2. Parasite occupies 61.6% of

the host-parasite complex. The parasite has an average diameter (taken at widest point

where relevant) of 16.6 im. The parasite nucleus is compact and round, having an area

of 12.8 im2. This is 9.3 % of the parasite. The host cell nucleus is stretched into a cap

covering 38% of the circumference of the parasite. The cap size ranges, but usually well

within 27-56%. Typically, the cap is thick with a smooth margin somewhat rounded ends.

Host cell cytoplasm is commonly (50% of the time) associated with the complex. When

present it makes up 10.4% of the host-cell complex, 21.2 jim2 is the average.

Microgametocyte: (n=19) Same as described above, with gender staining

associated differences. Stains lightly with Giemsa, usually appearing light pink.

Microgametocyte nucleus is pale, sometimes with portions staining darker pink. Nucleus

may be too indistinct to visualize. Nucleus has an average area of 40.8 jim2, over three

times that of the macrogametocyte. This is also 29.5% of the microgametocyte.

Taxonomic Summary

Type host: Common Sunbird-Asity (Neodrepanis coruscans), Sharpe, 1875,

Philepittidae.

Type locale: Andohahela, Madagascar, 24o35.6'S, 4644.3'E









Basis of description: Parasites are described from a blood smear taken from an

adult male Common Sunbird-Asity HAPANTOYPE: Blood smear SG-551C Neodrepanis

coruscans collected by Steven M. Goodman on 3 November 1995 in Andohahela,

Madagascar at 810 meters. Submission G463735, IRCAH.

Distribution: It is presumed that this parasite will be found throughout the range of

the Asities on Madagascar.

Additional Hosts: Velvet Asity (Philepitta castanea)

Remarks

The unnamed Leucocytozoon sp. reported by Greiner et al. (1996) has been re-

examined and is the same as described here.

Etymology

This parasite is named in recognition of Ellis C. Greiner, for his contributions to

veterinary parasitology and the current knowledge of avian hematozoa.













Table 3-1. Morphometric variation in the haemoproteids from the Brachypteraciidae, Vangidae, and Dicruridae. All measurements in
microns or microns2
H. goodmani H. forresteri H. vangii H. khani H. dicruri


Host Family Brachypteraciidae Brachypteraciidae Vangidae Dicruridae Dicruridae


Infected RBC n= 34 20 48 20 12


HPC Area 140.9 (24.8) 120.3 (22.6) 124.6 (12.6) 130.3 (15.2) 128 (15.8)
HPC length 18.8 (1.5) 16.2(1.5) 18.3(1.4) 16.2(1.6) 15.9(1.7)
HPC width 10.9(1.2) 10.0(1.6) 9.6(1.2) 9.7(0.9) 9.8 (1.5)

Parasite Area 68.97 (18.7) 76.5 (11.2) 60.9 (7.1) 110.6 (15.9) 106.3 (17.8)
Parasite Length 19.1(2.2) 21.8(4.3) 16.4(1.7) 28.11(1.3) 16.1(1.73)
Macrogametocyte 8.0 (2.6) 12.2 (12.3) 6.2 (3.1) 6.3 (1.2) 8.1(2.9)
nucleus area
Microgametocyte 22.0 (3.5) 16.9 (7.5) 15.7 (9.98) 27.6 (14.6) not visible
nucleus area
RBC nucleus 7.3(1.1) 6.8(0.8) 7.3 (0.6) 6.7(0.9) 6.7(0.7)
length
RBC nucleus 4.1(0.6) 4.7 (0.7) 3.5 (0.6) 3.4 (0.5) 3.5 (0.7)
width
RCB nucleus area 23.97(5.4) 27.1 (5.1) 20.8(3.4) 19.7(1.7) 19.38 (2.98)
NDR 0.66 (0.3) 0.51(0.4) 0.71 (0.3) 0.66 (0.3) 0.01 (0.05)















Table 3-1 continued
H. goodmani H. forresteri H. vangii H. khani H. dicruri


Host Family Brachypteraciidae 3rachypteraciidae Vangidae Dicrridae Dicruridae


Uninfected RBC 25 22 29 15 15
n=
Length 16.2(1.1) 15.0(1.5) 15.8 (3.1) 15.4(1.1) 15.4(1.1)
Width 10.2 (0.9) 12.4 (2.2) 9.5 (0.8) 9.7 (0.6) 9.7 (0.6)
Area 130.1 (13.9) 142.7 (32.1) 118.4 (9.3) 120.4 (8.7) 120.4 (8.7)
Nucleus length 7.4(0.9) 7.1(0.7) 7.4(0.7) 7.1(0.7) 7.1(0.7)

Nucleus width 3.9(0.5) 4.6 (0.9) 3.6(0.6) 3.7(0.6) 3.7 (0.6)

HPC= host cell-parasite complex, NDR= nucleus displacement ratio











Table 3-2. Morphometric variations of the Leucocytozoon spp. of the Brachypteraciidae,
Vangidae, and Philepittidae. All measurements in microns or microns2
L. frasci L. lairdi L. greineri
Host Family Brachypteraciidae Vangidae Dicruridae
Infected RBC N= 39 38 53
Host cell-parasite -- 224.4 (49.7)
complex area
Macrogametocyte 424.2 (85.1) 313.7 (48.7) --
host parasite complex
Microgametocyte 379.1 (135.0) 285.5 (48.6) --
host parasite complex
Parasite area 268.2 (99.8) 215.0 (42.0) 138.2 (26.8)
Macrogametocyte 16.3 (5.2) 13.7 (5.7) 12.8 (3.8)
nucleus area
Microgametocyte 86.7 (113.7) 48.2 (33.6) 40.8 (23.2)
nucleus area
Host cell nucleus area 88.98 (27.4) 61.6 (19.9) 83.9 (20.6)
Residual host cell -- 21.2 (19.97)
cytoplasm
Nuclear cap ratio 0.395 (0.1) 0.42 (0.1) 0.38 (0.09)

Uninfected RBC N= 55 44 41
length 16.4 (1.3) 15.8 (3.1) 15.9 (1.0)
width 11.3 (1.1) 9.5 (0.8) 9.7 (0.8)
area 143.5 (15.6) 126.1 (14.99) 124.5 (11.3)
nucleus length 7.4 (0.6) 7.4 (0.7) 8.1(0.7)
nucleus width 4.8 (0.7) 3.6 (0.6) 3.9(0.5)
nucleus area 27.3 (3.6) 23.1 (3.23) 27.4 (3.1)
















4A,5$
a(>IP: ii
7 ,I .,aL


Figure 3-1. Haemoproteus spp. of the Brachypteraciidae, Vangidae, and Dicruridae. (1,2)
H. goodmani macrogametocytes, (3) H. forresteri macrogametocyte, (4) H.
forresteri microgametocyte, (5) H. vangii microgametocyte, (6) H. vangii
macrogametocyte, (7) H. khani macrogametocyte, (8) H. khani
microgametocyte, (9) H. dicruri macrogametocyte.


IPS


























Figure 3-2. Leucocytozoon spp. of the Brachypteraciidae, Vangidae, and Philepittidae.
(1) L. frasci macrogametocyte, (2,3) L. lairdi macrogametocytes, (4) L. lairdi
microgametocyte, (5,6) L. greineri macrogametocytes.














CHAPTER 4
SURVEY RESULTS

One quarter (24.3%) of the 378 birds surveyed was parasitized by at least one

species of hematozoa. Seventeen of the 21 families had infected birds. Prevalence by

family was statistically different for families with at least one infected bird. At the host

species level, 32 of 43 species had at least one infected representative (Table 4-1). All

specimens of two species were infected by the same parasite respectively. For example,

all seven Accipiter francesii were infected with Leucocytozoon toddi, and 10 of 10

Ploceus sakalava were infected with Haemoproteus quela. In families where at least 10

individuals were sampled, the Zosteropidae and Ploceidae were the most parasitized

families, with a prevalence of 56% and 34% respectively. With reference to the

prevalence of the parasite genera, Haemoproteus spp. were present in 14%,

Leucocytozoon spp. in 11%, microfilariae in 6%, Plasmodium spp. in 2%, Trypanosoma

spp. in 1.1%, and unidentified Babesia sp. in 0.5%. Microfilariae in birds rarely have

been identified to the species, or even genus level, and these are the motile embryos of

filaroid nematodes. All of the others are single celled eukaryotic parasites.

Prevalence by host family

There was a significant (P < 0.001) difference in prevalence of parasitism by

family when all families were compared, and when families with ten or more birds

sampled were compared. The following parasites were identified:

ACCIPITRIDAE: Each of the seven hawks sampled were infected with Leucocytozoon

toddi, but no other infections were seen.









ALCEDINIDAE: One of eight kingfishers was parasitized with Haemoproteus halcyonis.

BRACHYPTERACIIDAE: Seven of nine ground-rollers sampled were infected with at

least one species of hematozoa. Four birds had multiple infections. Five birds were

infected with Haemoproteus spp.; three with H. forresteri, two with H. goodmani.

Leucocytozoon frasci is reported from five birds. Trypanosoma avium was present in

one bird. Microfilariae from an unidentified nematode were observed from one bird.

This is the first description of hematozoa from this family, and each species listed here

are new taxa, specific to the Brachypteraciidae.

DICRURIDAE: Two species of Haemoproteus, H. khani and H. dicmri, were observed in

three drongos. Additionally, microfilariae were observed in two birds. Haemoproteus

khani, a new species of haemoproteid in the drongos was described. Trypanosoma avium

was recorded from one bird.

MONARCHIIDAE: Two of forty-eight monarch flycatchers appeared to be infected with

an unidentified species of Babesia. No other hematozoa were observed in any of the

other individuals sampled.

MOTACILLIDAE: Two of three wagtails were infected with Haemoproteus anthi.

NECTARIINIDAE: One sunbird was infected with Haemoproteus sequeirae.

PHILEPITTIDAE: Four asities were infected with Leucocytozoon greineri, and

microfilariae were observed in five birds. This is the first species of the Haemosporida

reported from this family, and represents a new species.

PLOCEIDAE: Three species of Plasmodium were seen in three weavers. Two were

infected with P. rouxi, while a third was infected with both P. nucleophilum and P.

relictum. The same species of Haemoproteus, H. guelea, was observed in 15 birds. Six









birds were infected with Leucocytozoon bouffardi. Two birds were infected with

microfilariae. A single Trypanosoma everetti was observed in one bird.

PYCNONOTIDAE: Plasmodium rouxi was observed in one bulbul. Two species of

Haemoproteus were observed, H. sanguineus was observed in each of the three birds

infected with this genus. One bird had a concurrent infection with H. philippinensis.

Two species of Leucocytozoon were observed. Three bulbuls were infected with L.

brimonti and two more with L. pycnonoti. Three birds were observed to have

microfilariae.

STRIGIDAE: Hematozoa were observed in three of nine owls. Haemoproteus noctuae

and H. srynii were each observed in one bird, while the haemoproteid in a third was too

distorted for conclusive identification. Two birds were infected with L. ziemanni.

SYLVIIDAE: Four Old-world warblers were infected with H. wenyoni. Leucocytozoon

phylloscopus was observed in one bird. Microfilariae were observed in two birds.

TIMALIIDAE: One babbler was infected with microfilariae.

TURDIDAE: Schizonts of Plasmodium vaughani were observed in one thrush, and two

others were infected with Plasmodium rouxi. Haemoproteus minutus was reported from

three birds, H. fallisi from two others, and distortion prevented the positive identification

of two additional infections (one mixed with a H. fallisi infection). One bird was infected

with Leucocytozoon majors. Microfilariae were observed in three birds.

VANGIDAE: Four of five vangas were infected, each of which with multiple infections.

One bird was infected with Plasmodium rouxi. Two birds were infected with

Haemoproteus vangii. The three birds infected with Leucocytozoon sp. each harbored L.

lairdi. Three birds were infected with microfilariae and two more were infected with T.









avium. This represents the first examination of this family for the presence of

hematozoa, and the Haemoproteus and Leucocytozoon species mentioned here represent

new taxa.

ZOSTEROPIDAE: Only two genera of hematozoa were observed. Eight White-eyes

were infected with H. zosteropis and six had L. zosteropis.

Altitude

When prevalence was compared with altitude of bird collection, (120 m, 400 m,

800 m, 1000 m, 1200 m, 1400 m, 1500 m, 1875 m, 1950 m), there was a significant

(P<0.001) difference between groups. Pair-wise comparisons revealed three groups that

were different. At 120 m, prevalence was 75% with 15 of 20 birds parasitized. At the

highest elevation (1950 m) prevalence was 57%, with 4 of 7 birds parasitized. Finally,

the third group (1200 m) had a low level of parasitism, with 2 of 32 birds (6%) infected.

Because of the small sample sizes, the altitudes were consolidated into three ranges

and re-examined. The ranges were low altitude (120-800 m), medium altitude (1000-

1400 m) and high altitude (1500-1950 m). Prevalence was 36%, 19% and 22%

respectively. These ranges were then compared, and were found to differ from one

another, so once again pair-wise comparisons were performed. The low altitude group

was found to be statistically different from the others.

Reserves

The four regions from which the birds were sampled were compared to determine if

the prevalence of parasitism was greater in any of them compared to the others. When all

four were compared, a statistical difference was observed (P = 0.042). Prevalence for

Ambohitantely, Anjanaharibe-Sud, Andohahela, and Montagne d'Ambre were 21%

(26/124), 21% (17/82), 34% (44/131) and 20% (8/41) respectively. Regions were then









compared in pairs. Andohahela was statistically different from Ambohitantely (P =

0.023). When compared to Anjanaharibe-Sud, the p-value was 0.062. Compared to

Montagne d'Ambre, it appeared that there was no statistical difference (P = 0.129), but

the power of the test was very low (0.311) indicating a 69% chance that no difference

will be found when in fact there was a difference between the groups. No statistical

significance was observed between the other groups.

Habitat

Birds were classified into three groups: FED, FDE, and FDW. FED species are

forest edge species, living at the junction of forest and savanna habitats, and will cross

open, non-forested areas. FDE species are forest dependent species, and probably never

cross non-forested areas. Finally, FDW species are the forest dwelling species. These can

be found in the center of a forest, or in the margin between forest and savanna, and

almost certainly cross open areas. The majority of the species sampled were either forest

dwelling or forest dependent species. Prevalence for each habitat was as follows: FED

31%, FDW 27%, and FDE 19%. Groups were compared to determine any trends in

parasitism. There appeared to be no difference in prevalence observed in our study.

Gender

A total of 76 females and 111 males were conclusively identified. No gender was

recorded for birds where plumage was not indicative of gender, or birds that escaped

before the data could be obtained. There was no significant difference in prevalence of

parasitism between males and females.









Age

Birds were classified as either "adult" or "juvenile" by field ornithologists.

Prevalence by age was 23% for adults (65/284) and 8% (2/26) forjuveniles. No

significant difference was found in prevalence of parasitism between the two cohorts.

Breeding Condition

For the purposes of this survey, birds were considered to be in breeding condition

when a brood patch was observed, indicating the birds were actively incubating eggs.

Seventy-seven birds had brood patches, while 288 did not. Prevalence of parasitism was

27% and 23% for each group, respectively. There was no significant difference in

parasitism between groups.







42


Table 4-1. Prevalence of hematozoa in the avifauna of Madagascar, by avian family.


Family & species


Total Infected Plas. Haem. Leuc Micro Tryp Babesia
birds birds


Accipitridae
Accipiter francesii

Alcedinidae
Alcedo vintsiodes
Ipsidina madagascariensis

Brachypteraciidae
Atelomis pittoides
Atelomis crosslevi
Geobiastes squamigerus

Campephagidae
Coracina cinerea

Columbidae
Streptopelia picturata

Cuculidae
Cuculus rochii

Dicruridae
Dicrurus forficatus

Leptosomidae
Leptosomus discolor

Monarchiidae
Terpsiphone mutata
Total
% infected

Motacillidae
Motacilla flaviventris

Nectariniidae
Nectarina notata
Nectarina souimanga
Total
% infected


7 7


1 0


2 0


1 1


8 4


1 0


2 1


4.2 4.2


2 2


0
1 1
1 1
10 10










Table 4-1 continued
Family & species


Philepittidae
Neodrepanis coruscans
Neodrepanis hypoxantha
Philepitta castanea
Total
% infected

Ploceidae
Foudia madagascariensis
Foudia omissa
Ploceus nelicourvi
Ploceus sakalava
Total
% infected


Total Infected Plas. Haem. Leuc. Micro Tryp Babesia
birds birds


3 1
1 0
22 4
26 5
19.2


8 2
30 6
8 1
10 10
56 19
33.9


1

3 5
4 5
15.4 19.2


1 1
4 4
1
2 10 1 2 1
3 15 6 2 1
5.4 26.8 10.7 3.6 1.8


Pycnonotidae
Hypsipetes madagascariensis 23 7
Xanthomixis cinereiceps 8 0
Bemieria madagascariensis 25 4
Xanthomixis zosterops 22 0
Total 78 11
% infected 14.1

Strigidae
Otus rutilus 9 4


Sylviidae
Dromaeocercus brunneus
Harterula flavoviridis
Nesillas typical
Newtonia amphichroa
Newtonia brunneicauda
Total
% infected


Threskiornithidae
Lophotibis cristata


1 1
1 0
15 3
11 1
11 1
39 5
12.8


1 2


3 1


1 2 2

1 3 5 3
1.3 3.9 6.4 3.9


3 2


1

2 1
1 1
1
4 1 2
10.3 2.6 5.1


1 0


Timaliidae
Mystacomis crosslevi 1 1
Oxylabes madagascariensis 10
Total 11 1
% infected 9.1










Table 4-1 continued
Family & species


Total Infected Plas. Haem. Leuc. Micro Tryp Babesia
birds birds


Turdidae


Copsychus albospecularis
Monticola sharpei
Total
% infected


15 3
22 5
37 8
21.6


2 2
3 3 1 1
3 5 1 3
8.1 13.5 2.7 8.1


Vangidae
Cyanolanius madagascarinus 1
Tylas eduardi 2
Vanga curvirostris 1
Leptoperus viridis 1


Zosteropidae
Zosterops maderaspatana
Total
% infected


TOTAL


1 1 1
1 2


1 1


16 9
16 9
56.3

378 92


1.6 13.8 11 6.1 1.1 0.5


8 6
8 6
50 37.5


6 52 40 23 4 2


24.3














CHAPTER 5
DISCUSSION

There have been many studies to examine the prevalence of avian hematozoa

throughout the world. In the larger surveys, researchers have examined birds of the

Neotropics (White et al., 1978), North America (Greiner et al., 1975), eastern and

southeastern Asia (McClure et al., 1978), and central Europe (Kucera, 1981). Overall

prevalence varied considerably in these areas, from 10.5% in the Neotropics, to 36.9% in

North America. In each of these surveys, the most commonly observed genus of

hematozoa was Haemoproteus. A great deal of work has also been done in Africa, India

and island groups in these regions (de Mello, 1935a; de Mello and Fonseca, 1937;

Bennett et al., 1974; Bennett and Herman, 1976; Bennett et al. 1978; Peirce et al. 1977;

Peirce and Feare, 1978; Peirce, 1979; Peirce, 1984a; Peirce, 1984b; Bennett, 1990). This

depth of knowledge and accumulation of information over time is invaluable to current

and future investigations in this region. Many new species have been described in this

region, and others redescribed over time, owing to the diversity of the birds throughout

Africa and India. In 1990, Bennett reported 78 valid species of Haemoproteus,

Leucocytozoon and Plasmodium from the Indian subcontinent alone.

While a great deal of work has been done on the continent of Africa, relatively little

work has been done in Madagascar, a major continental island off the coast of

Mozambique. Three papers examining the avifauna for hematozoa have been published

to date. Bennett and Blancou (1974) examined 64 individuals from thirty-two species.

Of those, 14 (22%) birds were infected. Interestingly, Bennett and Blancou did not









observe any species of Haemoproteus, rather, Leucocytozoon was the most commonly

observed genus of blood parasite. Bennett and Blancou (1974) concluded that the

prevalence of hematozoa on the island was low, and that there was no indication of

unique species of hematozoa. The results of this survey indicate that these conclusions

are not the case, and that the prevalence of avian hematozoa falls within the range

reported elsewhere. Additionally, the endemic families and species were seen to harbor

new and distinct species of Haemoproteus and Leucocytozoon. Bennett and Blancou

(1974) were unable to draw conclusions on parasitism by family as a result of few birds

being sampled from each family. Greiner et al. (1996) reported six of 10 birds sampled

from southeastern Madagascar to be parasitized. In addition to the higher overall

prevalence, they also reported the presence of the genus Haemoproteus for the first time,

in addition to Plasmodium, Leucocytozoon and microfilariae. To have 60% of the birds

parasitized is a high number for such a small group, but the presence of four types of

hematozoa observed also speaks to the richness of all the forms of life in Madagascar.

Parasites in that study were identified only to the genus level. Most recently,

Raharimanga et al. (2002) examined 387 Malagasy birds and reported a prevalence of

35.9% overall. They sampled birds in 1995, 1996, and 2001 at six different sites. Again,

parasites were not identified to the species level. In this study, birds were collected in the

years 1994 and 1995, and four different sites were sampled. Additionally, parasites were

identified to the species level, providing a clearer understanding on the hematozoa in

Madagascar.

These investigations have been important as a means of establishing the presence

of avian hematozoa in multiple species, and over several years. The current investigation









represents the first large-scale survey of the hematozoa of the avifauna on Madagascar

including identification of the parasites observed. This information will be vital to future

investigations in Madagascar, and for more complete understanding of the biology of the

avifauna there.

This survey was intended to investigate the prevalence of hematozoa by host

family, and by several other parameters including age, gender, altitude, sampling site, and

habitat. Breeding condition (classified as here by the presence or absence of a brood

patch) was also considered. All birds included here were wild caught, none raised or held

in captivity. Additionally, the birds sampled here are from non-migrating, resident

populations on the island.

One quarter (24.5%) of the birds surveyed was parasitized by at least one species of

hematozoa. As mentioned earlier, reported prevalence in Madagascar has varied widely,

from 22% (Bennett and Blancou, 1974), to 60% (Greiner et al., 1996) and 39.5%

(Raharimanga et al., 2002). Prevalence in nearby areas varied greatly as well. For

instance, in the Mascarene Islands, Peirce (1979) reported 42% and 27% of the birds

were parasitized, while in Zambia he reported a prevalence of 48.2%, and finally he

reported a 12.7% prevalence through Africa (Peirce, 1984a; Peirce, 1984b). Bennett et

al. (1974) reported 24% of the birds examined in Uganda were parasitized, while in

Senegal only 11.5% were parasitized (Bennett et al., 1978).

In our study, 17 of the 21 families had infected birds. One or two birds only

represented each of the four families that was not seen to be infected. It is likely that

with a larger sample size, parasites would have been observed in these families, and

certainly further investigation is warranted. When compared, a statistical difference in









parasitism by avian family was found for the families that had at least one infected bird.

Further, there was a statistical difference in parasitism for families with at least 10 birds

sampled. The Zosteropidae and Ploceidae were the most parasitized, with a prevalence

of 56.3% and 33.9% respectively. These families were also the most parasitized in

Ugandan birds, as reported by Bennett et al. (1974). Other surveys have also reported a

variation in prevalence by avian family (Greiner et al., 1975; Bennett and Herman, 1976,

White et al. 1978). Variation by family or species may be a result of preferential feeding

by the vectors. Two studies have commented that some birds appeared to be more

attractive than others (Bennett and Fallis, 1960; Greiner et al., 1978).

Prevalence of parasite genera was also compared by avian family, for families with

twenty or more birds sampled. Two of these families, Monarchiidae and Philepittidae,

were not seen to be parasitized with species of Haemoproteus. There was a statistical

difference in prevalence of Haemoproteus spp. both with and without these families

included. Pair-wise comparisons revealed that two families were statistically different

from one another, the Ploceidae (26.8%) and Pycnonotidae (3.8%). Neither of these

families was statistically different from the remaining families in the comparison. There

was no difference in parasitism by the other genera of parasite by family.

On the species level, 32 of 43 species had at least one infected representative.

Some of these families were also represented in prior work. Bennett and Blancou (1974)

also examined two Streptopelia picturata, and like this survey, no parasites were seen in

this species. They examined 15 Fodys (Foudia madagascariensis) in 1974, and 4 were

infected with Leucocytozoon fringillinarum. Raharimanga et al. (2002) reported the









presence of either Plasmodium or Haemoproteus (not evaluated separately) or

Leucocytozoon in several of the same species examined here.

The prevalence of the parasite genera was similar to that reported elsewhere, with

species of Haemoproteus being the most commonly seen genus.

Families

Prevalence was not analyzed for families represented by fewer than 10 samples.

Three families included in this survey are endemic to Madagascar. These are

Brachypteraciidae, Philepittidae, and Vangidae. The Brachypteraciidae, or Ground-

rollers, were primarily parasitized by Haemoproteus and Leucocytozoon species.

Additionally, microfilariae were seen in one individual. Plasmodium spp. were not

observed. Raharimanga et al. (2002) examined five birds and reported the presence of the

same hematozoa. Four of their five birds were infected with either Plasmodium or

Haemoproteus, one with Leucocytozoon, one with Trypanosoma avium, and two with

microfilariae. In the Vangidae, four of five birds were parasitized and all had multiple

infections. Interestingly, species of Plasmodium, Haemoproteus, Leucocytozoon,

microfilariae and Trypanosoma were observed in these birds. In previous studies,

Greiner et al. (1996), reported the one Vanga examined to be infected with both a species

of Leucocytozoon and microfilariae. Out of 11 Vangas, Raharimanga et al. (2002)

reported four infections each of Plasmodium or Haemoproteus, Leucocytozoon and

microfilariae. While the samples sizes discussed are too small to make inferences with

regard to parasitism in the population, there is certainly a trend indicating high levels of

parasitism warranting further investigation. Finally, Philepittidae, the Asities, are small

birds endemic to Madagascar. There are four species; three have been included in this

sample of 26 individuals. The prevalence of parasitism is 19.2%. The same small









Leucocytozoon was noted repeatedly. Microfilariae were present in several individuals.

Because Plasmodium is typically rarely seen, it is not surprising that none were observed

here. It is interesting that no haemoproteids were noted. The genus Haemoproteus is

typically the most common hematozoon observed in surveys throughout the world. The

presence of Leucocytozoon greineri indicated that the family was susceptible to

parasitism. Greiner et al. (1996) reported the presence of an unidentified species of

Leucocytozoon and microfilariae. This slide was re-examined and the Leucocytozoon

was recognized to be the same species reported here and is considered L. greineri. The

lack of infection with haemoproteids warrants further investigation. Because it is an

endemic family and not present outside of Madagascar, it is possible that no species of

Haemoproteus has evolved to fill this niche. It is also possible that no appropriate vector

exists in the locales sampled. The parasites found in these three families represent new

taxa, and for the first time species names have been designated. This survey, then,

represents the first investigation of parasitism in these families to the parasite species

level.

The Monarchiidae examined in this survey are represented by one of the two

species of Flycatchers on Madagascar. Forty-eight Madagascar Paradise Flycatchers

(Terpsiphone mutata) were sampled and examined for the presence of hematozoa. These

birds are common in all forest sites, seen in degraded areas and in open areas, and have a

wide altitudinal range (sea level to 1600 m) (S.M. Goodman, pers. comm.; Morris and

Hawkins, 1998). Because of the wide range of habitats and elevations in which these

birds were sampled, one might expect that at least a portion of the birds would exhibit an

infection of some type. Interestingly, none of the birds sampled here was parasitized by









any of the expected genera. What appeared to be an undescribed Babesia species was

observed in two birds, but further examination of this family is needed to fully appreciate

this phenomenon, as ticks were noticeably absent from birds collected (S.M. Goodman,

pers. comm.). Based on the parasitism seen in other families, it was known that

ornithophilic vectors do exist on the island, at the same ranges. Greiner et al. (1996)

reported the two T. mutata included in their sample to be infected. The schizont of a

species of Plasmodium was observed in one, while the other was infected with either

Plasmodium or Haemoproteus, based on seeing only pigmented trophozoites.

Raharimanga et al. (2002) reported that three of 20 members of this family sampled were

infected with a species of either Plasmodium or Haemoproteus. The authors did not

observe any erythrocytic schizonts, all of the infected birds were T. mutata adults (V.

Robert, pers. comm.). Based on this information, it appears that a species of

Haemoproteus infective to the Flycatchers is present in Madagascar. Further examination

of this family is warranted, to determine the if the parasites are similar to those reported

for this family from other locations, or if the species is new and possible only found in

the Malagasy Flycatchers. Members of the same avian genus sampled in other

geographical areas were parasitized. Peirce (1984) reported one of seven Terpsiphone

viridis examined was infected with a Haemoproteus sp., one of two T. viridis and one of

seven Batis molitors were infected with Plasmodium in Zambia. Based on these reports,

we can conclude that the family is susceptible to parasitism. As yet, it was not apparent

that this species exhibited any behavioral adaptations that would preclude infection. In

considering the difference between the results of our survey and of the 2002 project, it is

likely that the vector for the haemoproteid reported by Raharimanga et al. (2002) is not









found throughout Madagascar, but more likely in isolated areas. It is also possible that

there were chronic infections existing at a subclinical level in the birds sampled for our

survey. These infections can result in a very low-level parasitemia that may go

undetected on microscopic examination.

Two species of thrushes (Turdidae) were represented in this survey, with a total of

37 birds. The family had a prevalence of 21.6% overall for parasitism, and had the

highest prevalence of Plasmodium, with 3 of 37 (8%) birds infected with species in this

genus. Schizonts consistent with Plasmodium vaughani were observed in one bird and P.

relictum-like gametocytes were noted in two others. Developing schizonts were seen, but

were not useable for diagnosis. Haemoproteus spp. were the most common parasite in

the family, with both H. minutus and H. fallisi recognized. Leucocytozoon majoris and

microfilariae were also noted. No parasites were seen in the single thrush examined by

Bennett and Blancou (1974). Raharimanga et al. (2002) examined blood smears from 45

individuals, and although no Leucocytozoon species were observed, 14 (31%) were

infected with either Plasmodium or Haemoproteus spp., and seven had microfilariae

circulating at the time of sampling.

The family with the highest prevalence in this survey was the Zosteropidae, or the

White-eyes. Of the 16 birds sampled, nine (56.3%) were infected with at least one

species of either Haemoproteus or Leucocytozoon. Five birds were infected with both

genera, and the two species of parasite observed were H. zosteropis and L. zosteropis.

The prevalence of both Haemoproteus (50%) and Leucocytozoon (37.5%) are similar to

those reported by Raharimanga et al., (2002). They reported 64% (11/17) birds were

infected with either Plasmodium or Haemoproteus, and 36% were infected with









Leucocytozoon sp. Additionally, they observed birds were infected with microfilariae.

Although Bennett reported this family to be highly parasitized in Uganda, only 15 birds

were included in his sample, and again, further investigation is necessary (Bennett et al.,

1974). All three sample sizes are small, but earlier work on Aldabra Atoll examined a

slightly larger group of Malagasy White-eyes reported 38 of 48 individuals infected with

elongated gametocytes of a Haemoproteus sp. Also, 14 of these birds were infected with

erythrocytic schizonts. Definitive identification was not possible as no gametocytes were

seen (Lowery, 1971). Finally, in the Mascarene Islands, Peirce et al. (1977) sampled a

total of 100 members of this family, 74 were infected with hematozoa. Peirce (1979)

reported four of four White-eyes to be infected with Leucocytozoon, after reviewing a

another sample from the same area. Further examination of this family in Madagascar

will be important in determining if the prevalence remains as high with more animals

examined.

The weaver family, Ploceidae, was the second most parasitized family in this

survey, with a prevalence of 33.9% in 56 birds. Four species were represented, with 8-30

birds from each. Multiple birds were infected with Plasmodium, Haemoproteus,

Leucocytozoon, and microfilariae. Plasmodium rouxi, P. nucleophilum and P. relictum

were observed, in addition to H. quelea and L. bouffardi. Trypanosoma everetti is known

to infect the Ploceidae, and was observed in a single Ploceus sakalava (Bennett et al.,

1994). Interestingly, each of the 10 P. sakalava sampled was infected with H. quelea,

while only five of the remaining 46 weavers were infected with it. One possible reason

for this may be related to the behavior of this species of bird. The three other species

sampled are solitary breeders, while P. sakalava forms dense colonies (S.M Goodman,









pers. comm.). This has been seen in this family before. In a survey from Senegal,

Bennett et al. (1978) reported the prevalence of parasitism in non-colonial nesting

ploceids was 3.1%, but in the colonial nesting ploceids, prevalence was substantially

higher at 21.3%. Each of the prior investigations in Madagascar reported parasitism in

this family from each of the parasite genera observed here (Bennett and Blancou, 1974;

Greiner et al., 1996; Raharimanga et al., 2002). Because Bennett and Blancou (1974) as

well as Greiner et al. (1996) reported positive individuals from small samplings (15 and 2

birds, respectively), this supports our observation of the high prevalence of parasitism in

this family. If the actual prevalence of parasitism in this family were low, then small

sample sizes would be less likely to contain infected individuals. There are other

indications that this family may be more susceptible to parasitism, either through their

behavior, or greater abundance of the vectors. For example, Bennett et al. (1974) also

reported that the ploceids were the second most parasitized family surveyed in Uganda.

They detected at least one species of hematozoa in the 116 of 206 weavers examined. In

the Mascarene Islands, 17% of the Madagascar Fodys were infected with either

Leucocytozoon or Plasmodium. For the Rodrigues Fody sampled at the same time, the

prevalence was 31% (Peirce 1979).

With 78 birds captured, the Pycnonotidae or Bulbuls, had the largest sample. Four

species were obtained, and the overall prevalence for the family was 14%. Plasmodium

rouxi, H. sanguineus, H. philippinensis, L. brimonti, L. pvcnonoti, and microfilariae were

all observed (Table 3). As the prevalence observed here is relatively low, it should not be

entirely surprising that none of 26 Hypsipetes madagascariensis examined on Aldabra

Atoll were parasitized (Lowery, 1968). A single, uninfected individual was included in









the Greiner et al. (1996) survey. Raharimanga et al., (2002) reported low prevalence of

Plasmodium or Haemoproteus (7%) and Leucocytozoon (8%). They reported 13 of the

84 (15%) bulbuls sampled were infected with microfilariae. In Uganda, only 14 of 244

(6.3%) bulbuls were infected (Bennett, 1974).

Haemoproteus wenyoni was the most common parasite observed in the warblers

(Sylviidae), with three of the five species infected. It appeared that there might be two

different species of Haemoproteus in these birds, but there was not enough material to

confirm this. One bird was infected with Leucocytozoon phylloscopus and two with

microfilariae. Raharimanga et al. (2002) also found this to be true, with 24% of the 45

birds they examined infected with either Plasmodium or Haemoproteus. They also only

found one bird to be infected with Leucocytozoon. This may be an indication that the

vector- host relationship was not ideal for Leucocytozoon, but was functional. Another

possibility was that there was a moderate vector population, which did not serve to

increase the prevalence, but rather maintained the level of parasitism by Leucocytozoon

spp.

Of the 11 babblers (Timaliidae) surveyed, only one was infected with the

microfilariae from an unknown filaroid nematode. Raharimanga et al. (2002) reported a

very low prevalence of parasitism in this family. Only one of 25 birds was infected with

Plasmodium or Haemoproteus, one with Leucocytozoon, and 5 with microfilariae.

Because of the small sample sizes, it is hard to draw conclusions with regard to

parasitism in this family. The presence of the one infection indicated that members of

this family do come into contact with a functional vector for their respective genus.









In the sunbirds (Nectariinidae), one of 10 birds was infected with H. sequeirae.

Raharimanga et al. (2002) sampled fourteen sunbirds and found that more than half were

infected with either Haemoproteus or Plasmodium, and two with Leucocytozoon. The

difference in prevalence may be a result of sampling location, as in their investigation, all

parasitized birds were taken from locations other than those included in this survey. This

difference may also reflect a true low prevalence, or birds with chronic infections may

not be being recognized as infected.

The Dicruridae is present throughout Africa, and has been examined before for

hematozoa. De Mello (1935b) described Haemoproteus dicruri, later redescribed by

Peirce (1984c). Until now, H. dicruri was the only haemoproteid known from this

family. In our survey, the four birds infected with Haemoproteus were infected both with

H. dicruri, and in addition, H. khani, a new haemoproteid from this family. Both species

were present, at high intensities, in the four infected birds. This is the first examination

of this family in Madagascar. Considering the unique flora and fauna of this island, it is

not surprising that there are parasites that may be unique to Madagascar. Further

sampling is needed to find pure infections of both haemoproteids, for the comparison of

the blood stages and perhaps elucidation of the life cycles of these parasites. If no pure

infections are found, the parasites may share the same vector, or perhaps infection with

one may insult the immune system enough allowing a rise in parasitemia of a chronic

infection of the other. Additional research may even indicate that the two parasites are

the same species, but with two distinct forms of mature gametocyte. This is rarely seen,

but at least two examples exist, H. sacharovyi in Columbidae, and H. forresteri from the

Vangas.









The remaining families were represented by a small sample sizes. Mentioned

earlier, there was 100% infection of Accipitridae by Leucocytozoon toddi. This suggests

a highly successful vector inhabiting the same range as those birds sampled here.

Perhaps the vector is very successful reproductively, or perhaps is well adapted to

obtaining blood meals from this host. Bennett and Blancou (1974) reported the same

species of Leucocytozoon in one of four Buteo brachypterus in 1964, while no members

of this family were included in the other projects in the area.

Nine owls were sampled and four were found to be infected with hematozoa (H.

noctuae, H syrnii, and L. ziemanni). The presence of hematozoa in this family on

Madagascar has been well established in earlier work. Greiner et al. (1996) reported a

species of Haemoproteus in this family. Also, each of the six birds examined by

Raharimanga et al. (2002) was infected with either Plasmodium or Haemoproteus, one

with Leucocytozoon and four with microfilariae.

A small group of kingfishers (Alcedinidae) was examined, only one of which

proved to be infected with H. halcyonis. This is not unusual, as water birds can be less

parasitized by hematozoa (Greiner et al., 1975). Raharimanga et al. (2002) examined

fifteen kingfishers, and none were infected with any Apicomplexan parasites, although

three were carrying microfilariae. On the other hand, Bennett et al. (1974) reported that

each of the kingfishers examined in Uganda were parasitized by either H. enucleator or

H. halcyonis.

Finally, two of three wagtails (Motacillidae) were infected with H. anthi. The only

previous work that included a wagtail was Greiner et al. (1996). They reported no









parasites seen in the one individual examined. This, then represented the first record of

hematozoon infection in this family in Madagascar.

The remaining families: Threskiornithidae, Columbidae, Leptosomidae and

Campephagidae were represented by one individual (2 in Columbidae) and no parasites

were seen. These families will not be discussed.

Vectors

All of the parasites discussed here are vector borne parasites. Therefore, for

infection to be maintained within the population, vectors and avian hosts must come into

regular contact, the parasite must exist in a reservoir population for the vectors to become

infected, and there have to be susceptible avian hosts for the parasite to complete its life

cycle. In many ways, the search for the ornithophilic vectors of avian hematozoa in

Madagascar will require starting from scratch. There are many aspects to vector biology

that must be appreciated before successful investigation can be performed. Behavior,

flight altitude, and altitudes at which the insects search for blood meals all need to be

determined. Each of these factors varies substantially from one species of vector to

another. For example, Pilaka and Elouard (1999) commented on the restriction that

Simulium has to running water during the aquatic stage. Within this restriction, a

preference for factors such as oxygen flow, pH, turbidity, temperature, and habitat all

vary for the different species of black flies. Different insect families forage at different

levels in the canopy. Bennett and Coombs (1975) found that most simuliids and

ceratopogonids were captured from bait birds suspended 3-4 m off the ground, indicating

a preference for feeding off the birds in the canopy. Also, ornithophilic flies are active at

particular times of the day. For example, ornithophilic biting flies were collected most

frequently in the evening crepuscular period (6-10 PM) in Newfoundland (Bennett and









Coombs, 1975). Different species respond to different attractants, uropygial gland ether

extracts, carbon dioxide or visual stimuli, that allow then to zero in on a potential blood

meal (Fallis and Smith, 1965; Bennett et al., 1972). The simuliid flies that are arriving at

sites to feed off the birds may actually not be very close to the sampling site, as Fallis and

Bennett reported (1966) these flies were able to travel a distance of two miles in 5-12

days. On the other hand, it is the limited flight range of Culicoides spp, in addition to

limited host dispersal and low winter temperatures, that was suggested to impact the

transmission of H. meleagridis in northern Florida (Atkinson et al. 1988). Additionally,

more than one species of arthropod may be capable of transmitting the parasite in

question. For instance, at least five species of Culex mosquitoes have been shown to be

either natural or experimental vectors of Plasmodium elongatum (Nayar et al., 1998). In

Canada, L. simondi is transmitted by two species of simuliid, which are present at

slightly different times, allowing for longer possible transmission of the parasite (Fallis

and Bennett, 1966). Because the natural vector, Simuliim spp., of T. avium in Canada

cannot be colonized, researchers were able to successfully use Aedes sp. in it's place

(Bennett, 1970). This, however, does not mean that any member of the same family of

insects can successfully vector hematozoa. In 1990, Work et al. (1990) demonstrated that

sporozoites of P. relictum failed to develop in three species of Culicidae, although the

primary natural vector for that parasite is a member of the same family. Fallis and

Bennett (1966) showed that sylvatic flies were not suitable vectors for L. simondi, a

parasite of ducks and geese. This indicated that there was a specific relationship between

habitat, omithophilic vectors, hematozoa and their avian hosts perpetuating parasitism.









Sol et al. (2000) illustrated how geographic variation in prevalence of parasitism

can be attributed to abundance of vectors using five separate populations of feral pigeons.

After investigating vectors of avian hematozoa in Newfoundland, Bennett and Coombs

(1975) suggested that the absence of Haemoproteus velans from the Picadae and H.

canachites from the Tetraonidae may be a result of low vector density in their study site.

They also noted that number of vectors is not likely to be the only factor influencing

vector potential. They concluded that low vector density and high vertebrate host density

can produce the same prevalence of parasitism as in high vector densities and low

vertebrate host densities.

The investigation and identification of the insects of Madagascar is in its early

stages. Like so many areas of biological investigation there, much remains to be

understood. Culicoides spp. are known to be present in Madagascar (Meiswinkel, 1991;

Sebastian et al. 2001). Duchemin et al. (2003) discuss the presence of the Anopheles

spp., with 26 species on Madagascar. Interestingly, the incredible endemnicity of the

island holds true for this genus, with 42% being endemic. Also, 29 species of Aedini are

present, ten of these are endemic. The Simuliidae, or black flies, are also present on

Madagascar. Species present are known to transmit animal onchocerciasis (Elouard

2003) in Madagascar, but human onchoceriasis in not present. In Madagascar, 22 of the

27 species if Simuliidae are endemic. Elouard comments that at least 13 species remain

to be described, and that less than 80% of the Malagasy Simulium spp. have been

discovered. So while the same genera that contain ornithophilic vectors are known to be

present in Madagascar, it remains unknown which of these may be responsible for

transmitting the hematozoa observed in this survey.









Altitude

Altitude in an important consideration when examining blood parasitism in birds.

With increases in altitude, vegetation and temperatures change affects which vectors will

thrive in a particular area. Hawkins et al. (1998) commented that the change in bird

density with elevation in Anjanaharibe-Sud somewhat reflected the decrease in canopy

height with elevation. In addition, the birds present can also change, as while some

species are elevation generalists, others are more restricted within a range of a few

hundred meters (Hawkins et al., 1998). Initially, nine altitudinal ranges were compared

to determine if any difference in prevalence by elevation was noted. A difference was

noted and altitudes were then compared to one another in pairs. Only three groups were

significantly different from the others. Two of which, 120 m and 1950 m, represented

small sample sizes, and further investigation with larger sample sizes is needed before

solid conclusions can be drawn regarding the prevalence of parasitism in these areas.

Additionally, at 1200 m, only 6% of the birds were infected. While this was low, it was

significantly different only from 400m. After consolidating groups, it became evident

only the low group was statistically different from the others. It is not surprising that the

lowest elevation had a higher prevalence of parasitism, and this concurs with

Raharimanga et al. (2002) who reported the prevalence of Plasmodium or Haemoproteus

and microfilariae to be significantly higher below 500 m. Based on the data obtained

thus far, it would be difficult to form definite theories as to why prevalence of parasitism

is higher at this elevation, as the sites sampled did not include the same ranges, and the

habitats in each range are not similar in all cases. This precludes us from drawing solid

conclusions, but certainly opens the door for more directed investigations. The

prevalence of Plasmodium infections by altitude has been examined in the Hawaiian









Islands (van Riper et al., 1986). In that study, elevation was found to have a significant

influence upon both population prevalence and parasitemia levels. In that case it was the

mid range elevations that had the highest prevalence of malaria, and it was theorized that

the cause for this is the overlap of the vector population and the native (susceptible) avian

host. As the vectors for the parasites in our study are identified, and the biology of their

life cycles are understood, this information can be compared to the findings reported here

to provide a more complete understanding of the factors driving the different prevalences

reported here.

Thankfully, a relatively large amount has been published in recent years about the

avifauna and its altitudinal ranges. With regard to Andohahela, a reserve from which a

portion of the birds for this study were sampled, Hawkins and Goodman (1999) discussed

species richness with regard to low, medium, and high elevation species of birds and

reported species richness slightly greater at mid elevations and lower at high elevations.

They reported that the two high elevation sites (1500 and 1875 m) had distinctly fewer

species, and at 1875 m there were fewer forest species than at the other altitudes (more

aerial species were noted here). When graphed, species richness for Andohahela peaked

at 1200 m and began to decline as altitude increased. In Anjanaharibe-Sud, species

richness was clearly lower at the highest range sampled, 1950 m, but the results were

unclear with regard to comparisons between the other three altitudes. The biology of the

avian host is an important factor affecting parasitism, and understanding this allows

identification of trends in the future. For instance, van Riper et al. (1986) suggested that

avian malaria may be restricting native Hawaiian populations to higher elevations and

drier areas, in addition to modifying behavior patterns to minimize contact with vectors









of the parasite. Monitoring the prevalence of parasitism through the ranges in elevation

discussed here, combined with data on the avian populations will continue to be an

important tool for conservationists and biologists.

Reserves

Prevalence was compared between the four regions from which birds were

sampled. The four regions sampled were reserves, Ambohitantely, Anjanaharibe-Sud,

Andohahela and Montagne d'Ambre. Ambohitantely is in the central highlands, and is

an area of forested fragments. Anjanaharibe-Sud lies along a chain of mountains covered

with humid forest in northeastern Madagascar; the climate is a humid and tropical

(Goodman, 1998; Goodman and Lewis, 1998). Relative humidity reaches 97% and

annual precipitation is reported to be slightly more than 2 m, two thirds of which falls in

the rainy season. The rainy season here begins in December, and the samples taken prior

to the start. On average it rains 271 days per year (Goodman and Lewis, 1998).

Andohahela is at the southeastern tip of Madagascar. All of this reserve falls below the

Tropic of Capricorn and is one of the southernmost "tropical" forests in the Old World

(Goodman, 1999). A variety of habitats were found in this reserve, including humid

forest, spiny forest, gallery forest, savanna, and riverine habitats. Additionally, the

elevation ranges from low (440 m) to high (1875 m). Because of these factors there is a

rich variety of avian species represented (Hawkins and Goodman, 1999). Andohahela is

considered comparable to Anjanaharibe-Sud. Finally, there is Montagne d'Ambre, an

isolated area of montane forest at the far north end of the island. This is mainly a mid to

high elevation site, as forest starts at 1000 m and reaches to 1300 m.

A statistical difference in overall prevalence by reserve was observed (P = 0.023).

Reserves were then compared in pairs. Andohahela was clearly different from









Ambohitantely. Compared with Montagne d'Ambre, the difference was not significant,

but the power of the test was low. Because Montagne d'Ambre had the same prevalence

as both Ambohitantely and Anjanaharibe-Sud, and the power of the test was low

(probably as a result of the smaller sample size in this reserve), I interpreted the results to

indicate a difference does exist. When compared to Anjanaharibe-Sud, the p-value was

0.069, only slightly higher than the generally accepted cut-off of 0.05. Because it is so

close to this mark as well as the pattern of parasitism for this region, I feel that there is a

difference and a larger sample size would clarify this. It is not entirely unexpected that

these two groups are more similar, as biologists working in the area have described the

sites as comparable. Andohahela had the highest prevalence of each genus of hematozoa

compared to the other three regions. The only exception was in relation to the

trypanosomes, where there was one infected bird in Andohahela and one in

Ambohitantely. Four birds sampled in Andohahela were infected with Plasmodium, the

two birds that were infected with the unnamed species of Babesia were also from this

reserve. Andohahela had the same amount of Leucocytozoon as the other three groups

combined, and more Haemoproteus and microfilariae infections than the other three

groups combined. As described earlier, this particular reserve has a variety of habitats

and elevations resulting in a high density of avian species. In addition to the variety of

potential host species, and microhabitats to suit difference types of vectors, there is also

known to be several types of aquatic habitats within this reserve. There is marshland,

slow moving rivers, and the reserve borders on the tributary of one of the larger rivers in

the area. Over 50% of the known Malagasy species of Simulium occur in the









southeastern region, specifically Andohahela, of Madagascar (Pilaka and Elouard, 1999).

No statistical significance was observed between the other groups.

Habitat

Birds were classified into three groups: FED, FDE, and FDW. There appeared to be

no difference in prevalence observed in between groups, but all Chi-squares were

performed with a low power, indicating that not enough data were included. The majority

of the species sampled were either forest dweller or forest dependant species, 365 of 378

birds total. Only 13 birds considered to be forest edge species were sampled. Data on

this factor should continue to be collected, so that habitat can be compared using a larger

and more balanced sample. Effort should be made to investigate a proportionate amount

of forest edge dwelling species for this classification to be compared to the other two.

Gender

Gender was established for 187 of the birds sampled, and prevalence was 22% for

each cohort. Seventeen of 76 females and 24 of 111 males were infected with at least one

species of hematozoa. No statistical difference was observed between groups.

Raharimanga et al. (2002) also reported no difference was found in the prevalence in

males compared to females. Sol et al. (2000) found no difference in prevalence of

parasitism by gender, pointing out that there is probably equal exposure to vectors.

Weatherhead and Bennett (1991) reported a higher prevalence of parasitism in male Red-

winged Blackbirds than in females, and as already mentioned, the males were more likely

to remain in the same area while females were less likely to remain or return to the area.

Based on this, it is possible that the higher prevalence seen in males in that study is

simply a result of greater exposure to vectors. Perhaps the other sites where these birds

move to are less ideal for the vector, reducing the rate of transmission. Although









Deviche et al. (2001) found a difference in intensity by gender, prevalence was not

different between genders in Dark-eyed Juncos. Additionally, they postulated that the

difference in intensity of Leucocytozoon infection observed was probably a result of a

difference in behaviors exhibited by the two genders.

Age

Sixty-five of 284 adults and 2 of 26 juveniles were found to be infected with at

least one genus of hematozoa. No significant difference in prevalence was observed

between adults and juveniles. There is some concern when sampling young birds about if

enough time has passed since hatching for the birds to become infected, and the infection

to become patent. Since all birds were mist-netted for this survey, the juveniles were old

enough to have fledged and therefore old enough for infection to become patent assuming

they were fed on early after hatching. The lack of difference between the two groups

indicates there is at least equal exposure to the vectors for both age groups. Sol et al.

(2000) found that intensity of parasitism by age class was similar. Weatherhead and

Bennett (1990) reported increasing prevalence of parasitism with age in male Red-

winged Blackbirds. Peirce (1984b) reported an only slightly higher prevalence of

parasitism in juveniles at one site in Zambia. In Hawaii, van Riper et al. (1986) reported

no significant difference in prevalence between juveniles and adults for either native or

introduced species. In Madagascar, no difference in prevalence by genus was found

between adults and juveniles by Raharimanga et al. (2002).

Breeding Condition

Reproductive status here was solely based on the presence or absence of a brood

patch, indicating that the bird was actively nesting or sitting on eggs. No statistical

difference was observed in parasitism between groups. Because of the nature of the






67


study design, there was no other way to determine reproductive status other than this

morphological cue. Results for this can be confounded, for instance if species were

sitting on nests that that time but perhaps do not develop brood patches. If they were

parasitized, but not showing a brood patch, they would have been inappropriately

grouped with those birds not laying at that time. Because of the variation in parental

effort and brood patch development in birds, it is perhaps best to consider this factor only

when examining the same species or family of bird.














CHAPTER 6
CONCLUSIONS

In this study, the presence of seven new species of hematozoa was reported,

possibly endemic to the island. This increases the current knowledge of the flora and

fauna of the island, and indicates that the parasites of these birds may also have a large

degree of endemism like their hosts. Additional research is needed before solid

conclusions can be reached. The prevalence of parasitism in one of the sites sampled,

Andohahela, was shown to be significantly higher than the other three sites. This was

also the area with a high degree of species richness and varied habitats within the

sampling area to support a variety of omithophilic flies. Prevalence also varied by avian

family, which could be a result of one or a combination of factors such as avian host

behavior, vector abundance, or vector feeding preference. Finally, prevalence of

parasitism was significantly higher at low altitudes. No difference in parasitism was

observed for habitat, age, gender or breeding condition. The prevalence of parasitism,

both overall and by parasite genera, was found to be similar to that reported from other

countries.

As entomological and ornithological research continues in Madagascar, a more

complete understanding of the parasitism can be developed. Many facets of the

relationship between the parasite, arthropod vector and avian host remain to be

investigated. For instance, it remains to be seen what impact parasitism by hematozoa

has on the avian populations, and if endemic birds are suffering more as a result of

infection than birds representing more broadly distributed species. Also, there are









isolated and distinct populations of birds within a species that do not mix. It is important

to compare these groups to determine if they are parasitized by the same species of

hematozoa, or if one population is more commonly parasitized or less so.

Only limited conclusions can be drawn considering the amount of data yet to be

collected. The results of this study provide a baseline for other islands that have not been

examined. A great deal remains to be learned about the flora and fauna of Madagascar.

This study and those done beforehand have established a baseline of knowledge upon

which future investigations can be developed.














APPENDIX
CHICKEN HEMATOZOA

Chickens and other poultry are an important commodity throughout Africa, but are

most important in small rural or impoverished communities. Chickens are of important

nutritional value, both as meat animals, but also for eggs. Chickens also may be used as

gifts or in celebrations. Conservationists are also interested in knowing if the importation

of birds may be bringing potential pathogens into a naive population. So, for several

reasons the health of chickens and other domestic fowl is important. In rural

communities, chicken are usually allowed to forage freely, and are not kept caged or

penned as commonly seen in larger production oriented farms. Foraging may increase

the exposure to vector species of omithophilic flies. This can be a result of birds

spending more time in the range occupied by vectors, or also by being present in those

environments at the preferred feeding times by those vectors.

Haemoparasites of poultry have been investigated, probably largely due to the

detrimental effects of Leucocytozoon caulleryi in Eastern Asia. This particular parasite is

the causative agent of chicken leucocytozoonosis, which is known to cause weight loss

and decreased egg production, and death (Isobe and Akiba, 1986). Several studies have

been published summarizing the known hosts and parasites, and species of

Leucocytozoon, Plasmodium, Haemoproteus, Trypanosoma and Aegyptianella have all

been reported (Fallis et al., 1973; Bennett et all, 1991; Earle et al., 1991; Huchzermeyer

and van der Vyver, F.H., 1991; Huchzermeyer, 1993).









Blood samples were taken from 15 domestic chickens, and examined for the

presence of hematozoa. The majority of the chickens were obtained from a village near

the limit of the Anjanaharibe-Sud forest. Two more were from Tolagnaro, while the

origins of the remaining three were unknown.

A single macrogametocyte of Leucocytozoon schoutedeni was observed in one

blood smear from these birds. No other species of hematozoa were seen to be present in

these birds.

Further investigation would be helpful in understanding the prevalence of blood

parasitism in domestic poultry in Madagascar, although it appears that prevalence is low.

It is possible that chickens infected with Leucocytozoon schoutedeni were imported,

bringing the parasite to the island. Low prevalence may indicate that the insect that is

vectoring this parasite is not the optimal vector, therefore, development of the parasite in

the arthropod host is somehow partially impeded. Another possible cause of low parasite

prevalence is that the parasite is highly pathogenic and kills infected birds rapidly. There

are no reports regarding high chicken mortality in Madagascar, and although there are no

published data to this point, it does not seem likely that the low prevalence in this study

was due to pathogenicity. Because these chickens were all from remote areas, it is

probable that that they were raised in that area from stock that the family or village had

maintained. Also, due to the low economic status of the population, it is unlikely that

most individuals would be able to import birds themselves. Therefore, the suggestion

that the parasite is being maintained at low levels in the country is more likely a result of

inadequate capability of flies to transmit this parasite. Finally, while this parasite

matches the description of L. schoutedeni, it is important to examine a larger number of









birds to determine the true parameters and morphology of Leucocytozoon species

present, to conclusively determine if it is a known parasite from Africa or a species

endemic to Madagascar. In addition, it will be necessary to determine import trends with

regard to poultry. If eggs are imported to raise replacement stock, it is unlikely that the

parasite observed here is of African origin. If young birds are more commonly imported

it is possible that they could become infected prior to shipping, and develop a parasitemia

and be responsible for the arrival of this parasite to Madagascar.

Although the margins have been set throughout this document correctly, please pay

close attention to the possibility of picture frames overlapping the margin. The base style

to use is Normal.















LIST OF REFERENCES


Apanius, V., and C.E. Kirkpatrick. 1988. Preliminary report of Haemoproteus tinnunculi
infections in a breeding population of American Kestrels (Falco sparverius).
Journal of Wildlife Diseases 24: 150-153.

Atkinson, C.T. 1986. Host specificity and morphometric variation ofHaemoproteus
melegridis Levine, 1961 (Protozoa: Haemosporina) in gallinaceous birds. Canadian
Journal of Zoology 64: 2634-8.

Atkinson, C.T., Forrester, D.J., and E.C. Greiner. 1988a. Pathogenicity of Haemoproteus
melegridis (Haemosporina: Haemoproteidae) in experimentally infected turkeys.
Journal of Parasitology 74: 228-239.

Atkinson, C.T., Forrester, D.J. and E.C. Greiner. 1988b. Epizootiology of Haemoproteus
meleagridis (Protozoa: Haemosporina) in Florida: Seasonal transmission and vector
abundance. Journal of Medical Entomology 25: 45-51.

Baker, J.R. 1968. The host restriction ofHaemoproteus columbae. Journal of
Protozoology 15: 334-335.

Beier, J.C., and M.K. Stoskopf 1980. The epidemiology of avian malaria in Black-
Footed penguins (Spheniscus demersus). Journal of Zoo Animal Medicine 11: 99-
105.

Bennett, G.F. 1961. On the specificity and transmission of some avian trypanosomes.
Canadian Journal of Zoology 39: 17-33.

Bennett, G.F. 1970. Development of trypanosomes of the T. avium complex in the
invertebrate host. Canadian Journal of Zoology 48: 945-957.

Bennett, G.F. 1990. Avian Haemoprotozoa of the Indian subcontinent- the species and
vectors. Proceedings of the Zoological Society, Calcutta 43: 49-58.

Bennett, G.F., Bishop, M.A., and M.A Peirce. 1993. Checklist of the avian species of
Plasmodium Marchiafava & Celli, 1885 (Apicomplexa) and their distribution by
avian family and Wallacean life zones. Systematic Parasitology 26: 171-179.

Bennett, G.F. and J. Blancou. 1974. A note on the blood parasites of some birds from the
Republic of Madagascar. Journal of Wildlife Diseases 10: 239-240.









Bennett, G.F., Blancou, J., White, E.M., and N.A. Williams. 1978. Blood parasites of
some birds from Senegal. Journal of Wildlife Diseases 14: 67-73.

Bennett, G.F., and A.G. Campbell. 1972. Avian Haemoproteidae. I. Description of
Haemoproteus fallisi n. sp. and a review of the haemoproteids of the family
Turdidae. Canadian Journal of Zoology 50: 1269-1275.

Bennett, G.F. and R.F. Coombs. 1975. Ornithophilic vectors of avian hematozoa in
insular Newfoundland. Canadian Journal Zoology 53: 1241-1246.

Bennett, G.F., Earle, R.A., Peirce, M.A., Huchzermeyer, F.W. and D. Squires-Parsons.
1991. Avian Leucocytozoidae: the leucocytozoids of the Phansianidae sensu lato.
Journal of Natural History 25: 1407-1428.

Bennett, G.F., Earle, R.A., and D. Squires-Parsons. 1994. Trypanosomes of some sub-
Saharan birds. Onderstepoort Journal of Veterinary Research 61: 263-271.

Bennett, G.F. and A.M. Fallis. 1960. Blood parasites of birds in Algonquin Park, Canada,
and a discussion of their transmission. Canadian Journal of Zoology 38: 261-273.

Bennett, G.F., Fallis A.M., and A.G. Campbell. 1972. The response of Similium
(Eusimulium) euryadminiculum Davies (Diptera: Simuliidae) to some olfactory
and visual stimuli. Canadian Journal of Zoology 50: 793-800.

Bennett, G.F., and C.M. Herman. 1976. Blood parasites of some birds from Kenya,
Tanzania and Zaire. Journal of Wildlife Diseases 12: 59-65.

Bennett, G.F., Okia, N.O., and M.F. Cameron. 1974. Avian hematozoa of some Ugandan
birds. Journal of Wildlife Diseases 10: 458-465.

Bennett, G.F., and M.A. Peirce. 1988. Morphological form in the avian Haemoproteidae
and an annotated checklist of the genus Haemoproteus Kruse, 1890. Journal of
Natural History 22: 1683-1696.

Bennett, G.F. and M.A. Peirce. 1990. The haemoproteid parasites of the pigeons and
doves (family Columbidae). Journal of Natural History 24: 311-325.

Bennett, G.F., Peirce, M.A., and R.A. Earle. 1994. An annotated checklist of the valid
avian species of Haemoproteus, Leucocytozoon (Apicomplexa: Haemosporida) and
Hepatazoon (Apicomplexa: Haemogregarinidae). Systemic Parasitology 29: 61-73.

Clements, J.F. 2000. Birds of the World: A Checklist, 5th Ed, Ibis Publishing Company,
Vista, California.

De Mello, I.F. 1935a. A contribution to the study of the blood parasites of some Indian
birds. Proceedings of the Indian Academy of Science I: 349-358.









De Mello, I.F. 193 5b. New haemoproteids of some Indian birds. Proceedings of the
Indian Academy of Science (B) 2: 469-475.

De Mello, I.F. and L. da Fonseca.1937. Further notes on the haemoparasitology of the
Indian birds. Proceedings of the Indian Academy of Science. (B) 6: 213-219.

Deviche, P., Greiner, E.C., and X. Manteca. 2001. Seasonal and age-related changes in
blood parasite prevalence in Dark-eyed Juncos (Junco hyemalis, Aves,
Passeriformes). Journal of Experimental Zoology 289: 456-466.

Duchemin, J.B., Ravoahangimalala, O.R., and G. Le Goff. 2003. Culicidae, Mosquitoes.
In The Natural History of Madagascar, S. M. Goodman and J. P. Benstead (eds.),
The University of Chicago Press, Chicago.

Earle, R.A., Horak, I.G., Huchzermeyer, F.W., Bennett, G.F., Braack, L.E.O. and B.L.
Penzhorn, 1991. The prevalence of blood parasites in helmeted guinea fowls,
Numida melegris, in the Kruger National Park. Onderstepoort Journal of Veterinary
Research. 58: 145-147.

Elouard, J.M. 2003. Simuliidae, Black Flies. In The Natural History of Madagascar, S.
M. Goodman and J. P. Benstead (eds.), The University of Chicago Press, Chicago.

Fallis, A.M. and G.F. Bennett. 1966. On the epizootiology of infections caused by
Leucocytozoon simondi in Algonquin Park, Canada. Canadian Journal of Zoology.
44: 101-112.

Fallis, A.M., Desser, S.H., and R.A. Khan. 1974. On species of Leucocytozoon.
Advances in Parasitology 12: 1-67.

Fallis, A.M., Jacobson, R.L., and J.N. Raybould. 1973a. Experimental transmission of
Trypanosoma numidae Wenyon to guinea fowl and chickens in Tanzania. Journal
of Protozoology 20: 436-437.

Fallis, A.M. Jacobson, R.L. and J.N. Raybould. 1973b. Haematozoa in domestic chickens
and guinea fowl in Tanzania and transmission of Leucocytozoon neavi and
Leucocytozoon schoutedni. Journal of Protozoology 20: 438-442.

Fallis, A.M., Pearson, J.C., and G.F. Bennett. 1954. On the specificity of Leucocytozoon.
Canadian Journal of Zoology 32: 120-124.

Fallis, A.M. and S.M. Smith. 1965. Attractions of some simuliids to ether extracts from
birds and to carbon dioxide. Proceedings of the XII International Congress of
Entomology. London, 1964.

Fix, A.S., Waterhouse, C., Greiner, E.C., and M.K. Stoskopt. 1988. Plasmodium relictum
as a cause of avian malaria in wild-caught Magellanic penguins. Journal of Wildlife
Diseases 24: 610-619.









Forrester, D.J., Greiner, E.C., and G.F. Bennett. 1977. Avian Haemoproteidae. 7. A
review of the haemoproteids of the family Ciconiidae (storks) and descriptions of
Haemoproteus brobkorbi sp.nov. and H. peirci sp.nov. Canadian Journal of
Zoology. 55: 1268-1274.

Goodman, S.M. 1998. Description of the 1994 Biological Inventory of the Reserve
Speciale d'Anjanaharibe-Sud, Madagascar. Fieldiana: Zoology 90: 1-7.

Goodman, S.M. 1999. Description of the Reserve Naturelle Integrale d'Andohahela,
Madagascar, and the 1995 Biological Inventory of the Reserve. Fieldiana: Zoology
94: 1-8.

Goodman, S.M. and B.A. Lewis. 1998. Description of the Reserve Speciale
d'Anjanaharibe-Sud, Madagascar. Fieldiana: Zoology 90: 9-16.

Greiner, E.C., Bennett, G.F. White, E.M, and R.F. Coombs. 1975. Distribution of the
avian hematozoa of North America. Canadian Journal of Zoology 53: 1762-1787.

Greiner, E.C., Eveleigh, E.S., and W. M. Boone. 1978. Ornithophilic Culicoides spp.
(Diptera: Ceratopogonidae) from New Brunswick, Canada, and implications of
their involvement in haemoproteid transmission. Journal of Medical Entomology
14: 701-704.

Greiner, E.C., Putnam, M.S., and S.M. Goodman. 1996. Blood parasites from birds in the
Reserve Naturelle Integrale d' Andringitra, Madagascar. Fieldiana: Zoology 85:
142-143.

Hawkins, A.F.A., and S.M. Goodman. 1999. Bird community variation with elevation
and habitat in parcels 1 and 2 of the Reserve Naturelle Integrale d'Andohahela,
Madagascar. Fieldiana: Zoology 94: 175-186.

Hawkins, A.F.A., Thoillay, J-M, and S.M. Goodman. 1998. The birds of the Reserve
Speciale d'Anjanaharibe-Sud, Madagascar. Fieldiana: Zoology 90: 93-127

Hayworth, A.M., van Riper, C., and W.W. Weathers. 1987. Effects of Plasmodium
relictum on the metabolic rate and body temperatures in canaries (Serinus canarius)
Journal of Parasitology 73: 850-853.

Herman, C.M. 1975. Plasmodium circumflexum: Age as a factor in isodiagnosis in
Anatidae. Experimental Parasitology 38: 83-86.

Huchzermeyer, F.W. and van der Vyver, F.H. 1991. Isolation of Plasmodium
circumflexum from wild guineafowl (Numida melegris) and the experimental
infection in domestic poultry. Avian Pathology 20: 213-223.









Huchzermeyer, F.W. 1993. A host-parasite list of the haematozoa of domestic poultry in
sub-Saharan Africa and the isolation of Plasmodium durae Herman from turkeys
and francolins in South Africa. Onderstepoort Journal of Veterinary Research 60:
15-21.

Isobe, T. and K. Akiba. 1986. Development of erythrocytic merozoites to gametocytes in
chickens recovered from sporozoite infection with Leucocytozoon caulleryi.
Journal of Parasitology 72: 190-192.

Jarvi, S.I., Scultz, J.J., and C.T. Atkinson. 2002. PCR Diagnostics underestimate the
prevalence of avian malaria (Plasmodium relictum) in experimentally infected
passerines. Journal of Parasitology 88: 153-158.

Kocan, R.M., and D.T. Clark. 1966. Anemia in ducks infected with Leucocytozoon
simondi. Journal of Protozoology 13: 465-468.

Kottak, C.P. 1980. The past in the present: history, ecology, and cultural variation in
highland Madagascar. University of Michigan Press, Ann Arbor, Michigan.

Kucera, J. 1981. Blood parasites of birds in central Europe. 1. Survey of literature. The
incidence in domestic birds and general remarks to the incidence in wild birds.
Folia Parasitologica (PRAHA) 28: 13-22.

Langrand, 0. 1990. Guide to the birds of Madagascar. Yale University Press, New
Haven, Connecticut

Langrand, 0. 2001. Family Brachypteraciidae (Ground-rollers). In Handbook of the Birds
of the World Vol. 6 Mousebirds to Hornbills, del Hoyo, J., Elliot, A., and J.
Sargatal (eds.) Lynx Edicions, Barcelona, Spain, p. 378-388.

Lowery, R.S. 1971. Blood parasites of vertebrates on Aldabra. Philosophical
Transactions of the Royal Society, -B 260: 577-580.

McClure, H.E., Poonswad, P., Greiner, E.C., M. Laird. 1978. Haematozoa in the birds of
Eastern and Southern Asia. Memorial University of Newfoundland; St. John's,
Newfoundland.

Meiswinkel, R. 1991. Afrotropical Culicoides: C. (avaritia) miombo sp. nov., a
widespread species closely allied to C. (A.) imicola Kieffer, 1913 (Diptera:
Ceratopogoniidae). Onderstepoort Journal of Veterinary Research 58: 155-170.

Morris, P. and F. Hawkins. 1998. Birds of Madagascar A Photographic Guide. Yale
University Press, New Haven, Connecticut, p. 298.

Nakamura, K., Ogiso M., Shibahara T., Kasuga H.,and T. Isobe. 2001. Pathogenicity of
Leucocytozoon caulleryi for specific pathogen-free laying hens. Journal of
Parasitology 87: 1202-1204.









Nayar, J.K., Knight, J.W., and S.R. Telford. 1998. Vector ability of mosquitoes for
isolates of Plasmodium elongatum from raptors in Florida. Journal of Parasitology.
84: 542-546.

Peirce, M.A. 1979 Some additional observations on haematozoa of birds in the
Mascarene Islands. Bulletin of the British Ornithologists Club 99: 68-71.

Peirce, M.A. 1984a. Haematozoa of African birds: some miscellaneous findings. African
Journal of Ecology 22: 149-152.

Peirce, M.A. 1984b. Haematozoa of Zambian birds. I. General survey. Journal of Natural
History 18: 105-122.

Peirce, M.A. 1984c. Haematozoa of Zambian birds. VIII. Redescription of Haemoproteus
dicruri from Dicrurus adsimilis (Dicruridae). Journal of Natural History 18: 789-
791.

Peirce, M.A., Cheke, A.S., and R.A. Cheke. 1977. A survey of blood parasites of birds in
the Mascarene Islands, Indian Ocean, with descriptions of two new species and
taxonomic discussion. Ibis 119: 451-461.

Peirce, M.A. and C.J. Feare. 1978. Piroplasmosis in the masked booby Sula dactylatra
melanops in the Amirantes, Indian Ocean. Bulletin of the British Ornithologists
Club 98: 38-39.

Pilaka, T. and J.M. Elouard. 1999. Aquatic biodiversity of Madagascar: Simulium
(Diptera: Simuliidae) from the Reserve Naturelle Integrale d'Andohahela and
surrounding areas. Fieldiana Zoology 94: 125-128.

Raharimanga, V., Soula, F., Raherilalao, M.J., Goodman, S.M., Sadones, H., Tall, A.,
Randrianarivelojosia, M., Raharimalala, L., Duchemin, J.B., Ariey, F., and V.
Robert. 2002. Hemoparasites des oiseaux sauvages a Madagascar. Archives de
l'Institut Pasteur de Madagascar 68: 90-99.

Raidal, S.R., and S.M. Jaensch. 2000. Central nervous disease and blindness in Nankeen
kestrels (Falco cenchroieds) due to a novel Leucocytozoon-like infection. Avian
Pathology 29: 51-56.

Sebastiani, F., Meiswinkel, R., Gomulski, L.M., Guglielmino, C.R., Mellors, P.S.,
Malacrida, A.R. and G. Sasperi. 2001. Molecular differentiation of the Old World
Culicoides imicola species complex (Diptera, Ceratopogonidae), inferred using
random amplified polymorphic DNA markers. Molecular Ecology. 10: 1773-1786.

Sol, D., Jovani, R. and J. Torres. 2000. Geographical variation in blood parasites in feral
pigeons: the role of vectors. Ecography 23: 307-314.

Stoskopf, M.K., and J. Beier. 1979. Avian malaria in African black-footed penguins.
Journal of the American Veterinary Medical Association 175: 944-947.









Van Riper, C., van Riper, S.G., Goff, M.L., and M. Laird. 1986. The epizootiology and
ecological significance of malaria in Hawaiian land birds. Ecological Monographs,
56: 327-344.

Warner, R.E. 1968. The role of introduced disease in the extinction of the endemic
Hawaiian avifauna. The Condor 70: 101-120.

Weatherhead, P. J. and G.F. Bennett. 1991. Ecology of Red-winged Blackbird parasitism
by haematozoa. Canadian Journal of Zoology 69: 2352-2359.

White, E.M., Greiner, E.C., Bennett, G.F., and C.M. Herman. 1978. Distribution of the
hematozoa of Neotropical birds. Revista de Biologia Tropical 26: 43-102.

Work, T.M., Washino, R.K. and C. van Riper. 1990. Comparative susceptibility of Culex
tarsalis, Anopheles franciscanus, and Culiseta inomata (Diptera: Culicidae) to
Plasmodium relictum (Haemosporina: Plasmodiiae). Journal of Medical
Entomology 27: 68-71.

Yorinks, N., and C.T. Atkinson. 2000. Effects of malaria on activity budgets of
experimentally infected juvenile Apapane (Himatione sanguinea. The Auk 117:
731-738.















BIOGRAPHICAL SKETCH

I was born in upstate New York, the youngest of four children. We moved to

Cheshire, Connecticut when I was young, and that is where I grew up. Before attending

the University of Florida, I attended the University of Connecticut and graduated with a

double major in pathobiology and animal science. I enjoy international travel,

particularly to the Andean nations. When I have the opportunity, I also enjoy reading and

studying European history.